title_body
stringlengths
114
4.06k
upvoted_answer
stringlengths
69
4.9k
downvoted_answer
stringlengths
34
5.14k
Seemingly random lines on the surface I'm getting seemingly random lines scattered across the top surface of my prints: Printer: Anycubic i3 mega Slicer: Cura 3.2.1 Printer chosen in Cura: Prusa i3, Gcode flavor changed to RepRap Cura Profile: Fine, "Outer before inner walls" enabled What might be the reason?
This is the result of travel moves passing through/over the top surface. The combing setting can prevent this type of move (for walls, infill or top), and z-hop can avoid making any mark during these travels. Both settings will affect speed. Z-hop will be active on every layer, for example.
It seems like the issue may only be that your nozzle has developed a bit of oozing. The g code likes to repeat the same patterns for at least a few layers every time it changes patterns and if it does the same pattern on the top 3-4 layers than by the time it reaches the last layer on the top that pattern will actually be the slightest bit raised above the rest of the layers because of the ooze building up along the same pattern for how many ever layers it followed the same pattern. Your nozzle is than rubbing on that raised pattern when it completes the last layer. This explains the randomness like OP said because the g code may not repeat the same pattern for 3 or more layers at the surface but without reading the gcode it would seemingly be random when and where they occurred. And truthfully if its a small oozing you could reprint this and it could be gone if you changed almost any setting in cura because the g code would be recalculated differently and new patterns would be chosen but so long as the new gcode did not have the same pattern repeat itself on the top 3 or so layers you would not have these raised patterns for the nozzle to rub on when finishing the top last layer.
Ender 3 Extruder First Layer Perfect But Then Starts Clicking Afterwards All of a sudden I seem to be having a lot of issues with under extruding on my Ender 3. The bottom layer (of height 0.1 mm) prints perfectly fine. This is done at 15 mm/s speed. However, The moment the print moves to layer 2 and above (at the default speed of 60 mm/s), I start hearing a lot of clicking noise on the extruder. So far I have tried the following Replace nozzle to eliminate clogs Cleaned the inside of the hot end assembly Calibrated extruder steps/mm Reduced the layer height from 0.3 mm to 0.2 mm Reduced feed rate to as low at 50 % Cleaned the filament feeder assembly and verified that it is able to push the filament properly (Extruding when the print is not happening works just fine with no clicking) Even with all the above, the issue is still persisting. I am not what else could be causing this. I am printing with PLA at 200 C
Not allowed to comment, so have to answer: The temperature sensor is a thermally sensitive resistor. Unfortunately, the temperature is near the high limit of that sensor, and the manufacturing tolerances are very significant. That is why a temperature tower is important for each printer, as well as each filament. (I have 4 printers and each requires a different temperature for the same filament. My worst-case is out by 25 degrees! - it's the one I bought second-hand because the original purchaser couldn't get it to work. I could replace the NTC, but it is easier just to have settings to suit that printer.) Filament does change over time. Lots of theories about why, but the practical response is to tune settings to suit the filament. The alternative is to modify the filament (eg drying, adding oil to surface, etc.), but even with really old filament, I've found adjusting settings (in the slicer, like Cura) to be the most generally workable solution. Bottom line is to test, adjust settings and repeat until the system achieves the result you need. Treat most recommendations as serving suggestions, so use them as clues (but not rules) for the puzzles presented as 3D printing.
A temperature of 200 °C is fine. You probably have heat creep from a Bowden style extruder. The filament starts getting softer due to heat creeping up from the hot nozzle. Filament expands and clogs the tube causing gear clicking. Usually there is a heat sink at the the Bowden tube connection. You need to cool that immediately. Which is, to set your fan to 100 % speed at layer height = 0. That should do the trick!
What is this called and where can I buy a replacement? In the image, there is an object circled in blue on the anet a8 printer that I need to buy a replacement of, however, I can not find it online, please help.
This is the throat block for direct drive extrusion, Anet8 is a cheap clone of Prusa printers, so it's easy to find parts for Anet printers. This is one extruder kit that may help your needs, 1 Unidades Impresora 3D makerbot MK8 Extrusora De extrusión de Aluminio Bloque de DIY Kit para Reprap i3. This is a link for Aliexpress - of course there are other dealers.
Like @ZuOverture said the name of this component is the filament drive. Most of the manufactures sell the whole extrusion device already assembled, to avoid mismatches between the components of the extrusion device. If your device is somehow damaged, and without possibilities to be used in the printer the easiest solution would be to buy a new extrusion device. If financially is unfeasible for you, you may have the option to generate the 3D model of this piece and share the file to someone who could print this component for you.
Makerbot Replicator 2 Halts during print I am running a MakerBot Replicator 2. During the print, the printer just stops executing and I am running out of ways to troubleshoot. After restarting, I can load and extrude filament. I have replaced the SD card, and even borrowed one from another working replicator, and the freeze still occurs. Likewise, I've regenerated the x3g file, and that did not work. The panel does not freeze, I can cancel the print during the freeze. I've made sure Z pause is off. It tends freezes randomly on the first layer. In general, it looks as though the print is "in progress" but not making progress (Timer ticks up, % completed does not) Does anyone have any idea as to what could be causing the problem?
There are a few options. First your board could be overheating etc. That is harder to verify without some overpriced replacements. So to start lets take everything apart (photo and labeling is Strongly recommended). Then simply put it back together. Hopefully it is just a loose connection.
Seems like others have had similar issues: https://www.thingiverse.com/groups/engineering/topic:3849. Their issue was a bit more mechanical than firmware, though. I hope this helps! You might also want to check the software you're using to slice the file. Sometimes the software doesn't slice the file properly, which causes problems mid-print.
What is this weird support line doing in this print? I am a newbie to 3D printing and ran into a weird infill line on my second 3D printing object on a new Qidi X-Pro machine (which works great). I've included a screenshot of the infill line, which is deliberately printed the full height of the object. I'm thinking this line has been deliberately inserted by the Qidi slicer for some reason, but I have no idea why. Do all slicers generate these kinds of lines? If so, why?
I do not know the Qidi slicer, but if you look closely, you will see this line is thinner than the normal support infill lines. You could try to visualize the G-code in a viewer, usually this can be done in the slicer itself, but online viewers are available. The viewer will not only show the printed lines, but also show moves by the print head (usually in a different color). You can check whether this extra line is actually printed or a move. If it is a move, this extra line is caused by your hotend which is leaking when it moves. You need to properly tune the hotend with respect to the retraction settings and temperature. There are numerous retraction test print objects to find on the internet. Depending on your slicer settings, some slicers are able to define where each layer starts printing (e.g. random, or start at sharp corner). The fact you see a support structure "printed the full height of the object" tells you that each layer starts at the same position. It is not uncommon in uniform simple parts where each layer starts at the same position (X/Y) as this is instructed by the slicer setting. In Ultimaker Cura such an option is called Z Seam Alignment. Bottom line, all slicers will do this when your printer is improperly tuned (incorrect settings for e.g. print temperature, retraction, coasting, travel speed). It is up to you to find the correct settings, test print objects help you with that.
Here have a posible solution -> How to remove unwanted filament trails from sharp corners I think, that can be a combination of z-hopping and combing-mode in the slicer that you use.
Wrong Z-Axis movement in G-Code Alright so I bought a broken UP mini. There was only a defect on the board so I connected the components to a Duet 2 Wifi. Everything is fully operational and I can manually move all axes correctly using the interface as well as homing them. The problem is that when I try to slice a model (I am using Slic3r) and upload the G-Code the bed will be raised instead of lowered during printing. How do I change it so the direction of the Z-Axis goes correctly? Additional information: The Z endstop is at the bottom When homed the Z coordinate is on 0 mm When placing the bed under the extruder the Z coordinate is 170 mm
You could put the z endstop at the top, and flip the motor connector to make it move the other direction. This should make it home to the top near the nozzle, and then move downward during the print. Otherwise, you'll have to tweak the firmware configuration and reflash.
You could put the z endstop at the top, and flip the motor connector to make it move the other direction. This should make it home to the top near the nozzle, and then move downward during the print. Otherwise, you'll have to tweak the firmware configuration and reflash.
Why is my print displacing along the y-axis by 2-3 cm? I am encountering a problem with this ID3 printer using ABS -- at some point during the print the print head displaces on the y-axis by 2-3 centimeters. I cannot pinpoint how or why it is doing this. It has displaced in the positive Y direction and in the negative Y direction on separate runs of the same piece (which is just a poker chip I found on Thingiverse). Is this a software issue (Simplify3D) or a hardware issue? Can anybody suggest a fix? See the following photos:
This has happened to me in the past and here are a few things that I determined could have been the cause: Limit Switch triggered. My coworker was able to repeat the issue by manually triggering a limit switch during the print. Somehow, this seemed to cause the relative axis to shift on his MakerBot Replicator 2. Skipped steps. It's possible that, if stalled due to excessive load, the stepper motor could skip steps. I have a dual extruder, so if a part curls really bad then the second nozzle gets caught in the print. I'm not sure exactly how this works, but my coworker suggested this as an issue. USB connection. I had sliced a model and tried printing over USB connection, however each time I attempted the print I encountered the "shift" at the same exact location. However, using the slice, I exported a G-Code file and printed from the machines SD card with no errors. Slicing Engine bug. This is in correlation to the USB connection issue. I noticed that the issue occurred after updating my slicing engine. My solution in the long term was to re-install the previous version that was stable for me. I was using MakerWare 2.4.2 and reverted back to MakerWare 2.4.1. If you're printing via USB connection, I'd suggest exporting to G-Code if you're able to print via SD Card until you can find a stable version of software for your machine. This has been the most repeatable solution for me in the past.
In my case the solution was to use Slic3r instead of Cura. Cura has some issues.
slic3r: Can I vary the infill percentage for different heights of my model? How can I vary the infill percentage for different layer heights of my model? Context: The bottom part of my model needs about 20% infill. The geometry of the top part of the model (mostly cones of various sizes) prints well with 0% infill and is of course a lot faster to print if I can specify this.
Your question is very similar to Different infill in the same part and Using multiple infill types within one model [duplicate]. The difference is that you specifically ask for Slic3r and a variation in layer height infill percentage. Actually this answer describes using "helper volumes" in Ultimaker Cura to set different properties for certain parts of the model (UPDATE: that answer now includes also Slic3r instructions), but it appears that this answer is very much applicable to Slic3r also. Please read this posting. Quoting from the reference: Finally, I fired Slic3r up and loaded the main part, then clicked on Settings... and then hit Load modifier... I loaded the new volume as a modifier mesh and I applied 100% solid infill... Secondly, this answer where 2 different infill percentage sliced models are manually combined at a certain height may also work for you (this is a perfect valid solution for Slic3r, but requires some editing skills). Basically, although you request for a solution for a slicer other than already described in other questions, boils down to a similar answer, the only difference is the implementation in Slic3r is called differently. To do this in Slic3r see this reference. The blog describes the use of a simple volume (the green volume loaded from an STL file). After loading: Right-clicking on the main part brought up the object settings menu. From there, clicking "Load Modifier" and selecting the previously saved model adds it to the part as a modifier. The green "+" was selected and "Fill Density" was added to modifier list and set to 100%. This shows that the functionality in Slic3r is very similar to the functionality in Ultimaker Cura.
Let's assume the model is 75mm high; the bottom portion (which needs 20% infill) is 40mm, and the top portion (which needs 0% infill) is 35mm. Load your model into Slic3r. Be sure your default fill is 20%. Right click on the model, and select Settings. Click "Load Generic". Select Slab. This represents a cube which is the same X,Y dimensions of your model. Specify H=35 (the height of the slab) and Initial Z=40 (where the base of slab starts). This creates a new part for your object. Select it and click "+" to add a new setting. Select Infill/Fill Density and set it to 0%. As a result of this, the bottom part of your model (which is not enclosed in the slab) will print at the default fill setting of 20%, and the top part of your model (which is enclosed in the slab) will print with 0% fill. You can verify this by slicing and looking at the preview. You will see the base with your default infill, and the top with no infill. The slic3r docs mentions several other ways to customize parameters for your model.
First time assemble of Anet A6 printer. Only fan works I bought an Anet A6-L printer (after some research I discovered that this is not a genuine Anet A6) and assembled it. The first time I plugged it in, the fan (fan 2, I believe) is the only thing that turns on. Also, the LCD screen (LCD2004 5 button display) only had 1 cord, and 1 jack for the cord to plug in on the back of the screen, but 2 openings on the motherboard. The fan turns on for around 30 seconds, then turns off. I have the anet A6 printer, with the V1-5 motherboard. I have everything else (fans, motors, hotbed, power) plugged into the right ports. I haven't plugged in the micro SD (TF) card yet, I turned it on to test it. I am using a modified part that replaces the windpipe/fan duct (white) and redirects the airflow around it and inwards from a ring around the tip, but that is the only modification I have made. I have not messed with the power supply (other than wiring it in). Same with the motherboard. I tested it also without the modification for the fan duct but it still has the same problem.
This definitely sounds like a problem with your wiring if you have a genuine Anet A6, the genuine A6 comes with a 12864 full graphic display. For sure, you are missing 1 flat ribbon cable (see below). Maybe this is causing the LCD not to light up and the SD card not functioning. As the "fan 2" is working, the board is powered by the power supply (this fan is using the constant power feed of the supply of the board). What you are actually describing as a boot sequence is the actual boot sequence of a printer. Once you power the printer, the fan that is cooling the cold end of the extruder should start spinning and keep spinning while the part cooling fan usually spins up but then powers down to standstill. While this is happening, the LCD should come alive and show the boot screen and finally the printer menu. If your screen is not showing any light, this implies that your screen is either broken, not powered or wrongly connected. You could connect the printer over USB and control the printer from an external program, e.g. Pronterface, OctoPrint, Repetier-host, etc. and see if the printer works (then you know that the display is broken). From a search on AliExpress I found that there are auto leveling printers sold with the Anet A6 branding that differ from the standard Anet A6 as written in Chinglish: Different Auto leveling A6 and Normal A6: 1.The auto-leveling version uses a proximity sensor to detect the aluminum print bed where the normal version of the printer uses a micro-switch to detect the end of travel for the Z-Axis movement (vertical limits). 2.Auto leveling A6-L work with LCD2004 screen, A6 work with LCD12864 screen The second remark from the quote above suggests that there is a 2004 LCD version that is only used by the Anet A6-L version (probably because they need a free pin for the auto leveling sensor). Such a display only has a single connection socket and needs to be connected to a single socket on the Anet printer board (named "LCD", not "J3") Note that automatic bed levelling is not magic, and a little more complicated to start with, if you order a printer without an auto bed levelling sensor, you will be able to update to one later. E.g. from here: It uses the "LED pin" which is an unused pin on the A8 (using the stock 5 button 2004 LCD). That is the third wire counted from below (where the red marker is on the cable). I simply spliced the cable and cut that wire. This will be the servo signal (yellow). If you have an Anet A6 adapted for auto levelling with a 2004 display When the LCD does not light up, this could be caused by incorrect placement of the flat cable, be sure to use the correct socket on the printer board and take care of the orientation. Once you have checked this, and it does not work and you are in the possession of a multimeter, you could measure the voltage over the "VSS" and the "VDD" pin; also look into the voltage over the "VSS" and the "VE" (see pin layout below). If there is power, but no light, the LCD is probably defective. You could try to hook up a computer to the board using a USB cable and use a program like Pronterface to interface with the printer to see if it works at all, the display is not required for printing (e.i. if you can access the printer over USB). If you have a genuine Anet A6 It is advised to install an extra flat ribbon cable and check all the wires, please do check for correct polarity and correct installment. Please do note the the installation of the Anet A6 LCD display (see this movie and the screenshot from this video below) requires 2 flat ribbon cables to function properly. Sidemark on fan ducts: Both the stock and most ring type ducts are not aerodynamically designed fan ducts. The stock fan converges too much, this narrowing of the duct causes extra pressure build-up which these fans are not able to handle, so they stall, causing a reduced flow output of the fan. The (semi) circular fan ducts usually also have a design problem. The (semi) circular ones all (but one that I have seen) have the same deficiency that the main passage area does not decrease when the duct loses air through a slot/ejector; this means that the velocity in the main ring decreases after each bleed slot! Note that these fans move air and do not build up a high pressure difference that is large enough to overcome the friction of those designs.
I also ordered an A8, but was sent a A6-L, after doing my homework, I am very happy. The A6-L is an upgraded A8. It uses same board as A6 but has a A8 display (5 button), that's why people have extra LCD spot. The A6-L is a better frame (full front), better carriage set up ( horizontal vs vertical, resulting in better weight distribution and allowing for higher builds ). As far as fans go, the cooling fan only comes on when printing and told to (there is a setting in menu (make sure fan is on, as well as check cura settings for when and how much fan should come on). The Auto leveler that comes with it says to plug into Z limit switch, ( basically it replaces the z-switch all together). Thingaverse has better fan/sensor holders you can print. Remember Google/YouTube is your friend. I found out after lots of searching. A simple google search answers everything. The A8 was the 1st printer released by Anet, followed by the A6. Check out links. Look at pictures and description and comparison. Or just Google it. A8= \$ cheapest, flimsy frame, off balanced extruder carriage, uses belt to drive Z. A6= \$\$\$ MOST EXPENSIVE of 3Full frame, better balanced carriage, LCD screen w/ knob, no Z belt. A6-L= \$\$ Middle Cost, A6 frame, carriage, no Z belt, but A8 screen and buttons ANET A6-L https://www.banggood.com/Anet-A6-L5-DIY-3D-Printer-Kit-With-Auto-leveling-220220250mm-Printing-Size-1_75mm-0_4mm-Nozzle-p-1209606.html?cur_warehouse=CN COMPARISION https://pevly.com/anet-a8-vs-a6-vs-a3-vs-a2/ AUTO LEVEL Anet a6 autoleveling
Anyone tried Varathane water-based Wood Stain on Hatchbox Wood PLA? I printed parts for a Harry Potter wand with HatchBox Wood PLA, now I want to apply stain. I intended to try MinWax PolyShades wood stain, then noticed Varathane water-based wood stain is available at my local Rona hardware store. This seems a "healthier" option, maybe not as fussy to apply as PolyShades, anyone had experience with this product?
I ended up using Saman brand water-based wood stain, also from Rona hardware (a brand of Lowe's Canada). The selection of colours for Saman stains was greater than the Varathane stains. I applied a single coat of colour #117 "Chamois" to the stem portion of the wand, and two coats of colour #120 "Dark Walnut" to the handle. I did not apply a varnish or sealer. The results seem acceptable, given that I Am Not A Carpenter, and this is my first go at printing and finishing wood PLA! Overall, I'm quite happy with the result. I left the PLA mostly unsanded to take advantage of the layers' wood-grain appearance.
Wood stains (as opposed to dyes, paints, etc.) work by having large particles that become lodged in the grain of the wood, yielding a result that varies in intensity with the grain of the wood and thereby brings out its beauty. It's unlikely that they will do what you want, or anything reasonable, on PLA that has wood particles mixed into it. You might be able to find some types of dyes that will work. I've used wood dyes on woods that don't take stain well and have had good results, and if the PLA+wood material consists of a significant amount of wood, it seems plausible that wood dyes might work well on it.
What is the best 3D modeling software for a beginner on a 3D printed mini barrel project? My goal is to 3D print a 5 liter miniature barrel with a side stand, similar to this wooden one on Amazon. I want it to have a removable top so that a boxed wine bladder may be put inside, and there should be a hole on the top as well so that the spigot may stick out and be used. I have no experience with 3D modeling or printing, but I have access to a public 3D printer at my local library. I know you can print parts individually (ex. curved wood-colored sides with staves and holes to interlock and make up the body of the barrel, the metal-colored hoops to go around the barrel). I don't know what software to use, though. I was thinking of starting to learn Blender? Would that be effective for this project?
Your question begins in an inappropriate format for StackExchange, but you've ended it with one more appropriate by asking if Blender would work. If you are willing to take the time to learn Blender, you are certain to discover that it will do as you require, and much much more. Your referenced model could be created using engineering-type design software such as Fusion 360 or SolidWorks or many of the free packages, but the free-form aspect is more suited to the flexibility of Blender. 2020 UPDATE: Fusion 360 now supports a sculpt feature, which combines organic modeling with the engineering-type for which it is previously known. Even though Blender is not an engineering-type program, it has internal support for precise modeling. Should you learn to use those features, you get the best of both types of software. If you construct your model in the software in segments/pieces as you suggest, your result will have greater flexibility at the printing stage, specifically with respect to color and filament choices. Instead of wood-colored sides, you can use wood-simulated PLA filament! Depending on the printer at the library, you could also use filamet, a filament containing 88 percent metal for the hoops. I use Blender for some aspects of modeling, often importing the STL into Meshmixer to address things I've not yet learned in Blender. I hope your reference to 5 liter is the original size and that your model will be a miniature of it. A 3d printer with 5 liter capacity would be a wonderful asset at the public library!
To the people saying 123design. I can only say I have been a blender user way before 3D printing was a thing and just cant get around learning 123design. It just feels so limited as it only have a few tools. Yes blender might have a hard learning curve but it does pay off in the future as you wont be limited to 3D printing models and its crucial to make more complex models
Nozzle heats up past the setpoint and increases I made a few successful prints since I got my CR-10 two weeks ago and I didn't run into any major trouble. The printer is new. Today I set it to "preheat" mode while I was preparing the SD card with the settings being 210°C for the nozzle and 60°C for the bed. When I wanted to start the print I noticed that the temperature showed as "actual temperature" on the printer's screen showed 233°C and it was going up steadily while the "requested" temperature was still 210°C. Thiking it might be a mis-manipulation on my part I powered it down for a few minutes (I got scared by the high-temp) and then powered it back on. I then immediately requested the print to start. The CR-10 heated up to the proper value, started printing and kept heating the nozzle. I stopped it at 217°C. I looked for an answer on the internet but all I could find is people having trouble with the nozzle not heating at all ...
This is not an easy one to solve, the firmware of the printer should be keeping the printer at a certain temperature depending on the temperature setting and the current value. If the firmware is not able to keep the temperature at the requested level, but goes beyond that level, that could be considered "strange". As it measures the temperature (and reports it on your display) it must know that it is over the limit and thus should not power the hotend. In this process there are a few possible candidates for you to look at: Check for a faulty MOSFET (sort of an electronic switch) on your controller board (is it leaking current to the hotend?). Check and or update the current settings for the PID values (settings for the control loop of the hotend). The PID values control the overshoot of the temperature. E.g. is this is very large overshoot? When incorrectly configured the temperature can get higher, but normally should never increase to infinity, are you sure it keeps rising? The determination of the new values is called PID tuning. Important commands (that need to be send over a USB connected printer with a 3D printer terminal application like Repetier Host, OctoPrint or Pronterface): The M503 G-code command shows the current settings (somewhere in the heap of all settings). The M303 G-code command can determine the values. Reflash the firmware Replace the printer controller board You could replace the thermistor and the heater cartridge (just to be sure, most definitely not the problem, but they are really cheap to replace). The thermistor works as it reports the temperature, and the heater element doesn't get powered by itself. As suggested below the most likely candidate for your problem is the MOSFET. These are pretty easy to replace (depending on your board) or replaceable by an external MOSFET module (if you happen to have one lying around).
A similar condition occurred in my 3D printer. I solved the same. I checked all my connection and I came to know that I connected the thermistor of the extruder in the wrong port. So just check the connection of your thermistor. Actually my 3D printer circuit board frequently failed because of over current. I then added a multimeter in series with my power supply and the load, also a voltmeter across the voltage regulator. I then corrected every motor driver DRV8825 to a reference voltage of 0.6 V so that the maximum current supplied to each of the motor will be less than 1 A. All these made my circuit checked ok. Then I uploaded the G-code, but I couldn't print because my extruder gets heated about 280 °C and got reset and suddenly shut down the extruder supply. This continues to happen. Then I reinstalled the Marlin firmware and I also changed the port of my thermistor. Now my 3D printer is ok and prints nicely. I also faced another problem while I gave the print command - it showed that the extruder and bed had started heating but it actually was not heating. So I reinstalled the firmware again and this fixed it. Now my 3D printer works OK. You just try for these steps: Check the connection. Reinstall the firmware. Try manually preheat the extruder, and just see whether it heats to infinity. These are my working experience please try for these, I think it will help you. Just don't leave it you will get the solution. Keep on trying.
A single angle, single phone model windshield mount -- a job for a 3D printer? My end goal is getting high quality dash footage from a 6 month road trip I'm going on. From my research, very few dash cams support 4k 30fps filming, and the ones that do overwrite their own footage really quick, so instead of that I'd like to use my iPhone. I have a wide angle lens for it, and I figure I can mount it to my windshield, behind the rear-view mirror. But here's the problem: there are no windshield phone mounts that allow for the angle I need. They're all designed to point the phone screen at the driver, and the little ball joints that let you set the angle just don't work to point the camera straight ahead. I've tried like 5 different ones, and they all have this problem. What I need is a solid thing that sticks to my windshield and holds my phone in the correct direction. Once stuck, it never needs to be adjusted. I think I could use 3M strips to stick something to the glass, so the only remaining part of the mystery is this: A piece of plastic the exact right shape to hold my phone and point it at a specific angle. My question is: Is this a good use case for 3D printing? And if so, how would a complete amateur get started on this? A few more requirements that I'm not sure if 3D printing can meet: It would need to withstand heat, as it would be left in the car on hot days in the south. It can't be too brittle, as speed bumps and dirt roads will knock it around a fair bit, and it has to support a large phone with an added lens.
You'd need to print in a heat resistant material - ASA for example - and design the part for your needs, but this project is certainly feasible and doable with 3D printing. If that isn't enough for you, you could drill a hole to the internal cavity (it's best to have an infill pattern that does not split the internal cavity into several ones. Gyroid is one of these) and fill it with resin to make it even more sturdy. With the right design, you could also go for SLA/DLP aks resin printing, but I am not well versed in the properties of printed resins but that they have some of the best inter-layer bonds. If you don't want to get a 3D printer yourself, order the part printed, which usually comes cheaper than an entry-level printer with better quality for a one-off project as you won't have to learn the ins and outs of your printer and how to ensure the quality in the material you choose. Some print services also provide really exotic materials.
Why not try a GoPro camera? They now have 4K, image stabilized camera with all kinds of accessory mounts. https://shop.gopro.com/cameras/hero7-silver/CHDHC-601-master.html
What is the biggest size of an object that iBox Nano can print? Since iBox Nano is the smallest public-production-available 3d Resin printer (and the cheapest so far), I assume it has a huge size limitation. So far I've only seen pictures of its outputs that are miniature things. I've never tried it nor have I seen it in action in person so I'd like to be sure. For example, my 3D models are of the size of beads to figurines to a standard sized pencil cup holder. I want to know in inches or millimeters the dimension (width, length, height) of the biggest possible object the iBox Nano can print.
On their website, I found the following picture, which states a build area of 40 mm x 20 mm x 90 mm (1.57" x 0,79" x 3.54").
From their website I found a comparison between 300 microns down to 50 microns print quality. My answer would be somewhere around this range.
Adding a Bowden extruder to a direct drive setup I currently have a single extruder (direct drive) Tronxy X8. I am thinking about adding a second extruder, that I would use only occasionally. I really like the characteristics of the direct drive extruder, so I would not want to switch over to dual Bowden. Also adding the weight of a full second direct drive extruder for only occasional use does not seem useful. Is it possible to use a direct drive extruder as main extruder and a (possibly detachable) bowden extruder as secondary extruder? Is there anything in particular that I need to watch out for when mixing extruder types?
I'm an amateur and I used dual extruder printers but never built any. It's probably a good idea, take care about the purge, retractions distances that may vary. On the filament feed side there shouldn't be interference from the different flexibility. Still it maybe good to place one reel on one side and the other on the opposite. Now the most complex thing is to make both nozzle not interfere. A technique is to rotate by few (like 5°) the dual setup to give few millimeter more to one of the two nozzles. If both heads are at the same height, you'll need to keep them hot, and avoid the passive one to leak. It would be good also (but not mandatory) to have a purge routine.
I'm an amateur and I used dual extruder printers but never built any. It's probably a good idea, take care about the purge, retractions distances that may vary. On the filament feed side there shouldn't be interference from the different flexibility. Still it maybe good to place one reel on one side and the other on the opposite. Now the most complex thing is to make both nozzle not interfere. A technique is to rotate by few (like 5°) the dual setup to give few millimeter more to one of the two nozzles. If both heads are at the same height, you'll need to keep them hot, and avoid the passive one to leak. It would be good also (but not mandatory) to have a purge routine.
Extruder motor not running during print job but working after I preheat the extruder and test it I have a Prusa i3 (from FLSun). I have assembled the printer, calibrated it (fairly decently), and have successfully demonstrated extrusion manually. However, when I try to run a print job (the G-code clearly contains E commands), the extruder motor never rotates. Is this a common problem? Any suggestsions for how to resolve/troubleshoot? I can confirm that the extruder was pre-heated for PLA (up to a generous 220 C). I load an STL file into Repetier v.1.0.2, use the built-in slicer (Slic3r) to generate the G-code. I include a sample of that below: ; generated by Slic3r 1.2.9 on 2017-11-08 at 10:54:07 ; external perimeters extrusion width = 0.50mm ; perimeters extrusion width = 0.72mm ; infill extrusion width = 0.72mm ; solid infill extrusion width = 0.72mm ; top infill extrusion width = 0.72mm M107 M104 S200 ; set temperature G28 ; home all axes G1 Z5 F5000 ; lift nozzle M109 S200 ; wait for temperature to be reached G21 ; set units to millimeters G90 ; use absolute coordinates M82 ; use absolute distances for extrusion G92 E0 G1 Z0.350 F7800.000 G1 E-2.00000 F2400.00000 G92 E0 G1 X76.952 Y76.952 F7800.000 G1 E2.00000 F2400.00000 G1 X78.055 Y75.850 E2.04824 F1800.000 G1 X78.650 Y75.324 E2.07282 G1 X81.121 Y73.387 E2.16997 G1 X81.780 Y72.933 E2.19472 G1 X84.467 Y71.309 E2.29186 G1 X85.167 Y70.940 E2.31635 G1 X88.143 Y69.602 E2.41730 As an update, I can no longer reliably get the extruder motor to run (even when I pre-heat) from the LCD. Sometimes, it works fine, but other times it doesn't seem to. Will confirm whether switching the motor connector to the z-axis to test the motor works when I get home this evening. I can confirm 1) that there is no tension pulling on the cable connections during the run, 2) the mainboard and motor connector aren't burned, cracked, and the mosfets have heatsinks on them, and 3) the extruder gear is fixed and rotates freely on the motor axis (I tested this by switching it over to the z-axis connection temporarily and confirming rotation). Here's a photo of my motherboard:
According to all the information: No reassignment of the extruder in the Gcode You can extrude if you push the filament by hand Extruder motor works when switched to Z driver Z Motor works I come to the conclusion that some of your electronics or the motor have fried. What have fried? I tried to make a step by step check: Switch drivers between E and Z, but plug the motors as usual: 1) Does the extruder extrude? Yes: Your driver has fried No: 2) Were you able to move the Z motor? No: Motherboard And driver has fried Yes: 3) Plug in the E motor on the Z driver, can you make the E motor turn? Yes: Motherboard fried No: Motor fried HTH and keep us updated if it worked or not!
This should have been a comment: when you say "manual extrusion", do you mean moving filament by hand or using extruder position commands from printer menu? Is extruder motor working at all? You can for example plug it instead of a Z-motor to check. Suggested points to check: Motor wire connections, both on extruder and mainboard. Are plugs experiencing tension (being tilted, pulled out) from improperly fixed cables? Hardware on mainboard next to motor connector. Does it look burned, cracked, lacks heatsinks on mosfets? Is extruder gear not fixed and rotating freely on extruder motor axis? You will hear sound of working motor, but it will look like it's not pulling the filament.
Nozzle jumps off the bed when using Marlin Mesh Bed Leveling I am using Marlin firmware with a RAMPS board on an Anet A8 printer. The bed size for the printer is 220 x 220 mm and that is stated in the configuration.h file. When using mesh bed leveling, the nozzle jumps to the first corner on the bed perfectly after setting the x-min to 5.0 but the next two points are off the end of the bed. Here are my settings: // Travel limits after homing (units are in mm) #define X_MIN_POS 5.0 #define Y_MIN_POS 0.0 #define Z_MIN_POS 0 #define X_MAX_POS 220 #define Y_MAX_POS 220 #define Z_MAX_POS 240 What could be my issue?
None of the answers address your question to solve it! The only sensible contribution comes from a comment of @TomvanderZanden. For the sensor to stay within the limits of the bed (considering the offset of the sensor and the size of the hotend carriage) you need to define where the sensor (plus carriage) may go to (to keep the sensor also on the bed, you also need to define the sensor limits). The suggested constants are important in defining the size of the bed and the build volume, but changing them does in fact do not solve your problem. Your problem is that you need to address where the sensor may probe within that specified volume i.e. the sensor probing area. I use ABL (Auto Bel Leveling) on my Anet A8 using a left-front mounted sensor at position (x = -26 mm; y = -40 mm seen from the nozzle center). In order to have the sensor on the bed area without running into the limits of the carriage, you have to calculate (yourself!) what the dimension of the auto bed leveling area is. This is because the sensor is off set from the nozzle. If the sensor reaches for the whole bed, you need extra space on the axis. To explain this, if your sensor is at the left front, as in my case, the amount of extra space you have left on the right side of the X carriage will determine how far the carriage may go and thus limits the right probe position. If there is no extra space on the carriage (the safest assumption) you just need to add the sensor offset to the maximum bed size (what the nozzle can reach). E.g. my probe X offset is -26 mm. The maximum probe distance is therefore -26 mm + 220 mm = 194 mm. This means that you need to set the following constants (amongst the settings to enable ABL; the probe and the type of leveling...) in the configuration.h of your Marlin Firmware installation: #define X_PROBE_OFFSET_FROM_EXTRUDER -26 // X offset: -left +right [of the nozzle] #define Y_PROBE_OFFSET_FROM_EXTRUDER -40 // Y offset: -front +behind [the nozzle] // Set the boundaries for probing (where the probe can reach). #define LEFT_PROBE_BED_POSITION (0 + 10) // 10 #define RIGHT_PROBE_BED_POSITION (220 - 26 - 10) // 184 #define BACK_PROBE_BED_POSITION (220 - 40 - 10) // 170 #define FRONT_PROBE_BED_POSITION (0 + 10) // 10 Furthermore the settings you do mention need to be: // The size of the print bed #define X_BED_SIZE 220 #define Y_BED_SIZE 220 // Travel limits (mm) after homing, corresponding to endstop positions. #define X_MIN_POS -33 // Distance from end switch to X = 0 of origin #define Y_MIN_POS -10 // Distance from end switch to Y = 0 of origin #define Z_MIN_POS 0 #define X_MAX_POS X_BED_SIZE #define Y_MAX_POS Y_BED_SIZE #define Z_MAX_POS 240 The -33 and the -10 define how much the nozzle needs to travel from the endstop position to the print origin! This is not necessary to change when using ABL with a sensor (unless you are using a different print head carriage with a different center of the nozzle). Why these values are -33 and -10 (or values close to this; e.g. for my printer they are -36 and -8) is explained in this answer.
The problem is in the code. Please use these: // The size of the print bed #define X_BED_SIZE 220 #define Y_BED_SIZE 220 // Travel limits (mm) after homing, corresponding to endstop positions. #define X_MIN_POS 5 #define Y_MIN_POS 0 #define Z_MIN_POS 0 #define X_MAX_POS X_BED_SIZE #define Y_MAX_POS Y_BED_SIZE #define Z_MAX_POS 240 Your problem will be ok.
How do I concave an image to create a 3D file for use in a 3D printer? How can I print an embossed image in a concaved shape? Like a big saucer. I will use this an a mold for a project. So far I've found lots of software with huge spread of features. It's sort of overwhelming. There is lots of ways to create images into 3D printable objects but to add the extra step and concaving that image is harder to find out. How would you do it? I'm open to suggestions. I'm new to 3D printing and would really appreciate the help.
What you describe, sounds like you want to create a lithophane; a pattern etched or engraved on a thin translucent base material (in your case a bowl) that can only be seen clearly when backlit with a light source behind it. Apparently you want to use it for another purpose. Special software and or scripts that transform the image to the base material exist. An example is e.g. this sphere which becomes a globe when lit from the inside. Recommending a tool for creating such bowl is a little out of scope as these types of questions become outdated very quickly as technology changes or tools cease to exist. With the provided information you should be able to find software that is able to provide what you want to do.
You may want to look up https://3dp.rocks/lithophane/ it lets you do things like that
How can I set the 1st layer after a raft in Slic3r to print at 1st layer speed When you add a raft in Slic3r, the first layer of the raft prints at the first layer speed. After the raft is finished, the first layer of the print prints at the standard speed. How can I make the first layer of the actual print slow down to the first layer speed?
This is still work in progress, and here is what I have so far, but first: A useful alternative for similar problems: A problem very similar to this would be to use different settings for different parts of a model in Slic3r. For most settings, this can be achieved through modifier meshes. Post processing scripts: As far as I know, Slic3r does not give you the option of setting the speed of the first layer after a raft directly, but they do allow you to run post processing scripts; that is, to automatically run a set of operations - programmed by you - on the g-code output. Although far from trivial, you can in theory make a program that runs through the output g-code, adjusts the settings to your preference, and then saves it again at the target destination. Tuning overall printer speed through g-code: As it turns out, there is a simple g-code command that sets the overall speed of your printer's operation: M220 S[some number] ; see the link above for compatible firmware A newsletter from Reprapwold explains that: For example M220 S50 will reduce the speed to 50% of the original sliced G-code. If you want to hurry your print to the finish in time for dinner, use M220 S200, to print twice as fast (200%) In other words, just like some printers allow you the change speed mid-print, you can use the M220 command to override the current speed used, either through a user interface such as PrintRun, or by fiddling with the original g-code itself. Manipulating the g-code output to adjust speed settings: The easiest way to achieve our goal would be to manually manipulate the output g-code file through a text editor, and insert our M220 command in appropriate places: Set M220 S50 just before the first perimeter layer (after the raft's interface layer), to slow down the first layer of the actual model. Set M220 S100 sometime after the first perimeter layer, to resume the normal speed settings. In order to do this, though, we need to be able to distinguish these two points in the g-code output. Distinguishing insertion points: Slic3r offers a setting under Print Settings -> Output options -> Verbose G-code that - when enabled - inserts written comments all throughout the g-code files generated. If one inspects a g-code file outputted for a model with raft, one will find the comment: ; move to first perimeter point <- lets call this A and ; move to next layer (x) <- lets call this B littered several places throughout the gcode. It is under my impression that the first occurrence of comment A happens right after the raft is finished, and before the actual model is being printed, while the first occurrence of comment B succeeding comment A can be used to set the speed back to normal. It should be noted, however, that the comments in the output g-code does not seem fully consistent, and I would therefore not recommend anyone to automate this logic into a script without possibly finding other, more reliable breakpoints, and thoroughly verify these through several different models. I have not looked into the details of writing an automatic script for this task as of yet.
You shouldn't need to. The purpose of a slower first layer is to help with need adhesion. With a raft the first layer of the model is printing on the raft so it can go at regular speeds.
Open Source 3D scanning I am trying to make a structured light 3D scanner using single camera, light projector and a turntable. After days on Google I did not find any reliable open source project which I can get to work. SLStudio really seemed a good choice but did not compile properly. I was wondering if anyone knows a good open source which they have used with the same kind of setup? Any help would be really appreciated. I am blocked here.
I did find only one 3d scanner which uses structured light. There is many projects using a laser diode. And these systems are completely opensource. Structured Light Structured Light 3D Scanning by kylemcdonald http://www.instructables.com/id/Structured-Light-3D-Scanning/?ALLSTEPS Laser diode Sardauscan http://www.instructables.com/id/Build-a-30-laser/?ALLSTEPS https://github.com/Sardau/Sardauscan http://www.thingiverse.com/thing:702470 ATLAS 3D Scanner http://www.freelss.org http://www.thingiverse.com/thing:1280901 https://www.kickstarter.com/projects/1545315380/atlas-3d-the-3d-scanner-you-print-and-build-yourse/posts/1475722 Ciclop http://diwo.bq.com/en/tag/ciclop/ https://github.com/bqlabs/ciclop 3D(ollar) Scanner https://hackaday.io/project/2021-3dollar-scanner http://www.instructables.com/id/Lets-cook-3D-scanner-based-on-Arduino-and-Proces/?ALLSTEPS "Super Make Something" Tutorial On How To Build a Simple 3D Scanner https://www.youtube.com/watch?v=-qeD2__yK4c http://www.thingiverse.com/thing:1413891 http://diy3dprinting.blogspot.de/2016/04/super-make-something-tutorial-on-how-to.html Different method $15 3D scanner consists of a tub of milk and a smartphone Interesting idea of using milk and phone. http://www.instructables.com/id/GotMesh-the-Most-Cheap-and-Simplistic-3D-Scanner/ http://www.3ders.org/articles/20160423-this-3d-scanner-consists-of-a-tub-of-milk-and-a-smartphone.html
BQ Ciclop 3D Scanner Scan Volume: Bigger than 5 cm x 5 cm and smaller than 20 cm x 20 cm Scanning Precision: 0.5 mm All the necessary parts for Ciclop are included in this DIY kit, which comes unassembled. This Ciclop 3D Scanner Parts List: 1 x Plastic parts( 1 set with 11 pcs) 1 x Arduino Uno R3 with USB cable 1 x ZUM Scan Shield 1 x A4988 Stepper Driver 1 x C270 HD camera 1 x Nema17 stepper motor 2 x Laser 1 x Power supply adapter 1 x 16014 ball bearing 1 x M8 threaded rod (9 pcs) 2 x Acrylic panel 1 x 8mm spiral wrapping band 1 x Set of screw and nuts
Are there biocompatible materials available to the general public? I am currently working on parts for a custom prosthesis. My main concern at the moment is to find biocompatible materials that can be 3D printed from a UP or a Reprap. The piece would need to be in contact with the skin for extended periods of time, probably around 17 hours a day on average. The main concerns I have are: Skin reactions caused by prolonged contact Skin reactions and bruising caused by friction Degradation of the materials due to prolonged exposure to skin secretions and sweat Risks of toxicity in the compounds generated by the aforementioned material degradation Which materials can you recommend? Any extensive data (from testing) would be greatly appreciated.
There are printers designed for medical use, and the manufacturers supply them with varying levels of certification and testing, however I've not seen a filament manufacturer certify their material as bio-compatible separate from the printer. The printing process changes the material slightly in the best case (and significantly with poor temperature control or badly set parameters), so even if bio-compatible filament were found, the resulting product might not achieve the same level of bio-compatibility. If your intent is to use hobbyist level machine for medical purposes, you might simply want to use an interface, such as a sock or a molded/cast polymer that you know to be bio-compatible between the printed part and the skin.
If you want to know something about what you are questioning, it is interesting to you to read http://e-nable.org/resources/prosthetics-students-consultation/ That's a link with complementary information how to print 3D as a volunteer. I know that is not enough information about products, but I believe Nylon is the best recommend because it is what generally printers are using, though. This website supports information with Dr. Chang by the e-mail eagle01@rocketmail.com but there insnt information about the components used in the prosthetics, but I know that Nylon is the best suited to it nowadays. If you prefer there is a site with makers loging information about PETG: https://www.matterhackers.com/news/how-to-succeed-when-printing-with-petg-filament
How to transport a 3D printer? Dismount needed? I need to transport my FDM 3D Printer because I am moving. What are the precautions that one should take? Should I dismount the motors and axes? I would definitively unplug the electronics as far as reasonable and fix the motors to the frame so they don't slide during transport. Should I have a housing to avoid dust and other mechanical issues?
Yes, fix the motors and any other loose/movable parts. Remove the bowden tube if it's there, and any other parts that are sticking out. Put the whole thing in a a bag to protect from dust, and put the bag in a box to protect it from getting beat up. Remember to calibrate it when you're ready to set it up again.
You just need to take basic security actions. like fixing all movable parts simple as that
Why the Ender 5 Pro can't remember the axis position or cant move back to home anymore? I am completely new to 3D Printing. I got my first printer a Creality Ender 5 Pro yesterday. My problem I did shutdown the printer without the axis being in home position (X: 0, Y: 0, Z: 320 instead of X: 0, Y: 0, Z: 0). I thought this should be no problem but after turning the printer on again the info screen showed the axis as 0, 0, 0 again... So I can't move the Z axis up again because the printer thinks its already at 0. NOTE: On the Ender 5 the bed is lowering for the value of the Z so 320 is the lowest and 0 the highest. The motor works because it tries to go down further if i increase Z position but I am scared of damaging the motor because it can't move further but it tries to (judging by that weird sound). My question Is this a normal behavior that the printer axis cant remember its position? Because I think as I built the printer the axis were also not at the 0, 0, 0 position and on the first start they moved back without any problems. Has someone any ideas how to solve this? Or is this a broken printer? PS: I could replicate this behavior on X and Y as well (moving them with prepare->move axis and then shutdown the printer) but here for I can easily disable the motors and move them manually back to 0, 0 which isn't the case for the Z Axis. I hope I could explain that understandably.
Once you pull the plug or disable power to the stepper motors, the printer forgets its location. That is perfectly normal and exactly how it is supposed to work. The printer knows where the printer volume is once you have "homed" the printer. Homing is done prior to printing with G-code G28 which should be present in your start G-code script of your slicer. Once homed, the offsets from the endstops determine where the origin of the printer is and the maximum dimensions determine the build volume. After you switched on the printer, the printer doesn't know where the origin is and movement is limited. E.g. when the following constant is defined: #define NO_MOTION_BEFORE_HOMING no movement at all is possible before the printer is homed, this can help prevent destroying the printer.
Slightly unscrew the motor. Manually rig the Z-axis above its lowest point. Rescrew the motor back into place tightly. Then auto home it.
Trouble printing Poppy Robot with SpiderBot For a university project, my partner and I need to print the robot Poppy. This is an open source robotic project, poppy-project.org. We are printing it with a double extruder SpiderBot with PLA and HIPS as support material. Our principal issue is the weakness of the pieces we print. It prevents us from removing the support material without damaging the piece. We don't have the chemicals to dissolve HIPS. Have you some advice to make the pieces stronger, or a more gentle method to remove the HIPS? Thanks for the replies
Having a bit more experience since my comment post above, I can offer up a possible solution. If your printer is able to print ABS, it works great with HIPS as support material. I've been printing boatloads of ABS models. When support is needed, the HIPS supports will bond reasonably well to the ABS, but when cooled, flick off with a sharp pointed instrument such as a razor knife point. As an additional reference, a good bond to the print bed is required for both ABS and HIPS and if your slicer allows support density adjustment, increase the support in that manner. My slicer (Simplify3D) would print an accordian type of support unless instructed otherwise. I set up for alternating 45° patterns making a denser but also more vertically rigid structure. If you cannot print ABS, PLA and PVA as Ogre55 suggests is about your only option. I see from the 'net that the SpiderBot Pro supports ABS, which implies the basic version might not have a heated bed.
Why not try using PVA(Polyvinyl alcohol) for support material. I use it for my support material with PLA, and it dissolves in water. here is a link to some on Amazon. https://smile.amazon.com/eSUN-1-75mm-filament-natural-0-5kg/dp/B00MVIQASU/ref=sr_1_3?s=industrial&ie=UTF8&qid=1508539512&sr=1-3&keywords=PVA
Bed design for magnetic easy release when printing PLA I print on a glass bed covered with BuildTak. Prints stick well, release is difficult. They sell a mod, self adhesive magnetic sheet that holds a steel plate which in turn you put the BuildTak on making it easier to get your prints off by twisting the flexible steel as opposed to chiseling your print off a glass plate while attempting to not peel chunks of the expensive Buildtak off. Magnetic sheets have different strengths proportional to thickness in milimeters. I can get 0.03 mm locally which provides 85 lbs pull per sqft. and was wondering if this might work. It's hard to find the stronger pulls, 0.045 mm and 0.06 mm which can pull 115 and 145 lbs respectively. I'd prefer cut my own new bed sandwich rather than pay 90 bucks for one. Has anyone made a bed using adhesive magnetic sheet> What thickness magnetic sheet has worked for you? What thickness steel are you using? Assume it has to flex easily to twist the part of and also flexible enough for the magnet to pull it flat. With respect to the comment by tbm0115: I don't think you need that much magnetic pull for this application. Adhering basic coin magnets to a material in a large array should secure the material appropriately for a desktop printer. I see where you are going with coin magnets. Have you tried this? In the original Makerbot Cupcake this is how it was done though with larger build surfaces you get warp on metal plates. Most metal sheet comes from the plant on a roll. Though your metal looks nice and flat, temperature changes will cause it to warp and as you know with SLA printers, a changing bed can cause a failure. This happened on the original Makerbot replicator 1st release followed by using a machined plate as a replacement.
Follow up: I used the 6mil magnet, cut a sheet of thin steel, (about .5mm) to fit, added a layer of Buildtak and now it's much easier to remove prints, build surface is very flat. I assume you could deform the metal beyond the magnet's ability to flatten it but it works quite well. The metal was cut with sheet metal snips, risky as they can bend the sheet. Probably better done with a saw though.
Try using buildtak flex. https://www.buildtak.com/product/flexplate/ or cheaper alternatives: http://www.printinz.com/printinz-plate-3d-printer-bed-upgrade/
How do I give 3D-printed parts in PLA a shiny smooth finish? The surfaces of my printed parts using PLA plastic look rough and uneven. Would changing filament to a better one make any difference? If not, what kind of methods can I use to achieve a smoother finish for my for 3D-printed objects?
PLA parts can be finished with a coat of epoxy like XTC-3D from Smooth-On. This will smooth out the part and give it a pretty nice shine. I've also had a fair amount of success sanding prints, giving them a coat of automotive filler primer, and using glossy spray paint. You can also get great results with an acetone vapor finish if you're willing to switch to ABS. Though that will require a heated bed and can be a bit more finicky to work with than PLA.
I suggest using XTC-3D since your prints are small also you can add some UVO pigments in. It's difficult to use in large prints because has small working time. Anyway here is a print I did for a friend: This now dead link shows more images and say how it is done.
How to calculate the strength of a printed object? If I'm working with standard PLA, and I want to print a box that I can stand on without any risk of it breaking, is there any good way to calculate the appropriate print settings? I know that structural strength comes from the infill. Knowing this, and knowing the dimensions of the box, the weight of my body, the surface area of my shoes, and the material I'm working with, is there any good way to determine the minimum infill percentage I'd want to use in order to safely bear my weight?
Strictly speaking, it is difficult to do calculations on these materials, but not impossible (I've heard about a few commercial analysis tools that do that). The FDM process (Fused Deposition Modeling) creates a product based of fused slices of material causing an anisotropic material (this means that the properties of the material are different in different dimensions). Basically, your product will be quite strong and similar in the X and Y directions, but fragile in the Z direction (layering direction). You can imagine that every layer may be a seed for cracks to grow when you're pulling at the part. When applying a compression load on a product like in your example, the walls need to be strong enough to hold the pressure (not all of the load as, based on the type of infill, the infill also can/should take part of the load!) and need to be of sufficiently high percentage, not only to take part of the load, but also support the walls to prevent buckling. I remember that stress calculations for buckling are difficult and require FEA (Finite Element Analysis) for more complex objects other than bars or beams. I think it is difficult to determine or calculate the infill percentage based on the compression load beforehand as you do not know the exact material properties and the buckling behavior. You do know that a 100% infill will give you enough strength and support, you could try to print at a lower infill, e.g. 75%, and test if that works for you.
A fast way to do this is by using SolidWorks. You can draw the box in it and run a simulation test with the max load expected. Here is a link on how to make dynamic load simulations work in SolidWorks, How to apply dynamic load in solidworks simulation ? The catch in the process is that SolidWorks takes cubes and most objects as complete solids, i.e. 100% infill in 3D printer terms. You would have to actually design your infill pattern into the cube so as to get the best and most accurate result.
Is Tetrahydrofuran viable for dissolving PLA and sticking PLA objects together? According to Wikipedia, Tetrahydrofuran dissolves PLA. It also apparently dissolves PET. THF is considered a relatively nontoxic solvent, with the median lethal dose (LD50) comparable to that for acetone. Wikipedia further states that it's not particularly dangerous provided you keep it away from air circulation, so that it does not form peroxides. One danger posed by THF follows from its tendency to form highly explosive peroxides on storage in air. So well, this seems acceptable compared to other possible solvents which are all ridiculously dangerous. The question at hand is now of practical nature: Is Tetrahudrofuran actually viable replacement for acetone, which only works with ABS? I am asking because the fact that it does dissolve PLA does not really at all mean that it will work well. It could damage PLA structure, be more toxic than wikipedia says or not dry well.
According to Shuichi Sato, Daiki Gondo, Takayuki Wada, Shinji Kanehashi & Kazukiyo Nagai: Effects of various liquid organic solvents on solvent‐induced crystallization of amorphous poly(lactic acid) film in Journal of Applied Polymer Science, Volume 129 Issue 3 (2013), p1607-1617source, Tetrahydrofuran is classified as a solvent for PLA. The specific entry on page 1608: Group Solvent Solvent type dd dp dh dt Result Ether Tetrahydrofuran Polar aprotic 16.8 5.7 8 19.4 soluble The values dd dp dh and dt are explained on page 1607: The effects of 60 liquid organic solvents on PLA are systematically investigated using the Hansen solubility parameter (HSP). The HSP is one of the digitizing methods for analyzing the interaction between polymer materials and organic solvents. In HSP analysis, all solvents have three parameters: energy from dispersion bonds between molecules (dd), dipolar intermolecular force between molecules (dp), and the hydrogen bonds between molecules (dh). All solvents were characterized by a point in a three-dimensional structure at which dd, dp, and dh are plotted on three mutually perpendicular axes. Generally, if the HSP values of the various organic solvents are near that of the given polymer, the solvent is considered compatible with the polymer material. The factor dt is the total Hansen solubility parameter - the bigger this is, the better it is a solvent. 19.4 is a rather good solvent but extremely explosive: in air 20000 ppm (2%) are explosive and thus the allowable concentration in an area is 2000 ppmsource Alternatives A similar potent solvent would be Benzene (dt = 18.6) which more easily available and less explosive, but more deadly (10000-20000 ppm fumes) and has an allowable concentration of 500 ppmsource Ethyl-acetate is also a solvent (dt = 18.2) and more available. It is explosive at an equal concentration as Tetrahydrofuransource, but it has just about half of its vapor pressure (73 mmHgsource vs. 132 mmHgsource), so can be stored more safely, and is less aggressive on the body. It is sometimes used to Smooth PLA via vapor chambersource, and only comes at a price tag of ca. 90 €/l for the pure stuff and also is used in some nail polish removers, put into a safe mix. Acetone is classified as a better solvent (dt = 20.1), and it is known to act as a glue and to soften PLA with some exposure time, but from experience, it can't smooth it. It is available most easy (nail polish remover and in the home depot), and is the least deadly option. Propylene-1,2-carbonate is classed as solvent and a far better at it with dt = 27.2. It has been used as an alternative to Ethyl-Acetatesource and its MSDS is rather gentlesource. It comes at a price tag of 130 €/l for the chemical-grade liquid. Conclusion Tetrahydrofuran is not a viable replacement due to its explosive properties. It is not an improvement above Benzene, which at least can be stored safely. Using Acetone as a benchmark, Tetrahydrofuran should not smooth the surface in a vapor chamber, as it is a worse solvent than Acetone. It should also take longer to soften and dissolve objects than Acetone, but a heated bath or coating the surface with it could help to generate the needed exposure times. However, its comparable ethyl-acetate has been claimed to be used as a cleaning, vapor smoothing, and brush on surface smoothing agent successfully and can be used better by helping the solubility via heating, which can be done much safer with ethyl-acetate than THF. In a proper chemical mix, its storage problem can be solved too. A better alternative is propylene-1,2-carbonate, which is a better solvent, and much less dangerous. tl;dr: No, Tetrahydrofuran is not able to dissolve PLA in a reasonable time1 without heat activation and it has worse characteristics than Acetone. It could arguably be used to weld parts, but Dichloromethane would be more effective. 1 - Sato, Gondo et al. in the aforementioned paper: Solubility tests were performed for 24 h at 35+-1 °C. which means we have a prolonged exposure of a thin film to a huge amount of solvent. how thin a film? Very thin: The PLA films were prepared by casting 2 wt % dichloromethane solution onto a flat-bottomed glass Petri dish in a glass bell-type vessel and by drying under atmospheric pressure at room temperature. Each solvent was allowed to evaporate for 48 h. The dried PLA films were then thermally treated under a vacuum for 48 h at 70 °C to eliminate the residual solvent and to obtain amorphous PLA films. Afterwards, the thermally treated PLA films were cooled at room temperature under atmospheric pressure.
From what I have read, THF is not super effective, it seems that Dichloromethane or DCM is actually a better solvent. Unfortunately I have no personal experience with either. It is rather nasty though, as the LD50 is just 0.5 to 5 g/kg[1][2].
Is it a good idea to include thermal fuses in a DIY 3D printer design? 3D printing should be relatively safe, however, the inherent nature of 3D printers, with all of the heated parts, constitutes a fire risk. A well designed 3D printer should be designed to be as safe as possible, especially one used in the home... Yes, the recommendation is, when printing, to watch the 3D printer at all times and never leave a print unattended. However, with some print times lasting hours and days, this is not always feasible, nor practical. So, some inbuilt safety features should be included, to at least mitigate the risk of fire, to some extent. Is the use of thermal fuses1,2 a good idea3? Would you use more than one? Where should one place a thermal fuse? Next to a particular component, or free standing, in the air, to get an average, rather than highly localised temperature? Against which components should a thermal fuse be placed? There are a number of places to choose from, such as next to: The hotend? The heated bed? The extruder? Each of the stepper motors? The power supply? The RAMPS stepper motor drivers? Of lesser import, which type should one use4, radial or axial? Has anyone added thermal fuses to their 3D printers? Or has anyone examined where the thermal fuses are placed in commercial 3D printer designs, if used at all? Background I have recently found myself having to repair rice cookers and fans in Thailand. In those, it is very often the thermal fuse (axial thermal fuses for the rice cookers and the square "radial" types for fans) that requires replacing, as they have blown before the device got hot enough to start a fire. This got me thinking about their use in a 3D printer. Footnotes 1 We are not talking about the standard, replaceable, thermo-fuse,or fuse, which blow upon a current surge, short-circuit, etc. These are thermal fuses that contain metal connector within them that melts (permanently) at a specific temperature (typically ~135°C), thereby breaking the circuit. 2 Nor am I referring to resettable fuses (AKA PPTC, multifuse, polyfuse or polyswitch) 3 Would a thermal fuse be preferable to thermal cut offs, in the case of fire? 4 The thermal fuses used in rice cookers are the axial type, and in the motors of fans are the radial type.
Whether you should use a thermal fuse or not depends on what other safety measures you've taken. You can't look at the safety features of a printer in isolation, you need to look at what other measures are in place. The main fire hazard in printers is unfortunately (still) the fact that some manufacturers use underrated connectors on their boards, and that some users put bare wires in screw terminals or use inadequate torque when tightening terminals. As the wire works itself loose, it starts arcing and burning the connector. A thermal fuse does not help in this situation (unless you place thermal fuses near all of the connectors, which is impractical). Instead: Properly tighten screw terminals, check them, and consider using proper wire termination (crimp lugs). Use strain relief on wires. Make sure wires don't rub against anything, and guide them so they do not bend in a tight radius. Since the extruder (or print bed) is constantly moving, those wires are subject to fatigue. Make sure connectors (especially those for the heated bed) are rated for the current running through them, and solder wires directly to the board if necessary. Using a regular fuse may protect against wires shorting against each other should their insulation be damaged. Fuses are usually already integrated into the main board. Most firmwares include some variant of thermal runaway protection, a feature that monitors the heaters and shuts the printer down if it notices something gone wrong. This protects against: The thermistor coming loose/reading incorrect values/etc... but not against: Bugs in the firmware itself Failure of the MOSFET Most printers use MOSFETs to switch power to the heating element. Unfortunately, when MOSFETs fail, they usually fail closed (i.e. conducting). This means that, even if the firmware detects something has gone wrong, it won't be able to do anything about it. Solid State Relays (TRIACs) can fail in the same way. To protect against this, mounting a thermal fuse (or resettable bimetallic switch*) on the heated bed may be a good idea. However, thermal fuses with ratings up to the operating temperature of a hotend do not appear to be available so this is not an option. Attaching the fuse physically to the part it is monitoring is the most reliable, but for instance with the hotend (if you wanted to protect it all) this might not be feasible to the high temperatures involved so you'd have to settle with monitoring the air temperature close by. Also consider thermal balancing. A thermal fuse is unnecessary if the component can not overheat to begin with. For instance, most MK2 heated beds struggle getting up to even 100C, so even with a shorted MOSFET they present no danger. However, if you have a powerful high wattage (mains-powered) heated bed, you should definitely install thermal protection. E3D supplies their hot ends with 25W, 30W and 40W heaters. The 25W heater is the safer choice, since it limits the maximum temperature the hot end can get to, while with the 40W heater you can reach higher temperatures (and reach them faster). Barring a very unlikely scenario in which simultaneously (1) the power supply fails and starts supplying excessive voltage and (2) the MOSFET and/or firmware fails, a heater that is sized appropriately to the load it is driving can never pose any danger. I don't think it's common to install thermal fuses on steppers, stepper drives or the power supply (which should have its own protection). For every possible location to place a thermal fuse, you can probably think up a failure mode in which that fuse would save the day, but at a certain point it just becomes overkill. The stepper drivers would likely burn out well before the steppers would get hot enough to pose a threat, and overheating of the stepper driver would probably (violently) destroy it but afterwards it should not pose any threat. Axial v.s. radial does not matter, just use whatever is convenient for your situation. * Note that some bimetallic switches short one of the leads to the (metal) case when tripped, which poses a danger, especially with mains-powered heaters.
I haven't found a good way to flood the entire machine, including the electronics, with CO2 gas or another extinguisher when the alarm sounds. A servo or some other actuator attached to a fire extinguisher, and then attach it to a thermocouple or some other sensor. Maybe a knockoff Arduino, which the small ones are 2 bucks a pop. Since Arduinos are open source, you aren't pirating anything, so clear conscience.
Weird Movement and not homing https://photos.app.goo.gl/O6yPf3sDeV1yhS0C2 I tried to illustrate my problem in the videos above, two of them show the weird movement and the other shows me clicking on the home button repeatedly. Some Info: Marlin 1.1.8 or 2.0.0 (same problem in both) Robotdyn RAMPS 1.4 0.9 angle stepper motors DRV8825 drivers configured at 0.8V Vref Anet A2 Plus stock for all the rest
As far as I can see on the attached videos your homing movement is reversed. as per Marlin, the homing for X shall move towards the left side and for Y to the back of the printer. That could occur when: cable connectors to stepper motors are reversed, or the motor is assembled the other way (you can set reverse direction in Marlin) The other issue is steps/mm calibration need to be done see source below. The high pitch in the video could also point that the drv8825 is shutting down the movement as it is overloaded. please also check that for vref Configuring Vref In order to measure Vref you first need to turn on your printer as you normally would. If you only connecct using USB, but not external power, you get a wrong reading. You need to turn on your multimeter and set it at 2v. Put the red one on the potentiometer and the Black one on the Gnd pin. Both are marked on the images here. Before starting this I read they come with a very high vref setting, and it is recommended to start around 0.5v Vref. After measuring mine, I can confirm they come with a very high initial setting. Mine both came at 1,65v or so! - Yours might be different, which just underlines the importance of doing this. Contrary to normal potentiometer usage, the ones on most copies/clones of DRV8825 are lowered by turning clock-wise, so that is what we will do, to we hit 0,5v on each. - A quarter of a full turn lowered it to 0,7v, - after that it goes very, very rapidly down, so aim for the quarter of a turn + a tiny tad more. If you buy your DRV8825 directly from pololu.com the Potentiometer are dialed up by turning it clock-wise: source
The problem was the logic of the endstops that were reversed
What is the least expensive 3D printer? What is the least expensive 3D printer available today? I am looking for something suitable for general use in a home office.
Depends on your definition of "available" and your definition of "suitable for general use." The cheapest 3D printers are mostly Kickstarter promises that take a year or more to ship, if they ever do. For example, the Peachy 3D printer Kickstarter just imploded and failed. There have been many other failed low-cost 3D printer crowdfunding campaigns. Another low cost Kickstarter printer, the 101Hero, is ongoing now (May 2016), but most competent observers I've talked to don't believe it will succeed at delivering working printers to all backers at that price point. If they do deliver, it will be painfully low-cost components and the printer will not perform well or last long. Stay away from crowdfunding campaigns for your first printer. At best, you get a beta product with lots of kinks to work out. At worst, you get nothing and lose your money. For actual products you can purchase today, there's a wide spectrum of quality/cost tradeoffs. Under \$200 there's nothing credible. The Tiko (\$179) might deliver, but post-Kickstarter units are widely expected to cost more. Around \$200-300 you get into low-quality Prusa i3 kits from China. These aren't a great value -- most people end up spending another few hundred dollars on upgrades to get them working reliably and with high quality. Around \$300-400 you can get an ok 3D printer, often with "chipped" proprietary filament so the vendor can make high profits on locked-in consumables. ("Razors and blades" model.) For example, the XYZPrinting Da Vinci Jr is \$350 but locks you into high-cost chipped filament. The Wanhao Duplicator i3 is currently a community favorite for value-for-money at \$399. The Printrbot Play is much higher quality/reliability but much smaller at the same price. If you get up around \$600, a big range of decent printers opens up. But this is no longer the "least expensive" option, so I won't get into it. If you want to tinker, the Duplicator i3 is a good choice. If you want a machine that just prints, the Play is a good choice. There are other printers and cheaper printers, but most of what you'll find below $400 is going to end up causing pain unless your goal is simply to tinker with printer troubleshooting and upgrades.
You can get an A8 3D printer on Gearbest at 149$, It's a version of an i3, with easy assemble and pre configured, and it have a suprising good quality. I thinks is the best price/quality that you can find at the moment. It prints with a good quality by default and you can upgrade it if you need it without spending hundreds of dollars. http://www.gearbest.com/3d-printers-3d-printer-kits/pp_337314.html
Choice of lead for lead screw This video brought to my attention the 8 mm lead of the Ender 3's Z axis screw, which seems like an exceedingly bad choice from a standpoint of accuracy with respect to common grid alignments in the Z direction. In particular, with the stepper having 200 full steps per rotation, the 8 mm lead consumes all the powers of two out of 200, leaving 25 full steps per mm - and 25ths of a mm are not a typical unit that layer heights/feature heights are going to be in. It seems like a 5 mm lead would be ideal, giving you 40 steps per mm, evenly divisible by 3 powers of 2 and one power of 5, for exact tenths and exact eights. Is there a motivation behind the choice of 8 mm lead? Is this common for other printers, and are there printers that use a 5 mm lead, or 5 mm replacements that work well?
I've not seen trapezoid lead screws with 5 mm lead, you can get 5 mm lead ball screws though. On one printer I use 4 mm lead screws to get native 0.02 mm resolution (so 5 full steps for 0.1 mm, 10 for 0.2 mm, etc.). I also geared down 8 mm lead screws with a 2:1 ratio (e.g. to use a single Z-stepper driving a belt that drives 2 lead screws), works fine.
You are delving into the darker recesses of 3D printing here! ACME threads, anti-backlash nuts, ball-screws, etc. will ALL follow... The simple answer is that the Z-axis screw on all the cheaper 3D printers are pure crap! They use 8mm With triangular threads because it they are cheap and most software works with it. The first major upgrade most folks make to a printer (after the print bed surface) is to replace the Z-axis screw and nut. Many are poorly mounted and/or warped, resulting in Z-axis wobble. Most suffer major backlash - offset only by the fact that most printing only ever goes upwards... The ideal upgrade would be to replace the rod and nut with a good quality ball-screw assembly that is properly mounted. But, if you do change the pitch, make sure you can alter the software to account for it... That said, if your prints are good, enjoy, and don’t sweat the details!
Cura slicer, enforce Z move before layer change I am slicing with Cura and Slic3r and one important thing that I recently took my attention is that cura positions head in start point of the new layer and then lifts the nozzle. That caused my few printouts to fall as they collided with the nozzle. Slic3r behavior is different: it raises the nozzle in last printed point and then moves to a new layer starting point (which for me is more obvious) Is there a way to instruct Cura to lift nozzle before it goes to the starting point of a new layer? excluding Z-hops.
To lift the head to prevent the nozzle to tip over your print you could use an option called Z hop in Cura. Just enter `hop' in the search box on the right side to make those options magically appear (in a recent version of Cura, e.g. version 3.x.x). Other than Z hop there is no default action, or series of commands, per layer to be specified before the start of the layer. There are 2 other ways to circumvent this: The first is saving you G-code to file and open the file in an advanced text editor (e.g. Notepad++). With a (recorded) macro you can find the words ;LAYER:, which are inserted by Cura before each layer starts, and insert a pre-copied list of commands that set the movement in relative mode, move Z up 2 mm, set into absolute mode again. When the next layer starts the extruder goes to the layer start from a 2 mm rise. Write a plugin for Cura to post-process (C:\Program Files\Ultimaker Cura x.x\plugins\PostProcessingPlugin\scripts) the G-code file to inject the code to Z hop before the start of every layer, or a plugin that adds a new option and/or category to the slicer settings sidebar of the GUI.
OK, after going via all the options I found that CURA has a combing mode which reduces retraction and offers another option which is Avoid Printed Parts When Traveling. That solves my problem.
Can I mix ABS and PLA when recycling filament? I'm thinking of recycling some filament from a couple of recently failed prints. I can reuse them in the future for basic prototypes, so I'm not concerned with whatever weird mixture of colors come out (they are of a few different colors). The thing is, I have both PLA and ABS, in small quantities. I originally intended to simply use each one separately, but it occurred to me that they could be mixed. If I recycle PLA and ABS together into one strand of filament, will there be any negative side effects (e.g. reduced strength)?
This is not a good idea. Both filaments have different melting points, that of ABS being much higher than that of PLA. To melt the ABS you have to heat the plastic to the point where the PLA starts to degrade.
yes but you would have to experiment with the settings to see were it afectivly melts.
DLP build plate adhesion Does anyone have tips on improving build plate adhesion in DLP printers? I've heard a thin layer of resin or UV glue applied to the plate will help, but we're not sure if we leave the resin/glue wet, or cure it before we start the print. Apologies for the ignorance here, but I'm just trying to avoid gluing my build plate to the bottom of the resin vat! Any advice would be appreciated. Update I had a product called ProtoGlass recommended to me in another forum, that apparently works as a good build plate primer for the resin we're trying to grow (BlueCast x5). Ordering today, and I'll update further with the results.
I have heard that adding a few pieces of masking tape to the print surface improves adhesion.
I have heard that adding a few pieces of masking tape to the print surface improves adhesion.
3D print a paint roller I was thinking about this question and thought of maybe printing a pattern-drawing roller painter. The question is: is it possible to print with an ink absorbing material that could make a paint roller possible? P.S: I don't own a 3D printer, nor have I any deep knowledge in this matter. I simply want to know if this is feasible, so I can start looking for someone to 3D print this for me. If it's not, knowing beforehand could spare me a lot of time.
While printing a roller stamp or rubber-roll from a flexible material such as [hard]TPU, [softer] TPE, or even a [super soft] foaming flexible filament is certainly possible. In any case, this would create soft, somewhat squishy prints. These prints will work quite easily as a stamp or woodblock printing stock, transferring ink from a pad to paper. The print pattern will depend a lot on how soft the stamp is: the harder, the sharper it will print. A massive roll of this material can behave akin to a rubber roll as one uses it in linoleum printing. A foaming filament might be able to take a little paint in its airgaps, but it will never be as soft and contain as much paint as a foam lacquer roll - making it at best an improvised tool, or one that is chosen for a specific artistic purpose. It behaves more akin to closed-cell foam, while foam brushes and rolls from artist supply are typically open-cell ones. So if you go for a roller-stamp, you'll need to have an ink reservoir in the shape of a soaky-roller that isn't printed. To top it off, it is nearly impossible to print a hairy wall painting roller: the hairs used in them can't be achieved with common print materials and slicers at the time. Even if stringing creates hair of similar dimensions, they are not affixed well enough to not get lost in the paint and can't be reliably created on demand. Notes on pricing: One of the few options for foaming print materials is colorFabb, who was the first to offer such. Most of their foaming filaments cost around 50 €/kg, their flexible NinjaFlex costs about 80 €/kg.
I'm relatively new to 3D printing, but I know some stuff. If you really wanted to print a paint roller with a pattern, I would actually go with nylon or TPR. TPE is okay, but I find that it is less cushy. I would highly suggest buying your own printer for this project. It would be much more cost effective in not only the long run, but for testing different materials. 3D printing services get pricey, so choose your printer wisely. If you do decide to buy your own printer, I would go with an Ender 3 pro. Mine is very smooth, and has an easy-to-use interface.
Issue on 2 corners but not the other 2 I am printing using an Ender3 Pro with eSUN PLA+ 215/45 and I am getting this issue on two corners, the other two corners look fine. Any idea on what can be causing this? The bad corners The good corners
This has nothing to do with speed, temperature, adhesion, and whatever you do, DO NOT extrude more material per line (increase flow rate), as this will make the problem that much worse. This is a fairly simple problem with an even simpler fix: you're over extruding. Reduce your flow rate by 5%, and see if that fixes the issue. It will definitely improve it, but you might need to lower your flow rate a little bit more. What can often happen when your flow rate is set too high is the extra plastic will concentrate at areas of relatively high acceleration (corners and the start/stop spot for a perimeter of a given layer), but depending on the size of the thing being printed and the degree of overextrusion, it won't concentrate at every spot like this on a perimeter. Usually, I see this happen where the perimeter moves start and stop each time, which (again, depends on the slicer and settings) tend to be the same spot for certain models, often a corner. I couldn't say for certain the exact mechanism, except that it seems like the plastic, given the right conditions, prefers to lay down evenly while the excess over extruded plastic builds up (probably carried by the nozzle, since it is hot and the plastic will want to stick to it) until too much has built up for the nozzle adhesion to keep it from sticking to the print, or the nozzle begins to decelerate (late at a corner or the end of a print move), causing the extra plastic to 'scrunch' up, like something that shoved in a distance too short for it. Knowing this, if you examine the corners, it should be quite obvious that this is what is happening. The perimeter is being extruded with more plastic than it should be, and the extra has a tendency to collect all in one spot each time. Sometimes it is one corner, sometimes it is every corner, sometimes it is corners that are maximally distant from each other (since it takes some time for enough excess plastic to build up to over power whatever effects are preventing it from adhering immediately. So in this case, the two good corners were just where not enough excess plastic had built up yet at the nozzle to cause problems). Another possible explanation is that those two corners are simply where the perimeters were started and stopped, but some layers it was one corner, and other layers the other corner. But you can see over extrusion artifacts lower down on the feet (or whatever they are), and your first layer as well. Do not increase flow rate. Do not increase infill. Do not lower your speed. Do not increase your temperature. None of those will help, and increasing flow rate further could cause the nozzle to catch on the print, potentially damaging your hotend. Just reduce your flow rate by 5%. You should see an immediate improvement, or even elimination of the issue. If it is still there, then reduce your flow rate a percent or two until it does go away. And remember this number, because you'll probably want to use that flow rate in general for your printer.
Looks like it treated those corners a bit differently in the slice routine. Normally something like this different treatment would be too slight to matter, but (probably due to the slight overhang?) it appears to have caused some layers to not adhere to the bottom ones and get pulled along to a different shape (red line in image). (Be sure to check some general layer adhesion/overhang topics too.) I'd try increasing the infill there (and using honeycomb style infill) or even using a cad software to hollow the interior to essentially a desired infill before slicing if the current infill is important elsewhere. Moving slower, Increasing the temperature a bit, and/or extruding more material per line and layer if the layer adhesion is due to the slight overhang (although it looks pretty small).
Is there a spiral lid mechanism I want to put a spiral lid on top of a container. When the lid opens, then the spiral mechanism will rotate into the container. Is there a name for this mechanism? If not, would something like this be possible?
It sounds like you're talking about an iris diaphragm. This has many parts that slide against each other, and would best be printed as separate parts, then assembled.
No spiral lids that I know of. I'm not sure how they would work, anyway. There are plenty of designs with hemispherical lids -- so-called "Venus" boxes. Check Yeggi: Yeggi : Venus box
Simulation tool software for 4D Printing In 4D printing technology or by means usage of Shape-memory alloy (non-metal, iron based, copper based or NiTi material) for 3D printing. Is there any software simulation tool which I can use to simulate this material change behavior with respect to time? For example, when introducing a change in humidity or temperature. Note: It would be best if the simulation tools targeted automotive parts (power train, cooling system, interior & exterior etc.).
I am going to say that this probably is a whole dimension out of scope for this group ;-) That said this new type of 3d printing is still at the University level. Also 4d is not necessarily 3d printing related at all. All it has to be is self assembling. Like http://www.selfassemblylab.net/4DPrinting.php Unless you have a connect with MIT. Then you aren't going to be simulating any 4d models. But if you HAD to do this, then you should write a paper about it and become a researcher. You could get published. You might need a PHD in material science. There might be some simulation in solid works.. but I would say you are mostly on your own and have to develop the models as they simply do not exist, especially outside of academia and stratasys. That said if you take the "4d" part and use models based on the current understanding of the raw material you would have more success.
(preface : I originally intent to write this as a comment, but the volume limit reached. In that tone, I'll utilize this answer space writing it..) to simulate those material change behavior in respect to time? COMSOL, ANSYS or equivalent ring a bell to be.. as long as you have the material properties (young's modulus, density, thermal transport & expansion properties, color?) right. Also. depending on your actual implementation resolution (nano-scale production.. macro/mini-lego sized assembly), you may want to adjust the mesh size for the finite element method solver. software for 4D Printing If your 4th D refers to time.. then the discussion ends here. If your 4th D is a space (not time) dimension.. Then you need to convert your 4D geometric object into a 3D shadow (just like generating the normal 2D shadow from a 3D geometric object) 1st to proceed. just like an 3D object (eg cube) can have multiple equivalent 2D form (or shadow). The same applies for 4D to 3D. Once you get the 3D coordinates of the chosen 3D form, you may print it as usual. Software wise.. mathematica (are used in the some of the example I've seen, others unknown/selfcoded) or equivalent will do.. but the key here is not the software.. it's the visualization algorithm. If you can get the "shadowing" algorithm done right, any 3D/math software will do. p/s : I didn't expect the materials modelling question coming when I read the title (I really thought this question refers to extra space dimension) .. but yeah.. why not. It's a good postgraduate/research topic to take on. ( :
Is a slower outer perimeter speed still meaningful? In the past we had printers with poor mechanics and with primitive software algorithms, therefore we used to print inner perimeters faster than the outermost one. See for example (generic, found online): However now we have pressure/linear advance which reduces extra oozing/extrusion in corners or areas with variable speed, and in Klipper we also have resonance compensation which takes care of imperfect mechanics allowing printers to be pushed to higher acceleration without visible artifacts (in my case from 2000 to 6000 mm/s^2), see (generic) photo: However printing slower has a clear disadvantage: E steps calibration is speed dependent with more filament being pushed out at lower speed, see Is there any reason left to print outer perimeters at a lower speed? Using one speed only except for specific areas (small perimeters, bridges, support) seems to make more sense to me to improve quality and reduce printing times.
Lowering speed on outer perimeter has always been mostly wrong, but possibly useful. Usually, it's a poor approximation for what you really want to do, which is lowering acceleration on the outer perimeter, to avoid surface quality and dimensional accuracy errors due to ringing and backlash. However, on bowden printers without compensation for pressure ("linear advance" in Marlin, aka "pressure advance" in some other firmware), slow acceleration and high speeds give really bad error in extrusion consistency, so you're better off just lowering the max speed too whenever you lower acceleration. It's also possible that you may want to print at extremely high speeds for inner walls and infill - speeds so high that you get a lot of extrusion consistency problems. If so, it would make sense to retain a sensible max speed on the outer wall. This won't avoid the part strength problems from printing too fast, but if your models are just decorative, it might be worth it. Generally, though, I'd recommend solving these problems right (using linear advance if your printer needs it, lowering outer perimeter acceleration to get rid of artifacts, printing at a speed your hotend can handle, etc.) and deem "slower outer perimeter" an idea whose time has passed. Also, note that if you're using a bowden printer without linear advance, sticking to the same speed for all extrusion will largely paper over the problem. This is probably the source of your observation that it "improves print quality" for you.
Thanks to Input Shaper, you can print faster, sometimes to the point of reaching your speed limit on the max nozzle flow speed. At that point, if all speeds are hitting that threshold without any changes in quality, you could print every part at the same maximum speed.
Printer thermistors read completely wrong after changing firmware I recently installed Marlin 1.1.9.1 onto my Creality CR-10S. After doing so, the temperatures for both the bed and hotend now read -14 °C. I have changed the setting in the Marlin configurations.h to several different thermistors, and it either reads 0 or -14 °C. I have now installed several different versions of Marlin and with each one I experience the same problem. Is there some tuning setting that I am missing or has installing the firmware somehow fried something on my board?
Okay, to anyone who is looking at this post, I am an idiot. I have spent the last 4 or more hours sifting through youtube videos and forum posts to figure out this problem. I have reinstalled firmware modifications dozens of times. All of this, to just now discover that I never had the bed or hotend cables connected. That's right, I never even considered the fact that I had yet to plug in those cables after transporting my printer to a new location. Don't be like me. Check the simplest solutions first.
If all cabling is correct, you have most likely chosen the wrong thermosensor type. You need to put the correct temperature table for your thermosensor into your Firmware. As explained here, you need to set the number at the end of these lines to match what you have installed. The "standard" CR-10 should use the Sensor type 1: 100k/4.7 k EPCOS. * :{ '0': "Not used", '1':"100k / 4.7k - EPCOS", '2':"200k / 4.7k - ATC Semitec 204GT-2", '3':"Mendel-parts / 4.7k", '4':"10k !! do not use for a hotend. Bad resolution at high temp. !!", '5':"100K / 4.7k - ATC Semitec 104GT-2 (Used in ParCan & J-Head)", '501':"100K Zonestar (Tronxy X3A)", '6':"100k / 4.7k EPCOS - Not as accurate as Table 1", '7':"100k / 4.7k Honeywell 135-104LAG-J01", '8':"100k / 4.7k 0603 SMD Vishay NTCS0603E3104FXT", '9':"100k / 4.7k GE Sensing AL03006-58.2K-97-G1", '10':"100k / 4.7k RS 198-961", '11':"100k / 4.7k beta 3950 1%", '12':"100k / 4.7k 0603 SMD Vishay NTCS0603E3104FXT (calibrated for Makibox hot bed)", '13':"100k Hisens 3950 1% up to 300°C for hotend 'Simple ONE ' & hotend 'All In ONE'", '20':"PT100 (Ultimainboard V2.x)", '51':"100k / 1k - EPCOS", '52':"200k / 1k - ATC Semitec 204GT-2", '55':"100k / 1k - ATC Semitec 104GT-2 (Used in ParCan & J-Head)", '60':"100k Maker's Tool Works Kapton Bed Thermistor beta=3950", '66':"Dyze Design 4.7M High Temperature thermistor", '70':"the 100K thermistor found in the bq Hephestos 2", '71':"100k / 4.7k Honeywell 135-104LAF-J01", '147':"Pt100 / 4.7k", '1047':"Pt1000 / 4.7k", '110':"Pt100 / 1k (non-standard)", '1010':"Pt1000 / 1k (non standard)", '-4':"Thermocouple + AD8495", '-3':"Thermocouple + MAX31855 (only for sensor 0)", '-2':"Thermocouple + MAX6675 (only for sensor 0)", '-1':"Thermocouple + AD595",'998':"Dummy 1", '999':"Dummy 2" } */ #define TEMP_SENSOR_0 1 [...] #define TEMP_SENSOR_BED 1
Cura 2.4 missing "split object into parts" I have an stl with multiple parts that I want to split up. Cura 15 had an option to "split object into parts" but I can't find that in cura 2.4. Did it get removed?
I don't think this feature was implemented at all with Cura v2.x. As the developers say on the v2.1 release, "Cura has been completely reengineered". Finding proper changelog documentation appears to be pretty hard because they have not posted any actual changelogs except the "user friendly viewable" changelogs which only list additions of new features but don't display what everything they changed between each version of their application. Here is the most complete changelog I could find. I do not see any mention of this feature. https://ultimaker.com/en/products/cura-software/release-notes Going through the Cura 2 manual or the Cura 2.1 FAQ, also does not mention this feature. https://ultimaker.com/en/resources/20406-installation-cura-2-1 Furthermore, searching around for version 2 "split objects" lead to forum posts of people suggesting to use some other software to achieve this specific task. If you decide to go this route, I recommend Meshmixer from Autodesk to manipulate your models and then export to STL and import them to Cura either as a whole new position set up or separate model files where you can change them there as you need to (meshmixer allows for object repositioning around a defined build plate so you can just import the whole assembly into cura and then print). It might also be worth to put in a feature request on the UM forums.
You can split STL files in Cura 2.4 Just open your STL file in Cura and right click it. Select "Split object into parts"
I can not really connect successfully to my printer via USB I have a Tronxy P802M (very similar to the Anet A8, but using a Melzi2.0V5 board) that seems to work fine (I just finished building, and axes movement and the integrated display work) but when I try to connect to the printer from my Simplify3D on Windows 10, I get the following: [...] Connected to machine! SENT: T0 READ: ok 0 READ: wait SENT: M105 READ: ok 0 READ: T:24.44 /0 B:23.33 /0 B@:0 @:0 Connection failed. My other printers all connect fine.
The Simplify3D support site mentions to disable the "wait for startup command" option in the firmware configuration for S3D. This allows me to make a rudimentary connection to control the printer, however e.g. during the bed levelling wizard of S3D, the connection still breaks off. It works reliably with Octoprint. I'll consider that good enough.
Try changing your port in the control menu. By default, mine is set to COM 1 and I have to change it to COM 3.
What causes this "stringing" and can my print recover from this? Here's what it looks like This is the model thingiverse linky It looks like it couldn't print the edge, but this happened many many hours after printing the brim. This did not happen with my 1st attempt at this print. The last print lost adhesion and I had to scrap it. This time, adhesion looks good so not sure why this happened. Printing with Monoprice Select V2 with ABS, sliced with Cura. 100C bed / 250C extruder. 15mm/s initial layer speed. 60 mm/s print speed. Update It looks like the printer is starting to smooth it out like so. Still not sure if this will lead to an ok print or will fail because of this layer. And it seems the stringing area does not have a brim underneath it. Did Cura just not calculate the brim size correctly? Update2 Here's a few screenshots from Cura to show that the model is lying completely flat. I let the print go on overnight and here's where I stopped it It almost seems like the print shifted completely after printing the initial layer. Have you ever seen anything like this or is there anything in my Cura model that would make it do this?
No, your problem is not related to slicing, this is a hardware problem. Your complete print has shifted, this is called layer shift. This could happen when the nozzle hits an obstruction while printing while the Y stepper continues. This could lead to skipping teeth on the belts, slipping of the pulley or missing steps. This results in printing over air as the print progresses. This manifests itself as stringing, but in fact is unsupported printing (in the air). In this case it is unrecoverable as the printer has lost the reference frame, it just continues to print with the new reference frame caused by the layer shift. A Prusa MK3, or any printerboard using trinamic stepper drivers would be able to recover (if the belt and pulley are correctly attached, and steps are missed) as the skipping of steps is detected, in case of a Prusa MK3 the machine will re-home when it detects skipped steps and continue printing. See also this answer for more details. Possible solutions are increasing the belt tension, increase the stepper torque by increasing the current through the stepper drivers or re-tighten the pulley on the stepper of the Y belt.
There are many problems caused with this. It could be a faulty motors, an unlevel bed, a dirty nozzle, or even something as simple as using the wrong filament settings. It would be best to go through a checklist of things that might be wrong with your printer. It could also just be caused by the wear and tear of a really old printer.
Can't connect Cura to my Anet A8 on OSX 10.11.6 I finished the mount of my Anet A8, tested everything and apparently it was ok. I installed the driver that came with it, CH341SER_MAC, turned on the printer, connect USB cable, but nothing happened. In Cura, I tried to add a printer many times and this message always appears The printer isn't connected. In OSX, I discover in System Information, an USB2.0-Serial, that I think can be the printer. Is it a printer driver problem? Is it the Cura setting? I'm completely lost...
There are a lot of problems with the CH340 chipset drivers to be found on 3D SE and various forums on the internet. To use this cheap CH340 chip that is used by a number of Arduino compatible (clone) boards to provide USB connectivity (a USB bus converter chip that converts USB bus signals to serial interface) you need to install a correct working driver. For Mac OS X you can try to download a working version for the OS system you are using. You can try to use this driver installer or this driver installer. Both reported to work with Mac OS. Edit: If you are using OSX El Capitan, please read this. To get the CH340 drivers to work you need to use the tool csrutil. Reboot and press CMD+R immediately after hearing the startup sound to boot to Recovery Mode Open Terminal Execute the following command: csrutil enable --without kext Reboot More information is found here!
What Cura version do you have? Cura only works with the Anet A8 via USB on versions 2.3 and lower according to this page. The page is from thingiverse(A 3d printing forum + print files page)
Why is Print Adhesion almost good, but still not right? It feels like my print is almost there, I spent a long time fine tuning the calibration with a fantastic set of G-code files from Chuck Hellebuyck. I am using an Ender 3, on which I have modified the heatbed to carry a glass sheet, otherwise factory standard. The prints are done on the bare glass. When I printed a window some 14 feet (~4.5 meters) away was open to the cold New York winter. The cooling fan on the Ender 3 is mounted on the right side of the hotend and uses a simple declector, it can be seen in picture 2. However, my print is not perfect yet. I'm using 1.75  mm PLA at a temperature of 200 °C for the 0.4 mm nozzle and a bed temperature of 60 °C. Out of the 4 corners, 3 printed perfectly, the back-left one did lift. What do I need to do to fix the bed adhesion issue?
Based on the pictures, this looks like the bed wasn't clean enough, thus why it didn't adhere correctly in some spots. The lifted corner is the worst, but I can see a couple more spots where the PLA didn't stick well enough, though the brim prevented it to completely lift off the bed. I would advise you to not print with the window open, even if it's only slightly ajar. Weather conditions and temperature will affect the quality of the print, even if it's only a small variation in temperature. I've had enough failed prints due to a shift in temperature to know that, even though PLA isn't affected as much than ABS by temperature changes through a print. You might also try to increase the bed temperature a bit, up to 70°C or 80°C. I've looked at the recommended temperature for PLA with my UM3E and it says 80°C for the glass bed. You should always make sure that the glass bed is clean: ie no dust on it and no grease. Yes, touching it with your fingers will leave a thin greasy residue in the form of fingerprints, and even one fingerprint can prevent good adhesion and result in the problem you had. There are various products to clean glass beds, I've personally found out that the cheap yellow-colored window cleaner from Karcher works wonder. Do not use the standard blue colored window cleaner products: the blue one has a chemical in it that is made with the express intent to prevent anything to adhere to the glass, which is obviously not what we want. The yellow-colored cleaner doesn't have that chemical, and it cleans without depositing an anti-adhesion film on the glass. If a thorough cleaning isn't sufficient, you can look into various adhesive products. While there is quite a lot of adhesive solutions marketed as being specifically for 3D printing, with insane prices most of the time, I've found that a simple UHU glue stick works wonders. I usually do not need glue when printing PLA, but I use it for Nylon, ABS and other filament that absolutely require it and it's a breeze to work with, and it cleans easily with soap and warm water.
I work with a glass surface on my machine and use a drop of superglue to attach the corners after the first layer is complete. The biggest contributors to not sticking are : Bed not level. Temperature of filament not hot enough. First layer print speed is too fast. Heated Bed not hot enough to compensate for cold room. Try enclosing print area. Head is snagging on print edges (no retraction when moving) and ripping print loose. Print bed surface is not clean. Use acetone or ammonia for cleaning glass. Alcohol leaves an oily residue. If you are printing on painters tape, make sure the bed is very clean when you put the tape down. Then make sure you rub the tape down well so it doesn't pull off of the print bed surface during printing. Then clean the tape surface to get the oil from your hands off of the tape. I use Glass Build Plate Wizard spray for delicate prints and a heated bed. It releases after the print plate cools down. That eliminates the need to chip the print loose. It releases itself after a few minutes.
Whats the difference between a basic rapid prototyping machine and a 3D printer? In general 3D printers are compact and smaller than RP machines. That's ok. But, what's the difference? 3D printers can be used as RP machine too.
All rapid prototyping means is automatically producing a physical part from a cad model. 3D printing is a way to achieve rapid prototyping. There are 2 main methods of rapid prototyping: additive, and subtractive. A 3D printer is additive- you add materials to an object layer by layer. Usually, when people talk about a subtractive machine, they are talking about a CNC mill (or lathe), which tend to be extremely large (most are over one ton). You start with all the material there, and you subtract the material that you don't want. This might be what you are thinking of.
A sintered metal printer is a version of a 3D printer that is rapid, but expensive. Seen 1 for 800,000$aud Uses laser to melt metal particles like titanium.
Speeding up the heating of the heated bed Currently I am using a 12 volts, 20 amperes power supply (Model S-240-12) The stepper motors and the extruder need 5 amp, and the heated bed build plate needs 11 amp. Technically you only need to use a 12 Volts, 16 amperes power supply, but I understand that you use the one with 20 amp because pulse currents from extruders and stepper motors can be stressful to supplies loaded to the max, so for reliability and performance, it is better to use a supply rated for 25% more than you need In the place where I buy the spare parts for my 3d printer they also sell 12 V power supplies capable of delivering 25 amp and 30 amp and they told me that if I use those ones you are going to be able to speed up the heating of the heated bed. Is that true? I understand that the heated bed is only going to take the 11 amp that it needs so is not going to make any difference to use power supplies capable of delivering more current
Changing the PSU with one with a higher amperage will not make the bed heat up any faster unless the PSU is underrated for the amperage required and the voltage is dropping as a result of the load. This can be checked by measuring the output voltage with a multimeter (when the PSU is loaded e.g. by a heating heat bed). In this case, the PSU has a marginal higher Amperage than the printer consumes (even has some room for the over-voltage; under the assumption that it is a good working PSU). Increasing the voltage will decrease the heat up time. There is a screw next to the 12 V connectors that can change the output voltage of the PSU. Usually, it is safe to increase the voltage up to 14 V, but that depends on your whole setup (and 14 V is applied to the whole setup, increasing the current for all parts, including your printer controller board, this board must be rated for the 14 V). Please do check the stability of the voltage during load. Although it can be done, it is not something I used. What is an extra minute on a print of several hours? You can do the math: say the heat bed has a resistance of 1.2 Ω. We only need two formulas: $U=R\times I$ - potential Difference U is Resistance R times Current I $P=U\times I=I^2\times R=\frac {U^2} R$. The power P of an item the potential difference times the current through the item. at 12 V that will draw 10 Amps (12 V / 1.2 Ω) resulting in a 120 Watt bed: $P= 12^2 \text V \times 10^2 \text A= {10^2 \text A}\times {1.2\ \Omega}=\frac{12^2 \text V} {1.2\ \Omega} $), at 14 V that same bed will draw 11.7 Amps (14 V / 1.2 Ω) resulting in a 163.3 Watt bed. Use at your own risk! What you could do to decrease time to heat the bed without changing the PSU or the voltage is to insulate the bottom of the heat bed with heat bed cotton sheets or cork (placemats from IKEA ;) ), put a sheet of cork onto the heat bed before printing and start heating the bed through the LCD panel of the printer or any attached printer controller programs over USB prior to printing.
I hate to sound like the Toolman Tayler from that old TV show. "But, it comes down to More Power!" Power is the ability to do work (move a mass certain distance) within certain amount of time. Power = mass x distance x Time. It can also be expressed in electrical terms, as the ability to heat something within certain amount of time. Power = Voltage x Current Since most systems have a fixed voltage, it is still possible to increase power by increasing the current, since, Voltage = Resistance x Current, And Power = Voltage x Current, So, Power = (Resistance x Current) x Current. So by switching a power supply with the same voltage, but higher current, it will provide additional power to the system. The larger current available would be able to flow through the heating elements, heating them up faster. However the caveat will be the amount of heat dissipation in the system, the large surface of the bed, will carry away enough of the heat due to air convection, that it may not make much of a difference. Or perhaps the heating element may not handle the larger amount of current and burn out. It would be worth testing it out, in my humble opinion. Hopefully without causing a fire somewhere. :-)
Which software do I need to start print something? I'm still new to 3D printing and I want to print something. I expect that I'll mess it up since I find nothing to adjust it but it is now laying around for 4 months and I'm sick of it. So my question is where do I find Windows software to print something and of course where do I get a 3D model? I own a Geeetech i3 Pro W.
First; find a model! To print something you require a model (usually this is in STL format, look into websites called Thingiverse and MyMiniFactory for examples). Once you have a model file, you need to make it readable for the printer firmware. If you can't find suitable model, then you need to design a model yourself (or ask someone to do it for you) or adjust an existing model to suit your needs. "Good (preferably free) Beginner Software for Part Creation?" is a good place to start. Second; use slicer software For a printer to be able to print the model, the model needs to be sliced into layers. These layers need to be printed at specific speeds, temperatures, etc. Search online and look at the filament packaging (usually the ideal temperatures are on the packaging) to find the ideal temperature for your filament. If you are not using the right temperatures, your print will most likely fail. Programs that are able to slice models are called slicers. The most popular free (and Windows compatible) slicers are Ultimaker Cura and Slic3r (or its Prusa distribution). The slicer produces a printer readable file called a G-code file (file filled with printer instructions for e.g. movement and heating). This G-code file can be sent to the printer using specific printer software (e.g. OctoPrint, Repetier-Host, etc.) but more common or simple is to put the G-code file on an SD card and print the file using the print menu on the printer LCD.
If you're just starting out then Tinkercad (website) is a good place to start designing your own objects. Later you can get to grips with OpenScad for more complex shapes. Both are free.
XYZ calibration without PINDA probe I have been happily printing with my Original Prusa i3 MK2S for a few weeks now. I have been thinking about replacing the PINDA probe with a BLTouch tactile sensor. If I do so, how will I be able to do the XYZ calibration? I believe that the tactile sensor would only be able to do Z calibration. If I need to redo the XYZ calibration in the future for some reason, is it possible to do manually? Or would I need to remount an extruder with a PINDA probe temporarily?
If you want to replace the PINDA probe (whatever reason for) then you can go with the BLTouch Sensor but only for certain operations. It will be good for leveling the bed before printing because here, only the bed level is important and not the skew of the bed. If it comes to calibration of the skew itself, the BLTouch Sensor will not be usable because the PINDA Probe detects the boundaries of the copper circles on the print bed. The BLTouch cannot detect these copper areas. The PINDA Probe is a proximity sensor. If you want to do a recalibration you have to mount the PINDA again. Therefore, it is possible but not recommended.
If you want to replace the PINDA probe (whatever reason for) then you can go with the BLTouch Sensor but only for certain operations. It will be good for leveling the bed before printing because here, only the bed level is important and not the skew of the bed. If it comes to calibration of the skew itself, the BLTouch Sensor will not be usable because the PINDA Probe detects the boundaries of the copper circles on the print bed. The BLTouch cannot detect these copper areas. The PINDA Probe is a proximity sensor. If you want to do a recalibration you have to mount the PINDA again. Therefore, it is possible but not recommended.
E3D V6 nozzle vs MK8 nozzle, first layer adhesion Does anyone notice that when they upgrade from an MK8Makerbot(?) to a E3D V6 hotend that when using the same settings and bed-leveling/z-distance-setting procedures, that the filament is much more likely to be pulled up and bunch up around the nozzle while printing the first layer? It occurs mostly when printing small details, such as 3 mm bolt holes, and not so much when laying down long lines. Is this due to the shape of the nozzle, which on the V6 is much more flat when compared to the MK8 which is more sharp, or is it due to something else that can be easily fixed?
When filament curls up this generally means that there is some sort of obstruction or burned material in the nozzle causing uneven flow. You can use the Atomic Method from this answer. Being a new nozzle, I would expect that this is not the case, so that it should just drop out of the nozzle and not curl up. It could be that the nozzle is not perfectly machined and a defect in the geometry is causing this. You could try to replace the nozzle with another nozzle. They are pretty cheap, you should buy a few extra. As far as why the filament sticks easier to the E3D nozzle when it curls up, is that this nozzle has a (hexagonal) flange that sticks out, while the other nozzle does not have such a flange. If it curls up, it has a larger probability to hit the flange and stick to the outside of the nozzle than a nozzle that has no such flange.
Turns out this can be fixed by increasing the bed temp to 65-70 °C and of course increasing the extruder temp by 30 °C or so, which is standard for the all metal hotends (no idea why)
Can model used for 3D printing be later used for mold mass production? I am working on a project that requires an enclosure that I am thinking to prototype using a 3D printer (which is pretty easy and overall awesome). My question is simply whether this same model can be used later on for mass production (i.e. mold injection)? What adjustment (if any) need to be done to model between prototype and mass.prod. stages? What areas in design to look out for that can make model viable for 3D printing non-viable for mass production?
I respectfully disagree with the hard no answer. There are many casting methods, some of which are not compatible with 3D printed parts and at least one that definitely is. See investment casting, aka lost wax casting. (Ref 1) Also search YouTube for "investment casting using 3D printing". Formlabs, the company that makes the Form 2 and Form 3 3D printers, sells a 3d printable resin specifically for investment casting. (Ref 2). There is a whole world of casting, so I will describe, as an example, a very simple process that will make a replica of your 3D printed part. I will briefly address the steps necessary for making multiple copies at the end of my answer. First you need to add a cylindrical extension to the shape that will create a sprue. See the two models shown in the image below. Next make a foil cup a little larger than your 3D printed part and place your part in the cup, suspended by the sprue. Pour liquid plaster of Paris (POP) into the cup, covering the 3D printed part, with just the top of the sprue sticking out. Once the POP has hardened, you can remove the 3D printed part by dissolving it in an organic solvent (acetone for acrylic) or by heating the part to several hundred degrees C so it will burn out (convert to gasses). You will now have a block of POP with a void shaped like your 3D printed part plus a cylindrical hole to the outside that acts as a sprue. Once you have heated the POP to drive out any remaining water, you're ready to cast. Fill the void (via the sprue hole) with, for example, copper powder and heat it to well above its melting point. Once everything has cooled, lightly tap the POP with a small hammer or equivalent, to remove it, leaving your final part plus the sprue. You can remove any remaining POP with some warm baking soda and patience. The final step is to cut or saw off the sprue shape. Please understand that the process described above is just to give you a basic idea of the process. There are many alternate or additional steps that may produce a better final product. Also, the process described should work of the rook model shown on the left of the image below, but not the model on the right, which has a small hole through the middle. There are ways to solve that issue, but as I said before, there is a whole world of this stuff. Casting is a way to create parts out of materials that you otherwise could not 3D print directly (on a home budget). For mass production, you would need to start with an inverse of the final shape. For example, a cylindrical slug with a rook-shaped void. Then you would use the investment casting process to create a mold out of a rigid material that you could use to, for example, create rubber copies of the original part that could, in turn, be used as the "wax" for copies in the final material. I urge you to experiment, maybe following one of the tutorials on YouTube. Good luck. Ref 1 - Investment Casting https://en.wikipedia.org/wiki/Investment_casting Ref 2 - FormLabs 3D printable casting resin https://support.formlabs.com/s/article/Using-Castable-Resin?language=en_US
That's a hard No. Models used in industrial injection molding are very specific to the process sometimes even the machine. When they build those molds they have to include constructs like spurs and runners; things to get the molten plastic to where it needs to go. Further the wall thickness of an injection molded part can only be so thick, because the plastic will contract when cooled, which will distort the final part if it is too thick.
What could be causing my y axis to slip? Occasionally, while printing, my y axis will slip and the layer will, from that point forward, be shifted, ruining the print. What might be the causes of an axis slipping? I have tried cooling the motor which seemed to have been getting warm, and the belts are not too tight. This does not happen with every print, and seems to be an intermittent problem. My printer is a MendelMax RepRap, and the y axis is my moving bed.
(source: all3dp.com) Your printer is skipping steps in the y-direction. This can have several causes. Take a look into Shifted layer guide on RapRap.org which lists 29 possible problems that can cause this issue and how to fix them. First items of the list: Driver current is too low Driver current is too high Belt too Loose Belt too Tight Loose Set Screw/Grub Screw Belt or Bearing is binding Speeds are too high Acceleration is too high ... When I was dealing with this issue on my RepRap I had to increase current to the particular driver.
My Y axis runs on a channel and I believe there was some grit or metal flakes in the channel left over from manufacturing. The wheels in the channel got stuck on the debris and caused the belt to slip. It made a horrible grinding noise when this happened. So I blew out the channel with pressurized air and tested all the wheels. I'll update if necessary as I test my fix with longer (taller) prints. Update Actually, the print had messed up g-code. The gcode file was corrupted.
Delta Printer: Slighty incorrect print alignment on the build plate After building a Delta printer, I noticed that my whole prints are slightly tilted around the Z-axis in comparison to the slicer (e.g. Cura). There is no twist layer wise. This means, the prints themself look actually perfect. I just don't know what could be the reason of the rotation. I do not believe it is a build issue of the printer, because I tried to keep the printer frame pretty stiff and symmetric. Could it be, that the Auto-Calibrate Feature of Marlin can add such a rotation? The picture below illustrates the problem. I expect the black alignment of the print and get the orange one. Note that the print is still a rectangle with ~90° corners.
I figured out that the reason is probably a slightly translated slider construction. Instead of using a proper centered slider as shown in red, I used a slider construction like illustrated in yellow. When all sliders are translated on each tower like this, the print should be tilted by the same amount. This seems to have no influence on the general shape of the object. However, for my next printer I will use a proper centered uni-body slider.
If I am reading this correctly, your prints are being either stretched or your prints are shifting / leaning on more complicated prints. In this case, given that you are on a Delta printer, my answer is the same for all. I usually do Cartesian based 3d printing but the concept is the same for any drifting or leaning. You simply need to recalibrate your steps per MM for each motor, and tighten your belts. You will have the complication of the interaction of the 3 arms, that others will be able to answer better. But in the end, if each arm moves as it should, the belts are not slipping, and you do not have issues with moving too fast (jerking can cause the belt to shift, and a loose belt can cause whiplash / and other print artifacts). My bet is your steps per MM is off on one of the motors, or you could have an overheating issue (not likely). There are many guides to help with Delta specific calibration. I can provide a better answer with photos. See Stackoverflows guide on asking questions. Edit with the diagram (not a photo), you issue might be caused by stepper over voltage and you will need to adjust your pololus. If you hear a repeating Thud noise, you have your voltage too high.
Can 3D printers print details in the 1/10 of the micrometer for metals? I'm starting to get familiar with 3D printers. I wish to know if printing details the size of 10-7 m (3.9*10-6 in) is possible these days with metals or any other material. If anyone has information or articles as leads, I would really appreciate it.
You can get the 0.1 micron (100 nm) resolution with a 2-photon 3D printer, but only in a polymer resin. Nanoscribe, in Germany, pioneered this technology, see Mechanical Microstructures. Their commercial printer, the Photonic Professional GT, is about $350,000 US with software and accessories. There is some work being done to replicate the 3D printed polymer in metal using electroless plating or ALD (atomic layer deposition). Other techniques are in development. None of the direct metal 3D printing processes come close to your 0.1 micron resolution, although the field is rapidly evolving. Only a couple of years ago, direct metal 3D printing was all based on powder bed fusion. Now binder jet technology as been adapted to metals and, very recently, Xjet has developed a nanoparticle 3D printer. It prints "ink" composed of metal or ceramic nanoparticles suspended in a liquid. The minimum layer thickness is 1 to 2 microns. They have not released XY resolution data yet. A good overall reference for the various 3D printing techniques (including Xjet, but not Nanoscribe) can be found at Explaining The Future - 3D Printing.
Today, only SLA/DLP will give features of that size (if even these technologies do), and I am unaware of anyone using these technologies to print with metals today.
ABS Filament safety concerns I have the XYZPrinting da-Vinci-1.0 with ABS filament. I am concerned about ventilation. If this is used inside, what safety precautions are necessary, which are recommended, and/or which are optional?
Yes... The issue with all 3d printing materials. Not just ABS, but worse with ABS is the fine air particulate and Ultra fine it creates during the 3d printing process. PLA is considered safer than ABS. But I fear people will use this as justification, it is like saying I only smoke one cig a day instead of two so I am safe and healthy. No it really should be taken seriously. There are a number of scientific papers and articles proving that this is an issue. Specifically that 3d printers release ultra fine particles into the air. Which can damage the lungs over time. I would STRONGLY advise not using a 3d printer around children, or at least putting it in the garage where you will not contaminate your homes air supply. I.E. http://www.sciencedirect.com/science/article/pii/S1352231013005086 http://built-envi.com/portfolio/ultrafine-particle-emissions-from-3d-printers/ https://www.ncbi.nlm.nih.gov/pubmed/11139166 http://www.3ders.org/articles/20160201-new-study-shows-health-hazards-of-3d-printing-suggests-pla-could-be-your-safest-bet.html Lot of these have some scary looking graphs. Note that I am being a bit sarcastic when I say "scary looking graphs" the take away is ABS has twice the Fine Partical emission as PLA. However should you be worried, is still up to debate. The idea is that FPE can maybe contribute to cancer or other illnesses. FPEs are thought to irritate the lungs. These graphs are of the ultra fine particle emissions. as you can see PLA is MUCH safer in this regard. I cannot find the paper at the moment, but the recommendation is a full air cycle several times an hour. As someone that lives in a Cold state I personally just use PLA and am rolling the dice. There is another SO where I cover my future plans for an air scrubber. I will note that other materials such as Polycarbonate should just be plain avoided. I also want to provide these links on WHY ultra fine particles are bad. In summary they really upset your lungs and are thought to cause lung cancer over time. Wikipedia on Ultra fine particulates. White paper on ultra fine particulates.
your fine at practical temperatures. source: https://en.wikipedia.org/wiki/Acrylonitrile_butadiene_styrene#Hazard_for_humans recommended would probably be set your controller to not go above 380c if your really worried. but it's not needed.
When to use 1.75 mm vs 3 mm filament? Why do we have two standard filament sizes, 1.75 mm and 3 mm? Does it really make a difference when printing? Or is the 1.75 mm just for smaller printers? In what situations should I be using 1.75 mm? When should I be using 3 mm?
There's no appreciable difference. Just use the filament that fits your particular printer. If you don't yet have a printer, then I'd get one that uses 1.75 mm filament: 1.75 mm is increasingly becoming the "standard", thus being easier to get. Some filaments are not available as 3 mm. 1.75 mm filament allows for finer control, because feeding in 1 mm of filament corresponds to less plastic extruded. 1.75 mm filament requires less force to extrude. Compressing 1.75 mm down to 0.3 mm takes less force than doing the same to 3 mm filament. However, the advantages are fairly minor. I don't see any reason to replace a functioning 3 mm extruder with a 1.75 mm one (yet).
I agree with all the answers above but only would like to add if you want to build a very large printer using 0.8 mm or 1.2 mm nozzle, i would advice to go 3 mm. 1.75 mm filament then just can't be fed in fast enough.
Will this MOSFET allow the heat bed to run at a different voltage than the control board I am currently running my Tronxy X5s with a MKS Gen L board. So far I have not ran the heat bed over 50 degrees C since I have only printed with PLA so far. I plan to try PETG and/or ABS in the near future and I have a spare power supply 12V/360W laying around. When I first got my printer I purchased this external MOSFET board after reading about X5s "upgrades", but so far have not used it. My plan is to now run the heat bed using a separate power supply than the one running my control board using the external MOSFET to switch it. Since this power supply will only be powering the bed, I would like to bump up the voltage, via trim-pot somewhere from 12V-15V, to gain some watts per square inch on my heat bed. Will this MOSFET isolate the heat bed circuit from my control board to allow it be ran at a higher voltage? Is it safe to run the power supply/heat bed at a higher voltage than it is rated for a significant amount of time?
Short answer YES. You can run it from a different power supply at a higher voltage. Also it has a PC817 Optical isolator (for some reason) therefore the second power supply and your main board should not be electrically connected at all.
For reference,I done tracing this module as shown. So you can use isolated power supply for load. Also you can use up to 24V without any problem. Edit: This module still work with higher voltage up to 24V. But according to question. Using higher voltage supply more than rated is same as my question here. For summary. Adding more voltage to resistive load results to higher current flow. Recheck your wire and connectors for current rate so you will know the limit of voltage you can go with. One more thing, mosfet can fail like short circuit. At this situation the temperature of bed cannot be controlled. At normal voltage rate even we left the bed connected to power supply the temperature will rising to about 100C if you add more voltage it will go higher than that can the heated bed will broken or burned or start to fires
Changing the home position on a Prusa i3 MK2 As part of a project with my university, I have developed a new extruder to attach to a Prusa i3 MK2. My problem is that both the nozzle and PINDA probe have moved 17mm forward and 0.5mm to the right. As a result when I try and calibrate the printer it moves to the home position and the PINDA probe is too far out over the heatbed so it doesn't detect the printing surface. What is the simplest method of moving the home position so that the printer can be properly calibrated? UPDATE: I am planning on removing the heatbed and placing spacers that will move the printing surface 17mm forward. This should then prevent the printer losing any printing area and hopefully prevents me having to edit any code. Can anyone see any problems with this I'm overlooking? The simplest thing to do would be to move extruder 17mm closer to be the same as the original printer but my deadline is fast approaching and I haven't time for a redesign that large.
Consider the original installation with the orientation of the Pinda probe to the nozzle. Let's say for argument's sake that the Pinda probe is 3 mm to the right and directly in line with the nozzle on the y axis. If you examine your new nozzle, I would expect that the relationship of the nozzle to the Pinda probe no longer matches the original spacing. If possible, re-design the mount to place the Pinda probe in such a way as to match the original design. Thanks for pointing out my oversight, Mac. If the relative position of the nozzle and pinda probe are as the original, the solution is then in changing the appropriate parameters in the firmware. I found a reference for someone who had a bit smaller error in home position, but the concept is the same. The link above points to information reading thus: In Configuration_Prusa.h: Code: Select all // Home position define MANUAL_X_HOME_POS 0 define MANUAL_Y_HOME_POS -2.2 define MANUAL_Z_HOME_POS 0.15 // Travel limits after homing define X_MAX_POS 250 define X_MIN_POS 0 define Y_MAX_POS 210 define Y_MIN_POS -2.2 define Z_MAX_POS 210 define Z_MIN_POS 0.15 it will be necessary to connect the printer via USB to a computer running an Arduino IDE and to load the Prusa specific files for that printer. Edit the noted location, save/write the configuration and test. I would suggest small adjustments in only one or two parameters at a time, to avoid ambiguity in the cause/result sequence.
What is the simplest method of moving the home position... I think the solution outline by @fred_dot_u is very elegant, so I would go with it. ...so that the printer can be properly calibrated? I'm not sure that will be possible. Because the physical lenght of the axis hasn't changed, by moving the nozzle/probe, you have actually reduced their reach in the opposite direction, so the probe may be unable to travel on top of the intended calibration points (the usable printing area has also shrunk, but that's less of a problem). If that is the case, I can't think of an easy solution (bar not using the auto-calibration feature altogether).
Best material for compression? I'm designing a mount for a cylindrical speaker to attach to my bicycle. It will mount on the bottle cages. I've printed a few iterations with various infill settings (using PLA) and the weak point is always the bolts holding the entire mount to the bike. They can't handle the compression needed to secure it properly. I had thought about using an imbedded metal part to distribute the load, but it's not easily replicable and I want to make the design public and fairly accessible to others with the same speaker. I currently have PETG and ABS, would one of those perform better or should I order a specialty high strength polymer?
Infill has minimal effect on the strength of printed parts, so I would expect the part to break in the same spot regardless of what infill percentage you used. PLA is especially poor in this exact application, and it undergoes significant creep/cold flow under mechanical compression over time, so even if achieved the necessary strength by changing settings (which you can), it would require that you periodically tighten the bolts more and more, as the PLA would slowly deform under the mounting pressure. Perimeter width and number of perimeters are what primarily influence the strength of a printed object, infill has very little impact on strength in comparison, and unless you're using an exotic pattern like gyroid, what impact it does have is not even close to isotropic (will add strength in some directions while doing nothing in others). But even then, infill only really has an effect when we are talking about forces that are spread evenly over the entire object, not concentrated strength of a specific spot of the part. And that effect is always much weaker than what perimeter count or width will have. Just bump up your perimeters to 4 or even more and that should make a huge difference. And also, don't use PLA. I think PETG is a much better choice in this situation. It is more ductile only slightly less rigid than PLA, making it much more durable overall than PLA, and less prone to cracking under compressive forces. PLA theoretically has higher tensile strength, but that often doesn't mean much. I would not recommend ABS, it tends to have similar issues with brittleness and is one of the weaker materials one can 3D print. It terms of ordering a special high strength polymer.... unless printer and hotend is rated for in excess of 400°C, no such 'high strength polymer' exists, at least not that you can print. PLA and PETG are close to the best you can get, with Polycarbonate inching out ahead but not by a huge amount (~20%). Despite what filament companies would like you to believe, carbon fiber reduces the strength of PLA, ABS, PETG, PC, and probably nylon, and instead simply makes those polymers more rigid and increases dimensional stability. The only filaments that would actually be made stronger with added fibers short enough for filament manufacturing processes are ones with glass fiber. But you don't want to print those filaments, trust me. They will dull the teeth on your hobbed gear(s), even if made from steel, and will just ruin all but ruby nozzles very very quickly. And they still wear out ruby nozzles even then. There are exotic polymers, but none of them print at less than ~350°C, and are generally exceedingly expensive. All the polymers that can be used at normal printing temperatures are all fairly similar to each other in terms of tensile strength at least.
Try heating op your PLA while mounting it. The PLA wil temporarily weaken and will become a bit moldable. PETG is a bit more brittle than PLA is my experience so I'd say stick to PLA. Otherwise you might try Arnitel ECO. I can try to find an amazon purchase link for you if you want. Anyway, it is more rubber like and stays flexible, it will never crack but depending on your design it might become a bit more wobbly. If you share a picture of your design or an STL i might be able to help you a bit better.
Out-gassing of printed material when heated We know from this answer, 3d printed materials continue to outgas after printing and being cured. My question relates to this: How much does heating the printed object after printing (or being cured) affect out-gassing? Does the continued out-gassing degrade the stability/quality of the print?
This is not an answer to your question, but it relates to outgassing so I am sharing it here. I have used the acetone vapor method of smoothing the surface of ABS prints. It works really well, and the surface becomes much smoother and glassy. I printed a large coffee mug (lets, for the moment, ignore food safety issues) and made it very smooth. After a couple of weeks to allow the acetone to fully evaporate, I poured a nice, hot cup of tea. Unfortunately, the acetone had not fully left the print, and the surface was immediately covered with dozens of bubbles as the acetone evaporated and pushed against the ABS. This effect was only because of the acetone. Another cup that had not been vapor smoothed worked perfectly and was unaffected by the hot water. I have not seen similar outgassing from PLA, ABS, Nylon, or PETG. If I may hazard an opinion about the substance of your question... I don't think that outgassing is the biggest contributor to the aging of prints and their properties changing. I suspect that (perhaps not in order) these are larger factors: exposure to UV light. UV light breaks polymer bonds and reduces the strength of plastic. absorption of water vapor, which can both expand the material which causes stress, and chemically break polymer bonds. long-term crystallization of the material fatigue from repeated sub-failure stress NASA used to have resources that spoke to outgassing rates related to suitability for space applications.
The effect of temperature on outgas rates is a subject for Engineering.SE . As to continued outgassing - depends on the material in question. To pick an extreme example, solid CO2 (dry ice) will outgas until it's gone. But stuff that's got VOCs is expected to outgas; said volatiles leave by design and the remaining material stands on its own.
What is the melting temperature of a 3D printed part? Here is the context I've got an old car for which I have a small plastic piece who is broken. As it's an old car and a very specific piece, I can't find it anymore. So I was thinking about 3D printing it. My problem is this piece is on the carburetor, so close to the engine. This means, it can heat a lot, close to 90-100 °C. My question Do the pieces created with the common 3D printing techniques melt at 100 °C? If yes, what kind of other 3D printing technique can I use? Here is the piece I want to recreate (sorry for the bad quality), the scale is in cm.
The number you're looking for is the glass transition temperature (the lowest temperature at which the material can flow or warp), not the melting point. This depends on what material you're using; approximate temperatures for common printable materials are: PLA: 60˚C PETG, high-temperature PLA: 95 ˚C ABS: 105˚C Nylon: typically 70˚C or above ("Nylon" is a large family of similar polymers) Polycarbonate: 145˚C Any plastic under your hood is probably either nylon (for its durability, impact resistance, and chemical resistance) or ABS (for its strength and heat resistance). These are both difficult materials to print: ABS emits toxic fumes while printing, and tends to warp if you're not using a heated enclosure, while nylon readily absorbs water from the air, causing the filament to bubble as it's printed. Further, many printers can't handle the high temperatures needed to work with these materials. If you're going to print this yourself, I recommend using PETG and inspecting the part after a few days of use to see if it's warping. PETG is reasonably easy to print and comes close to your target heat resistance. If you're going to get someone else to print it, I recommend using ABS. It's probably what the original part was made of, and anyone willing to print ABS for you will have the heated enclosure and ventilation system to deal with printing it. I'd avoid polycarbonate unless you know the original part was made of it. Although PC is strong and heat resistant, it's also somewhat brittle and vulnerable to scratching. High-temperature PLA is also brittle, and requires a heat-treating step that will change the dimensions of the part. It will likely take several tries to get something that comes out the right size, and even then, you risk having the part break when your car hits a bump.
Nylon is probably your best bet. It is resistant to some chemicals as well. Figure on printing it at around 250 °C. It might do the job.
Are large format prints more brittle? Recently on one of her videos a YouTuber stated that prints from large format printers are more brittle than if you were to print them in parts and glue them together. This seems to contradict the testimonials from the customers of a large format printer, who say that they get good prints from those printers (which print have a print area of a meter square). Would a print form a larger format printer be more brittle than a print made of smaller pieces super glued together? (with all other aspects being equal e.g. the nozzle, the temps, the material and the shape of the object). The YouTube didn't cite any source information to back up her claim.
If you break up a large piece into multiple smaller pieces and properly glue them together, you basically add stiffeners (as a result of printing walls). This could lead to a more stiff model; this might have been confused by calling large prints more brittle opposed to constructed models. If printing is conducted at similar conditions on large printers, there shouldn't be a reason why the model becomes more brittle unless the conditions aren't the same. But that would be true for printing at small printers too, e.g. if one print was printed in a draft.
I'd recommend getting the object to fit together by design, rather than glue - though I tend (if the item is never to be disassembled) use Zap-a-gap - that stuff sticks like crazy though you must not squeeze the parts together but let it naturally sit.
Looking for information on modifying Boots Industries 3D printers About two years ago I purchased a Boots Industries v.2.5 3D printer in a delta configuration. The wiki is still up at this location where you can see what the printer looked like. The printer is prone to break and slip because the design uses thin wire to move the carriages up and down. I'd like to convert my printer over to using a belt like the one on this page, but the conversion kit is no longer available. I believe I can 3D print the carriages, but would anyone be able to point me toward the resources to convert my printer over to using a toothed belt?
Contact Brian or Lykle. We all have Boots Industries printers and have heavily modified it to belt drive and magnetic effectors. The print quality has improved drastically. The old Boots support group has since moved on to Slack and we are in constant contact everyday. https://biv25.slack.com/messages/@slackbot/ https://www.thingiverse.com/thing:1843195
Contact Brian or Lykle. We all have Boots Industries printers and have heavily modified it to belt drive and magnetic effectors. The print quality has improved drastically. The old Boots support group has since moved on to Slack and we are in constant contact everyday. https://biv25.slack.com/messages/@slackbot/ https://www.thingiverse.com/thing:1843195
Best method to make dissolvable supports? Sometimes supports are very difficult to remove (physically) when I print with ABS. The image below, from Thingiverse - MOF-5 unit cell, is after significant effort to remove the yellow ABS supports from a black ABS model. I've learned about polyvinylalcohol (PVA) as a 3D-printing filament which is soluble in water. It seems like a great option for dual-extruder printers, where you can print the model with ABS or PLA and the supports with PVA, then throw the whole thing in water and let the supports dissolve. I've considered other options since PVA is ~4x more expensive than ABS. Doing a PLA print with ABS supports, followed by dissolution of ABS with acetone, is my best idea currently. Has anyone found success with another option?
I've had great success printing with HIPS (high-impact polystyrene) as a support for both PLA and ABS. Most sites recommend it for use with ABS because the materials melt at similar temperatures and work best with heated beds, but I've had good luck using it as a support material with PLA on a bed at 60°C. It doesn't stick as well to PLA as it does to ABS, so supports tend to peel away very readily. The downside is that, if you need the support to anchor your print at all, it doesn’t really stick well enough to accomplish this task. For that, you must pair HIPS with ABS. When you print with ABS or have complicated interwoven support structures, HIPS can be dissolved with D-limonene, a citrus based cleaner sold under various names like Citrisolv (others exist), or with dipentene (a mixture of L and D-limonene that doesn't smell as pleasant). Regarding cost: I've found HIPS to be slightly more expensive than PLA/ABS, but only 1.5x the cost, not 4x like PVA. Additionally, it isn't hydroscopic in the same way as PVA so it lasts longer out of the package. Since you're using it as support, you also tend to use far less filament than you do for the main print (sparse support structures as opposed to solid print structures). Water-soluble alternatives: There are a few proprietary blends of polymers sold by the big commercial printer manufacturers (3DSystems, Stratasys) that only work in their machines… these are generally soluble in basic solutions (water + sodium hydroxide or sodium carbonate). These are usually very expensive and you'd have to rewind the filament on a spool, as they come in cartridges made for specific printers. You'd also have to experiment with the right build conditions and solution blends to remove the material afterward. Airwolf has a support material called Hydrofill that purports to be soluble in plain water… I'm not sure how this is different from standard PVA, though I assume it is different. Hopefully more companies will work on developing water-soluble options to help us keep the 3D printing world full of renewable, less-environmentally-harmful options for filaments (both print and support). Update: Ultimaker now has a material called Ultimaker Breakaway. After using it for a few models, it works remarkably well, allowing me, for the first time, to print nice rounded surfaces on the bases of my prints. It really does just break away from the surface, much like HIPS but without the lack of adhesion problems between HIPS and PLA.
While I haven't used PVA yet, think of it this way, it may be 4x the cost but you use significantly less material for support structures even if you have a lot of support.
How does this multiple-curve-beam flexural pivot work? I found this in Handbook of Compliant Mechanisms (2013), page 162, or at the start of "Chapter 11, Elements of Mechanisms," subsection 11.1.2 Revolute I don't understand how it's supposed to go from 1 to 2 when b rotates around c. The description reads: This element is a rotational flexural pivot constructed by three curved beams to achieve a large range of motion. Theoretically, this element will rotate without axial-drift motion, because of the symmetric arrangement about the axis. (1) Rigid body a is fixed. Rigid body b rotates about c-axis. (2) Deformed configuration (3) Photo of the device. It's unclear which part of the beam is attached to what. I can understand how one curved beam could switch its curvature (in general, like they do in bi-stable latches), but I don't see how they could both at the end of of the rotation end up curved in a way that's opposite to how they started. How could c2/c3 go from the configuration in 1 to the configuration in 2 ? Or could they be two different iterations of the same idea ? I can see how (1) or (2) would resist rotation of c around b, and snap it back to its original position. The text claims (2) is the deformed configuration though. I can also see how (1) with just c1, c3, c5 would deform to (1) with c0, c2, c4 if (b) was turning anti clockwise. Also, what would be an ideal material to print this kind of compliant mechanism ?
The picture looks as if there is a printed back surface to which the central part is attached. This can not be the case, since otherwise the curved members would not be able to flex. Everything inside the outer ring must be detached from the back shell. Like all couplers, especially flexible couplers, there is a limit to the amount of torque the coupler can transmit. Within the elastic limits of the material this looks like a good design, and a good match for 3D printing. The forces are along the layers, not across the layers. This looks like a good design for a rotary coupler. I'm resisting calling it a "shaft" coupler since neither side is equipped with any connection to a shaft. One could modify the design to have a larger central hub with a shaft hole (and set screw(s)). As it is, there is an implied method to hold the outer ring, and a fairly explicit three-pin adapter to drop into the open slots in the arms which connect to the center. I would use ABS rather than PLA, although it depends on the stiffness you require and ability to sustein abuse. PLA is stiller than ABS, while ABS, within the elastic limits, is more compliant. I am not confident that either of these plastics would stand up to thousands of millions of flexures. I would prefer to print this of TPU or Nylon. Both of these are tougher than ABS and PLA. They withstand greater flex with fewer problems with micro cracks and degradation. I have printed another shaft coupler of TPU, and it was both compliant and still. It isn't possible to really nail down a material without knowing the application. As to your question about one example being a transform of the other, I don't thing they are. The A and B drawings are similar in function, but are not stressed and unstressed versions of the same part. Either will work as a coupler.
I would try and print it in PLA because it is quite flexible. ABS is harder in my experience and breaks more easily. If you are able to print PET, you should also try that for the same reasons. There might also be better materials I've never heard of, also please leave a comment if I'm wrong with anything.
What are the tradeoffs of a cantilever printer design? There are several cantilever-style printers around like some DIYs (2018), the TronXY-X1, the Monoprice Select Mini, Tiertime's Cetus line and the 2019 Prusa Minilinks for information, no affiliation. While the specs of the printers are very different, what are the main benefits, tradeoffs and drawbacks that are inherent in the design compared to a more traditional Core-XY or Prusa/Portal design?
Cantilever printers constrain the X-axis carriage only on one side and have a free hanging side. This means, that the X-beam has to be rather sturdy and is limited in length, making them usually quite small machines. Another tradeoff stemming from this source is, that they have to stiffen this one connection as much as possible. Any play on the connection leads to increasingly large errors the further the printhead is out from the Z-Axis and can lead to bad prints, banding and total failures of prints. This behavior, at least on the TronXY-X1, is sometimes called Z-Wobble and has brought up some aftermarket modifications. For example, this bracket constrains the movement of the X-cantilever. While all axis would be affected by bad constraints, a cantilever suffers the most: The cantilever (if not well constrained) can deflect in both XY and XZ plane, pretty much opening a cone with the tip at the leadscrew nut. A one side-leadscrew portal design can only deflect in the XZ plane, opening an angle along with the center in the leadscrew nut. A double leadscrew design has, constrains both ends of the X-axis and prevents deflection but could suffer from bending the portal bar. The main benefit of a cantilever design is that being of a rather simple design with few moving parts, understanding as well as repairing and modifying the motion control system is fairly easy. The ease of access and low component number also allows reducing manufacturing costs to some degree. Another thing of note is, that these printers usually have a fairly small form factor, allowing them to be used in small workshops without taking up too much space. However, their build-volume to occupied-volume ratio usually is not as good as other designs - printers with a much larger build volume just get a much better ratio there.
The point of the cantilever design is for reduced cost. There are no other benefits beyond that.
Filament isn't going into the Bowden tube, instead it goes "into the room" The pictures explain my problem. I have already tried to reduce the retraction but that showed no effect. Thank you for your help. The effect is a total stop of the print (no material is extruded). Creality CR 10 Cura 3.4.1 I recently added this new feeder aluminium block because the 3D printed stock version was bad quality.
This is typically caused by resistance in the tube or hotend but in your case it appears to be mostly caused by a very poorly designed extruder. The filament needs to be constrained closer to the drive gear. You may be able to drill out the PTC connector to allow the PTFE tube to reach closer to the gears or print a spacer to fit in between but you need to support the filament in that gap. You can try raising the hotend temperature as a band-aid until you can fix the problem, do not exceed 240 C if you have a PTFE lined hotend. Long retractions can also pull molten filament into the cold zone where it solidifies and make extrusion harder. As an example, here is a picture of a Bondtech BMG extruder. Note how the extruder constrains the filament path all the way from the drive gear to the hotend entrance. While this example is extreme for normal PLA/PETG/ABS, it is required for flexible filament. A 4mm gap (or closer) should be fine for PLA/PETG/ABS or other hard filaments.
You just have to cut the filament at a 45° angle. Then push down the extruder and push the filament in. Then heat up the nozzle. Next, hand feed the filament until the filament starts oozing out.
Should you use hairspray on a metal bed 3D printer? I've heard that using hairspray is useful for keeping the 3D objects from peeling off of the bed, but every example I have seen where someone uses hairspray, they use it on a glass bed. Is it okay to use it on a metal bed as well?
I have been using a sort of a very strong hairspray called 3DLAC for about 2 years directly onto the aluminium heat bed of the Anet A8 printer I have. Basically, all those sprays contain copolymer constituents, PVA (PolyVinyl Alcohol), Vinyl or Acetate. These are also found in certain glue sticks or wood glues. For me this spray works perfectly! On day one I assembled the printer, the paper tape tore and I was too anxious to wait for new tape to arrive. This worked so well that I have not changed it for that printer. Cleaning is very easy as PVA or any of those constituents are solvable in water, so a moist cloth or paper towel over the plate is all to clean it. Furthermore, you do not require to spray before every print. To answer your question if you should use a PVA based spray like hairspray directly onto the metal build plate is a matter of preference, but you definitely could use it as I have been doing it for about 2 years. To address the comments: I spray the heat bed platform whilst it is attached to the printer. I do pull it forward and gently spray the bed or just the location where the print is going to be build. Note that you do not need to do that for every print. I recently did notice very little spray on the X guide rods (maybe I have been careless once or twice), but that has not been a problem for my Chinesium iGus ripoff plastic bearings. It is very easy to clean with a damp cloth. It also works great on the glass bed of my Ultimaker 3E, but I usually (unless when I'm lazy ;) ) remove the slate of glass before printing. You could consider shielding the rods with a piece of paper, but it has not been necessary for 2 years.
Do you mean bare metal or metal with some film on top? You can apply hair spray to bare metal, but you will have troubles cleaning it off. Solvents do not evaporate hair spray, they only turn it into thick sticky goo you will need to clean off. I recommend that you try glue stick or even beer (seriously) before hair spray. You should also know that any adhesive is required only when your printer has bed leveling issues. After I finally dealt with bed leveling, I print ABS, and even ABS/PC mix (nasty) without any adhesive, on a bare kapton film over a metal bed. I rub it with alcohol before printing.
Is there any way to shrink 3 mm filaments down to 2.85 mm? Some 3 mm filaments seem to actually be 3 mm - is there any way to shave off the excess and use it as 2.85 mm?
I would not recommend you to try and somehow re-size the filament, since even the smallest of irregularities and error in diameter occurring from such a process would ruin your prints with sporadic over and under extrusion. Rather, if you have the tools available, you could grind the filament into pellets, and use a filament extruder to make it anew with your desired diameter. Alternatively, depending on your printer setup, you might very well extrude true 3.00 filament with your 2.85 mm filament printer. If you try to do that, make sure to: Adjust filament diameter in your slicer Check that your filament isn't getting squashed by the extruder wheel Check that all mechanical parts actually can pass through your filament freely I do not own a 2.85 mm printer myself, and therefore have not tried this procedure. There are, however, several people who seem to have done this successfully.
you could put a 3 mm extruder on you're printer. or (if you're r brave) put route filament in the oven for 10 minutes and unroll it than stretch it out. if you do this outside you can attach one ent to a stride licht or something an pul on the other end, if it is hot anoth it should hold its new position. but don't get it to hot or the filament will stick to itself, you can put talk powder on it to prevent this but be shore to watch it off before using your filament. i have never tried this and not komw is it will work. you can try it on i short pease of about 2 m first to test it oud. if it works, please tell me. i would love to know.
Why all the excitement about linear rails? Whenever a 3D printer that uses linear rails is announced (case in point: the cetus), the Internet (well... at least that corner of it dealing with 3D printing) gets abuzz with excitement. I researched a bit the topic myself, and while I understand that linear rails can be produced to a fantastic degree of precision for super-heavy machinery, it escapes my comprehension why they are considered superior "by default", relative to the classic linear bearings on a shaft. 3D printing is a lightweight application, and the motion of at least 2 axis does not happen against a solid surface (where you could bolt a linear rail every few cm) but suspended between the 2 ends of the axis. Furthermore, the internals of the bearings used on linear rails are substantially identical to those used on a shaft. The cetus site says under the heading "Quality Linear Rails": Self-lubricated | Maintenance Free | High Precision | Long Lifespan | Quiet but this - in my experience - can be said of "Quality Linear Bearings on a Shaft" as well, and in some cases even bushing deliver to a high standard on 3D printers. So, what am I missing?
The following is a compilation of the input from a number of sources. Linear rails in general are mechanical components that - when designing equipment - offer great flexibility. The profile of the rail can be designed in nearly infinite ways. This in turn allows for: Different levels of stiffness in different directions (for example you may have stresses only on a given plane, or you may actually want the rail to slightly flex in one plane but not in another one). Placing the surfaces for the rollers strategically, for example in a location that is unlikely to get contaminated, or where the maximum force will be applied. Curved paths, so that the carriage can move along a line that is not straight. Because the contact surface between the rollers and the bearings is flat, cylinders can be used instead of spheres. This in turns diminishes the mechanical stresses, and the amount of play, increases longevity and allows for more bearing capacity, among others. Linear rails can be anchored along their full length, rather that at their extremes, thus increasing the accuracy of their positioning, their stiffness and their bearing capacity. Linear rails can be machined while pre-loaded, thus achieving maximum accuracy when in use, rather than when coming out of the factory. The bearings on a linear rail only allow for one degree of movement. There need to be two rods with linear bearings/bushes to achieve the same result. All that said, when it comes to the specific application of consumer-grade FDM 3D printers, it seems that none of the above is very relevant, nor confers any real advantage to the printer in terms of quality of the final print: the mechanical stresses involved in 3D printing are very small, the movements all happen along straight lines, most of the axis cannot be anchored to a large, rigid body, ... On the other hand, the design with rods + linear bearings is cheap, effective, simpler and lightweight, all characteristics that are highly desirable in a 3D printer. All in all, it seems that there is no good reason to prefer linear rails over rods in general. Still, there may be specific designs that may benefit from their adoption. I postulate that the Cetus printer linked in the question may be such a design: the cantilever arrangement of its axis - for example - is well served by the fact that a single rail locks movement in all but one direction, and the orientation of the X rail offers maximum rigidity against the action of gravity.
Linear rails will always produce a high degree of accuracy and stability and more so than round rod with PTFE bushings and/or bearings. One may argue the fact however even as a product developer and one who is involved with the mechanicals and development of machinery on a day to day basis comparing the two we see “significant” improvements using rails over rods and if properly utilized on a Z axis print bed you will have a bed free of leveling issues that can smoothly and accurately operate with one driver versus two. I will further add that getting rods aligned perfectly is a difficult task for the average person and even a slight twist or angular position can affect final print quality. I’ve seen many linear rods that appear visually straight and when chucked into a lathe spindle with 0.0000 accuracy there will be 0.005 or more in runout. In fact I’ve yet to see one perfecting concentric motion that is longer than 6 inches. This tells me that they cannot plausibly be as accurate and that while they may function they will never function with a high degree of accuracy. Do we need higher accuracy in 3D printer axis? Sure we can have quality control boards that compensate to some degree however the mechanics of the machine are utmost important before you choose the quality of board and software. Why install a \$300 motion control on a cheap linear rod printer if you’re not going to see the full benefits? With technology further advancing into 64 bit and eventually 128 bit and higher degrees of precisional accuracy 3D printing is turning a page and will if not already be capable of micro resolutional accuracy and can only do so if all the components function properly together. So sure, your rod guided printer works. However, it will never work as well as my linear rail guided printer with ballscrews and servos. You can have your layered fuzzy prints. I will keep my smooth finished injection molded looking parts that are made from materials a typical desktop cannot even print. So to argue it’s not needed is arguing that high quality isn’t accepted in a lower price point market. One other addition here. Ask yourself how level and square is your printer? I’m not talking about using a carpenter's level for checking your machine they can be inaccurate up to a 1/4” per 10 feet. When you can dial your printers bed down to 0.00005 or less every direction and your structure is just as accurate than you know what a quality printer and print looks like. I guarantee no printer priced \$300-\$1000 comes even close to that degree of accuracy. The average consumer is so drawn into the technology of a final print itself they overlook the precisional aspects involved and learn to settle for less. Then you wouldn’t expect your \$500 printer to compete with my \$10,000 printer either. Bottom line you get what you pay for.
Connecting Anet A8 to PC issue I am still at calibration stage and need some info from the PCB. I connected the USB and ran Repetier. The PCB wants to talk at a higher baud rate than my serial port says it can do. I tried setting the serial port to its highest setting 125k and reduced the PCB baud in Repetier setting to 125k. No joy. PC port reverts to 9600 every time I check it. Thoughts? PC running Windows 7 Home Premium 32bit.
Sorted. Repetier Server was hijacking my com port. Uninstalled it and Repetier host worked fine. As I have no desire to monitor or control prints remotely I have no use for the server software. Hope that helps others.
My A8 just started doing this. I shut down repetier server, and voila, all is good! Since I will go do OctoPi eventually, I just uninstalled RS.
3D prints keep suffering from warping I have a Prusa i3 made by Geeetech. My 3D prints keep suffering from warping when printing with PLA. Whenever I print something with a base at about 10 cm x 10 cm, at least one corner of the print would warp up. I've read numerous articles about warping and tried all sorts of methods. My printer's bed is level, and heated to 60°C. My bed is made from clean glass. I've tried all sorts of adhesives. I tried blue tape, and used hair spray. The only way for me to combat this is gluing the base to blue tape with 502 Glue. I used brim and the whole brim just warps up. I sometimes leave the model printing over night. For the first few hours it's perfectly flat. When I go back to it the next morning I'd find one corner warped up. This is very dysfunctional to my prints. Is there a reliable way to stop this warping from happening?
For ABS it will warp unless you build a heat chamber. That said the tricks to reduce warping come down to: Material, i.e. PLA is less likely to warp; Use a fan, it helps so much; Make sure you have temps calibrated well - Too hot is more warp; Use a raft. The Makerbot uses a raft and no heated bed; Make sure the room is not drafty. Having it by the window will result in warping; Adding a large brim also helps; I find good ol' glue sticks work the best at keeping the print to the bed; SMASH the first layer. This one is controversial. I personally do first layer at 130% and print speed of 30%. You get elephants foot sure, but it's on the bed real good. Tom is right. It is very very hard to print that big of a piece without warping. That said I have done very large pieces on my Ultimaker, using a fan, glue stick, MatterHackers PRO PLA and no raft. But again that's on an Ultimaker. Note you can build a heat chamber pretty easily. Specifically a passive heat chamber.
Source (at least in part): SolidWorks 2015 Tutorial with Video Instruction, page 11-6, and; 12 ways to fight warping and curling, June 23, 2011 by MakerBlock Curling As printed plastic parts cool the different areas of the object can cool at different rates. 1 Depending upon the parts being printed, this effect can lead to warping and curling. Although PLA has a much lower shrinkage factor than ABS, both can warp and curl, potentially ruining a print. There are some very common ways to deal with this potential problem, the most notable being a heated build platform. However, sometimes that might not be enough. Use a heated build platform. A heated build platform helps keep the lowest levels of a print warm as the higher layers are printed. This allows the overall print to cool more evenly. A heated build platform, sometimes abbreviated as HBP, helps tremendously with just about any ABS print and large PLA prints. Print with a raft. Rafts are a printing option in ReplicatorG and Skeinforge. They’re basically a large flat lattice work of printed material underneath the lower-most layer of your printed object. They’ll also help reduce warping and curling by allowing your printed object to adhere better to your flat build surface. Other variations on this are to print with a larger raft and/or a thicker raft comprised of more layers. Calibrate your starting Z height. A good first layer makes all the difference. If your starting Z axis height is too high, the extruded filament won’t be able to make a good bond with the platform. If you think your Z axis starting height is too high, try lowering it by 0.05mm increments until you find a good first layer. Get the right build surface. Some people have experimented with different surfaces such as steel, titanium, glass, different kinds of plastic, different kinds of tape, and foam board. However, I find both ABS and PLA seem to stick really well to hot or warm Kapton tape. Clean your build surface. ABS and PLA stick better to a clean build surface. Keep it clean of dust, pieces of old prints, and any other debris. Print slower. Printing slower allows finer detail, better adhesion to the build surface and lower layers, and gives the printed part more time to cool evenly. Print cooler. Printing at a lower temperature isn’t always an option. Ideally, you should be printing at the lowest temperature required for extrusion and that allows good interlayer adhesion. However, trying lower temperatures isn’t for the faint of heart. Printing at a too low a temperature could cause harm to your extruder motor or extruder. Eliminate drafts or enclose your robot. Forrest Higgs found that having his 3D printer too close to an open window caused very uneven heating across his build surface. This in turn caused the side of his prints closest to the window to curl. Since keeping the window closed wasn’t an option for him, he compensated for the window drafts by adding a heat lamp. Cupcake and Thing-O-Matic owners might have an easier time of eliminating drafts by simply enclosing two or three of the sides of their robots. It will also have a fortunate side effect of helping to control fumes. Design with mouse ears. Zach Smith’s solution was to add little discs to corners of an object to help those corners stick to the platform. These essentially serve as “mini-rafts” to give those corners more surface area and better adhesion without having to print an entire raft. Design with aprons to hold down corners. Forrest Higgs suggested adding “aprons” around an object to be printed, while that object was being printed on a raft. These low thick pieces of plastic help keep the raft flat and help prevent any curling or warping from affecting the desired printed object itself. Design with surrounding thermal walls. While Forrest Higgs’ apron approach provides a mechanical advantage of essentially holding down corners with a chunk of plastic, Nophead has added thin surrounding walls to his designs to act as baffles to keep warm air around the printed object as it moves around. He’s postulated that a very thin surrounding wall could have the same beneficial effect as printing inside an enclosed build chamber. Interestingly, it seems that Nophead suggests that designing objects with more rounded corners might also help avoid curling and warping at those corners. Reduce infill. When printing a model you can chose to print it hollow, completely solid, or some percentage between zero and 100. However, as Nophead points out even the plastic inside a model exerts a force on the entire printed object as it cools. It stands to reason that the more plastic you have, the more those pieces of plastic will pull against themselves and the build surface as they cool. By reducing infill there will a reduced amount of internal tension as the object cools. Reducing these internal forces by printing with a lower infill ratio can help reduce curling and warping as well.
Danger in 3D printing over a lithium ion battery So I bought a Lulzbot Mini a couple months ago and finally downloaded Ultimaker's version of Cura... Boy... have I been missing out... One feature Ultimaker Cura implemented that I've been looking for is a "pause at z-height" feature ("post-processing tool"). I'm building prototypes of an electronics device, and creating two pieces that snap together looks a lot worse than printing a single piece. If I could pause my print, insert my electronics, and continue printing, my device would look a lot more professional (even if it took longer to make). My one concern is the lithium ion battery. Right now I'm printing in TPU. With a heated bed of 40 degrees Celsius, and a heated extruder at 240 degrees Celsius, there seems to be a significant risk that the lithium ion battery reaches a temperature above 60 degrees Celsius (damaging the cell, causing a potential explosion). Granted, I am not sure what "60 degrees Celsius" actually means. It could mean only one part of the packaging needs to reach this temperature, or it could mean the entire LiPo's internal temperature would need to reach this. In either case, the numbers don't look good. On the other hand, the heated bed surely doesn't need to remain heated beyond the first few layers? Additionally, I can create a "roof" for the LiIon battery that I can slip it under, providing some insulative TPU before the rest of the device. I think the print would happen safely like this, but obviously, an explosion would be really really bad. Like it would probably burn my house down, and I would be asleep when it happened. Does anyone have any experience doing this? Is there a way to turn off the heated bed mid-print? I guess I can insert a g-code line during the pause? Will this affect the remainder of the print you think? Am I being paranoid? Can the extruder actually pass heat through a 1–2 mm of insulation and cause an explosion? Anyone know how heat travels from the initial, liquid print material through the rest of the structure? Any more advice or things I should consider before pursuing this? A more specific pause type might be helpful, if anyone knows of any.
I would have to see your design to comment more but why not just change your model with the lid or top that can be either fused on later or attached in some other fashion? It is possible to turn off the heated bed after a certain layer. It looks like there is a discussion here with the G-code: Can Cura Turn Off a Heated Bed Partway Into a Print? I would personally avoid exposing the battery to the heat generated by the hotend or the build plate. Printers alone can burn your house down. Why combine it with the awesome power of a lithium ion battery?
Discharge the battery entirely, then there is no danger of ignition if the membrane breaks. The battery cells are usually protected by Kapton tape (LiPo) or maybe even metal (LiIon). Both will withstand the temperature of 240°C. The downside is that the cells are damaged after a deep discharge. You should discard the battery afterwards. What I would expect when you do that: A deeply discharged battery will bloat up because the solvents will gas out. This will also happen if you heat up the battery enough (>>60 degree). The gases themselves can be ignited again but are captured in our air sealed Kapton envelope :) Igniting these gases will be comparable to a short flash with a spray can. The temperature is not enough to damage the envelope. The solvents will be also not very dangerous in that temperature range. In case of mechanical breakdown of the membrane nothing will happen anymore. From Solvents and Additives:
Are there any "best" or at least "common" practices to handle allowances in OpenSCAD code? I'm designing a few mechanisms with OpenSCAD, and one of the parameters that need adjustment between printers/filaments is allowances between pieces that need to be assembled together (for example, if I want a "5 mm diameter pin" to fasten two pieces together, how much larger than 5 mm will the holes / how much smaller than 5 mm will the pin need to be). Currently I approach the problem by setting a global variable allowance and manually using it in the code, something like: module pin(radius) { cylinder(r = radius + allowance) } module hole(radius) { cylinder(r = radius - allowance) } I have no real world experience with design though, so I wonder if there are common or coding best practices to account for allowances when designing parts like for example: specific modules to be used, conventional names for variables specific techniques to be used (scaling? vectors?) conventions (like only use tolerances on the fastener, not the fastened object) ...? To clarify: I'm not looking on advice on how to plan the dimensions of my designs. Rather, I am looking for advice on how to organise the OpenSCAD code generating them.
It helps to understand the different aspects of dimensions, so you can use the terminology correctly. This will help you define your variables in OpenSCAD with correct names. (Tolerance is the wrong term to use.) And once you have correct names, you'll understand how to specify the dimensions in OpenSCAD. Tolerance is the amount of random deviation or variation permitted for a given dimension. Allowance is a planned difference between a nominal or reference value and an exact value. Clearance is the intentional space between two parts. Interference is the intentional overlap between two parts. Two other terms Accuracy is the maximum dimensional variation between parts. A machine cannot produce parts with a tighter tolerance than its accuracy. Precision is the size of the steps your machine is capable of. It is often confused with accuracy. In your case you need to define the allowance in order to create the clearance you desire. To design your 5 mm pin and 5 mm hole, you need to understand your machine's accuracy. The printer could print the pin larger than 5 mm or smaller than 5 mm. Or it could print the hole larger than 5 mm or smaller than 5 mm. You'll need to print some pins and holes and measure the differences between what you defined and what you printed. The difference between the largest and smallest measurements you take is your machine's accuracy. And be sure to check the accuracy in your X, Y, and Z dimensions; your printer might have a difference between them that would impact the roundness of the parts. Let's say that your printer's measured accuracy is ± 0.2 mm. Then, we move to clearance. What is the minimum gap between parts you are looking for, and what is the maximum you can accept? Let's say you want a clearance of at least 0.2 mm between the pin and hole, but no more than 1.0 mm. Since your accuracy is ± 0.2 mm, your pin will be 5.0 ± 0.2 mm, so the hole must therefore be 5.6 mm ± 0.2 mm. The minimum tolerance condition would be an minimum sized hole (5.4 mm) and a maximum sized pin (5.2 mm); the maximum tolerance would be a maximum sized hole (5.8 mm) and a minimum sized pin (4.8 mm). Note that a clearance of 1.0 mm might be too sloppy for your application. You might think to tighten the tolerances to 0.05 mm in order to reduce the clearance. But if your printer can't produce a part that meets your specified tolerances, you would need to find a different way to manufacture or finish the parts.
Well, the tolerances will depend on material to be used for fabrication of the required part and also where the part will go and fit. Remember the all parts need some clearance to fit properly. Few years ago (10 years) I was working as Quality Engineer and some Design Engineers were complaining about a Dupont pin was not fitting on the PCB so they told me that I need to force the PCB manufacturing to increase the holes to the higher tolerance. Which I had to ask him firstly the pin size and told me 0.70 mm and hole size 0.80 and maximum 0.90 - hmmm and maximum size of the pin? I asked, and they told me proudly 0.78mm so the part will fit perfectly. - Oh, so one square pin of 0.78mm will fit on one hole of 0.9mm, but what about the diagonal dimension? if the pin on the higher dimension is close to 1.2mm. Imagine what happened later, engineering changes and modifying other designs due improper tolerances. pro-engineer software was unable to calculate what the designers needed. Material has shrinkage, warpage, and other conditions that is needed to know like malleability and hardness and some of this data is on the specification material or the part specification.
How to automate printing of multiple parts continuously? I would like to print multiple parts continuously (non-interactively), so I can leave the printer alone for a longer time. So after finish, parts could be moved somehow out from the printing area, so the next can start. Are there any methods of achieving that with standard desktop printers without having to use multiple printers?
The only thing I can think of off hand is an old mod for the early MakerBot machines. It first was released for the Thing-O'-Matic I believe, but is compatible with Replicator 1 machines (and its knock-offs). Here's the Thingiverse page, but look up Automatic Build Plate. Essentially, you can use the Replicator G slicing program and there is a setting for "ABP" or Automatic Build Plate. This will basically tell the ABP to run its routine after the controller receives the response that the printing program is done and roll the finished part off the edge of the build plate, then start the same program over again. Drawbacks: I don't think it's easily compatible with newer machines/slicers. But, it's open source Pretty sure you have to use Replicator G, which is outdated now and may make your machine sound like it's going to fall apart (I know from experience) Going off of @Pete's answer about solenoids. It reminded me that someone integrated a solenoid "ejector" (aka Boxing Glove) for their machine. Update (06/08/2016): Forgot to mention that if you choose to create your own "Boxing Glove" or conveyor belt, some software such as Octo-Pi and Repetier-Host allow plugins. So, you could interface with your hardware via customized code and integrate the functionality directly into the slicing application for the full closed loop operation.
I Don't really think that it is possible without hardware modifications, or maybe some small parts that will fit in the bed of the printer all on the same time
Filament extrusion always stops at some point during print I own a Creality CR-10 (using Ultimaker Cura for slicing) and I am experiencing terrible printing problems. My problem: I am not able to print anything that takes several hours to print. For parts printed in 30 minutes or so, my printer usually works. The prints end up like this (unfinished and with a lot of stringing) This below is actually the best result I got so far.. (important note: there is no under extrusion during the print, it suddenly stops out of nowhere..) (Although I have to use terrible retraction settings and need a lot of post processing because of all the stringing - you might have an answer for that too.. I just couldnt get the 3d printer to print parts properly with good retraction settings.) The problem is that at some point of the print, no more (PLA) material is extruded and the printer moves without actually printing anything. The first couple of layers usually work (you can see that in the images) but after wasting almost 600 grams of PLA, I am not able to find a solution myself. . I have some suspicions: Could the problem be the angle, the PLA is inserted into the feeder? (so that it is almost a 90 degree bend)? The filament comes from a spool in the right hand side. But I dont think that this would cause such a problem.. Is it a software problem? Here are all my cura settings (I even reduced the printing speed to 30mm/s at 205°C - still didnt work..) :-( Could the length of the bowden tube and it tangling up be a problem? (as you can see in the image below) And annother important thing: The feeder always grinds into the filament (even at those low retraction settings) and it is always very hard to pull the filament out of the bowden after a failed print. Sometimes its almost impossible and i have to use heavy tools for it.. that should be the probelem I already did some atomic pulls, replaced the nozzle and switched the bowden tube. I have a dream: My printer printing a part without any stringing and actually finishing the print. Please help me to achieve this dream.. Thank you for your help in advance. :) (and the filament I used, sorry for the bad quality of the pic)
Stringing? The stringing is explained by your relatively low retraction settings, 1.5 mm is not much for a Bowden setup. As do too high printing temperatures. Stopping mid printing? What you are experiencing is called clogging, the extruder cannot push the filament through the hotend and cause the grinding you report. Clogging mid printing is usually caused by insufficient cooling (for the temperature you print at) of the cold end (causing heat creep; slowly increasing temperature of the heat break), especially all-metal hotend assemblies are notoriously known for this problem. Your Bowden tube does make a lot of kinks, maybe you can improve the path of the Bowden tube. Also if you are concerned about the sharp filament intake angle, you could print a filament guide:
I had the same thing! On my anycubic I3 mega. Tried a lot of stuff and got so sick of it so I installed a 10:1 gearbox on my extruder motor. So now it provides 10x more torque and since I havent got the issue anymore where the extruder is just grinding on the filament and the motor is just skipping steps. No because of this and because my anycubic won't connect to a pc I can't change settings on my printer so I resolve the ratio with 1000% flow in my cura settings. And other stuff is also affected. But at least it won't stop and just be sitting there and skipping steps.
How to convert png to stl? Is it possible to convert an image image to STL file format? E.g. I don't need the coloring, I need the lines.
I suggest your objective can best be accomplished by converting the image to a single color vector file. You can do this with Inkscape (free, Linux, Windows, Mac) by combining the built-in bitmap tracing feature with some manual editing. I attempted to do so, but the coarseness of the image would result in excessive manual edit time. It may even be faster to reduce the opacity of the image and create a manual tracing on a second layer. Once you have a vector file, you can use any number of 3D modeling programs to convert the file to a 3D STL. Fusion 360, Tinkercad and OpenSCAD allow import of SVG files to be extruded to user-specified thickness.
For things like a coat of arms, you do not require a full 3D conversion, essentially all you need is a lithograph-like effect. The best program I've found for such conversions is 3D Builder which is a free Microsoft download for windows users (yes, I was surprised too). It can use either color or degrees of greyscale shading to determine the z-axis depth of each region in the picture. Makes the cleanest conversion I've found so far. Exports as a .3mf file but can be readily converted to an .stl for 3D printing (using something like SwiftConverter for example). Also has an option to invert the conversion which I've found quite useful.
Centering model with CuraEngine when creating g-code? How can I center a model at the middle of the printing area of the printer when creating a g-code with CuraEngine. Are there any parameters I can add to ultimaker2.def.json to achieve this? Thanks.
Found a solution. This need to be applied under "settings" "command_line_settings": { "label": "Command Line Settings", "description": "Settings which are only used if CuraEngine isn't called from the Cura frontend.", "type": "category", "enabled": true, "children": { "center_object": { "description": "Whether to center the object on the middle of the build platform (0,0), instead of using the coordinate system in which the object was saved.", "type": "bool", "label": "Center object", "default_value": true, "enabled": true } } }
If this is over the commandline tool "CuraEngine", then you will have to read the sourcecode. According to the Author, 'Nope. Only documentation there is in the code, readme and my head.' (cringe!). If you're talking of the GUI program, then right click and click "Center". But this requires GUI usage. Not so nice if you want to automate using curaengine as your slicer.
How to directly send G-code to printer from a Linux terminal? Should it be possible to directly send G-code to the printer serial connection using pipes under Linux? Example: echo M106 > /dev/ttyUSB0 My controller runs at 250000 baud, I have tried setting the TTY baud rate to 250 kBd with: stty -F /dev/ttyUSB0 250000 But, unfortunately, this particular baud rate appears to be unsupported under Ubuntu, giving the error: stty: invalid argument ‘250000’
For direct low-level printer control from a terminal, without specific software, I found the following solution with full credit thanks to user: http://stackexchange.com/users/6463673/meuh Sharing here as may be of use to other users in the 3d Printing community, and I was unable to source a "complete" solution to this elsewhere. Step 1) Create a custom python script that allows you to set arbitrary baud rates (make executable with chmod u+x). #!/usr/bin/python # set nonstandard baudrate. Original Question: http://unix.stackexchange.com/a/327366/119298 import sys,array,fcntl # from /usr/lib/python2.7/site-packages/serial/serialposix.py # /usr/include/asm-generic/termbits.h for struct termios2 # [2]c_cflag [9]c_ispeed [10]c_ospeed def set_special_baudrate(fd, baudrate): TCGETS2 = 0x802C542A TCSETS2 = 0x402C542B BOTHER = 0o010000 CBAUD = 0o010017 buf = array.array('i', [0] * 64) # is 44 really fcntl.ioctl(fd, TCGETS2, buf) buf[2] &= ~CBAUD buf[2] |= BOTHER buf[9] = buf[10] = baudrate assert(fcntl.ioctl(fd, TCSETS2, buf)==0) fcntl.ioctl(fd, TCGETS2, buf) if buf[9]!=baudrate or buf[10]!=baudrate: print("failed. speed is %d %d" % (buf[9],buf[10])) sys.exit(1) set_special_baudrate(0, int(sys.argv[1])) Step 2) Run the script to set your baud rate. ./set_custom_baud_rate.py <> /dev/ttyUSB0 250000 Step 3) You can now monitor your printer output in a terminal window simply with: tail -f /dev/ttyUSB0 Step 4) And finally, open up a new terminal window, and you can directly send M or Gcode to your printer; example: echo "M115" >> /dev/ttyUSB0
That's work solution: you need 2 terminal, one for in and another for out flows
What is stopping us from mixing 3D filament colors in an Extruder? This came up in one of my groups today. That we could not color bend, or mix 3d printing filaments. I have researched but I am not finding anything talking about Plastic mixing in an extruder. Why is it that we cannot take say a Diamond hotend, or a hotend with 5+ inputs, and mix any color we want? (assuming all the same type, ABSm, PLA). I think it would be interesting to at the least get a gradient effect on prints. The best I have seen is natural plastic and a marker system. Or a powder / advanced / out of hobbyist price range process that sprays ink. The only Color Bending I know of is with Recycled plastic that uses multi color. Not quite what I am looking for. Thanks!
I just started with google and phrase "3d printing color mixing" and on the first place (in fact first two were valueless adverts) I got this Instructables - DIY Full Color Mixing 3D Printer. How it works? It uses magenta / cyan / yellow filaments and mixes it while printing with Diamond hotend. It definitely does what you are asking for and it's exactly the same idea you come up with ;) Overview Getting the controller board ready for three extruders... I hacked a RAMBo board to drive three extruders, however, you can use any board you want... (most people use a RUMBA due to it having all the pins/components needed for 3 extruders native on the board) Rewriting Repetier firmware to get color mixing working on your machine. How to install, configure, and use the diamond hotend - tips / tricks / lessons learned / etc... My original Bowden extruder design and various ways to mount the three extruders for your set-up My universal magnetic effector plate and accompanying hotend mounts for quickly swapping various hotends. (delta specific) How to design multi-color models and making STLs that can be exported and used as a individual STLs or combining them into an AMF file for slicing... Configuring color mixing in Repetier and Slic3r to print above mentioned multi-color models. Anything else I can think of later that I can't think of now. Comprehensive overview of Quantum Mechanical Entanglement as it pertains to multi-color printing (just kidding, I don't understand that... But I will cover multi-color printing throughly)
If you clearly observe the geometry of the print it has printed in different colours .the part is designed in such a way that it has the small height gap between the each colour but we cannot seen this gap in one projection plane.. We can print only in direct drive extrusion .by changing the filament of different colours without pausing the print by immediately changing the material . It is printed in single nozzle. The above image clears your doubt.it is designed with different heights.
High temperature flexible filament I need to cool some liquid (250 °C) while it’s flowing through a tube which has to be able to bend and flex. My idea is to make a flexible tube with a second tube spiraling around it through which coolant will flow. I’d like to 3D print this tube if possible so I wonder if there is some printable filament that: doesn’t melt at 250 °C is flexible enough that it can print some tube that can bend (bending radius of 30 cm) optimally also has good heat conductivity Is there any 3D printer filament available that has these properties?
Ok, so to answer the primary question: What flexible filament will operate consistently at 250°C? Man... this is a tough one. Some filaments, like PEEK and ULTEM 1010 can operate up in the 200c range.... but they're not flexible at all. Silicon might be able to work, but you're still pushing boundaries. Now.... I'm lucky to be in a 3D Printing company.... and we're testing a super-high-temp flexible material, very similar in temperatures to ULTEM. I'll definitely check back and let you know how it goes, but.... Honestly, that's so hot, readily available thermoplastics may not be an option unless you're in Aerospace with an unlimited budget..... which, based on the requirements, would make sense, lol! I'd say the most readily available way to get this done would be 3D Printing a mold, in which to put your silicone, and bam -- you've got the part.
3D printing nerd showed a couple of filaments that fits this in his latest video "Printers at RAPID + TCT 2019": Firstly a Nylon 6 high temperature filament: Another part of the video shows another flexible print, created on a four axis 3D printer, using TPU filament:
Managing LEDs with NanoDLP I'm building my own DLP printer with UV LEDs. I use a 20A relay to power them on, since they require high power and can't be directly driven by arduino or Raspberry. I'm planning to use them with NanoDLP on Raspberry + GRBL on Arduino. Is it possible to make NanoDLP tell GRBL to power on (send digital 1/+5V) the relay when the print starts and power it off (send digital 0/GND) when the print ends? The other way would be to just install an on/off switch and do it manually, but I feel this step should be automated somehow. Is there a way to do it?
I have no experience with either GRBL or DLP printers, but the M7 M8 M9 coolant control codes should be able to be sent by NanoDLP to GRBL. Those seem to allow for direct digital output. Apparently you can set the pin you want them to use in the cpu_map.h file, with the standard being Analog Pins 3 and 4 for the M8 and M7 commands respectively. // Define flood and mist coolant enable output pins. #define COOLANT_FLOOD_DDR DDRC #define COOLANT_FLOOD_PORT PORTC #define COOLANT_FLOOD_BIT 3 // Uno Analog Pin 3 #define COOLANT_MIST_DDR DDRC #define COOLANT_MIST_PORT PORTC #define COOLANT_MIST_BIT 4 // Uno Analog Pin 4 Hope this helps! Source: https://github.com/gnea/grbl
I think you don't have to use the Arduino. There´s an option in NanoDLP to control the z-axis through the Raspberry GPIO.
Creating positives for a food-safe silicone mold: safe? First, a little background. A couple of years ago, I was researching making my own candy, and I came across this page: Lego brick shaped gummy candies, describing how to use real Lego bricks as a positive to make silicone molds for Lego brick shaped candy. Now that I have a 3D printer, and inspired by the usual description of ABS filament ("It's the same plastic used to make Lego bricks") it occurs to me that I can now make any positive I want. The question is, would that be safe? I know the filament I'm using is not food safe, but if I create positives for a food safe silicone mold, would toxins leech into the mold? And if so, is there a barrier I can use to prevent this, such as some kind of coating?
In principle, ABS is safe for contact with (cold or room-temperature) food. The two main concerns specific to 3D printing are, assuming you start with a filament that is not itself contaminated: Pores and holes in the printed part which may harbor bacteria Impurities introduced into the plastic during the printing process I doubt that the silicone mold will capture the pores and holes with sufficient detail to be of any concern (it certainly won't capture the internal structure, only the surface). That leaves us with 2. It has been noted that brass nozzles contain trace amounts of lead. This lead can contaminate the printed part, which may in turn contaminate your mold, which may in turn contaminate your food. I don't think this is of realistic concern, since we're looking at trace amounts of trace amounts of lead. The nozzle might also have burnt plastic stuck to it (which might be carcinogenic) so you should make sure to do the print with a very clean nozzle and at a temperature that is not too high. ABS is food safe for contact with cold or room-temperature food. It is however not food safe for contact with hot food, because at higher temperatures the food may leach certain chemicals out of the plastic. Your application is one of low temperature, but silicone is not food and might perhaps leach some contaminants out of the plastic, regardless of temperature. However, this concern is not specific to 3D printing, as it applies to the method of making moulds out of Lego bricks as well. Therefore, making moulds from 3D printed positives does not appear to be different in a food safety perspective from making them out of LEGO blocks.
The plastic is not quite your main concern (though it still can be). You should worry about the cracks and crevices in FDM prints. Bacteria loves to hide there. For most people, this is the first concern when it comes to "Is X 3D printed food-safe?" If the end product is a hard material, you should sand or smooth your print to prevent the layer crevices from appearing in the end use product. Also consider food-safe epoxies for filling in gaps. If the end product is made of a flexible silicone, then this is less of a concern. ABS is not food safe. PLA as a material is considered food safe by the 3D printing community, but I have not seen a scientific study on this. However, many filament makers do not extrude pure PLA. Therefore, the answer to your question is that it depends on the manufacturer. You will need to contact the company to know what is in the filament first. Beware of untrustworthy manufacturers that claim food safety without backing that up.
First layer eventually sticks to the hotend and tears I am at my wits end with this problem. I start a print and the skirt goes down fine, then the outline of the parts go down fine (usually) and then when it goes to fill in the first layer, it will always get stuck to the hotend at some point and rip apart the layer. Any ideas on how to solve this? Printing PLA at 210°C; First layer temp is 225°C; Bed temp at 60°C; 1.75 mm filament, and; 0.4 mm nozzle. Maker Select V2.1, using Cura to slice.
Step Zero: is always to check/adjust the bed level - if the height over the bed varies while putting down the first layer, it's hard to ever get first layer settings that work. Step One: is to adjust your first layer settings - height, temperature, extrusion width until you find a set that work for your setup (knowing that they may change somewhat when you change filament.) Some folks find more success with thin first layers, others with a thicker first layer to pump more plastic, still others use the same layer height but increase the width to pump more plastic, and others combine these approaches. Increasing the temperature is common, though in my case I found that the "usual" +5°C was not enough for the present setup. You can change one setting at a time and have an idea of what works better or worse in each case, or you can change lots of settings and hope you get lucky. I prefer the tedious approach, it's less maddening. Are you using any surface treatment on the bed? A bit of gluestick or hairspray may help you stick.
Try lowering your first layer temp to 205°C, and apply a gluestick to the bed before you print anything. By lowering the temp, this allows th plastic to cool faster, and doesn't risk the plastic sticking to the nozzle as much. Applying the glue stick will help the print to stick to the buildplate better, lowering the risk of it getting stuck to the nozzle. You might also want to try using a raft, as this will provide more surface area for the print to stick to the bed with, and lower the chances of it attaching to the nozzle.
What can I do about a printjob in progress that starts to warp/curl on the edges? I currently have a print job that is about 50% done, been running for 2 hours with 2 hours remaining. One side is curling/warping pretty bad, and I'm afraid there's no possible way this is going to finish without serious problems if I don't intervene. So what I'm doing is either brilliant or idiotic, I'm not sure which: I've paused the print job, stuck some elmers glue below the curling part (with toothpicks, careful not to budge anything else), added a couple degrees to the heatbed (for pliability hopefully), put a small book on top of it to smash it on the glue and let it rest for a little bit (I'll report back if this was a horrible idea or not). So my main quesiton: Is there any other techniques that you folks can recommend for a scenario like this? McGuyver'y techniques to repair your in-progress print jobs? Has anyone tried this technique I'm attempting and if so how successful was it? In case it matters, I have an ANET A8 and generally send my print jobs to Octoprint (Raspi) from Cura with a Octoprint plugin (Windows). Printing with PLA filament. I've done quite a few successful prints recently, but this is the first one that goes from corner-to-corner on the heatbed (this specifically). Printing at 207c with 60c heatbed (bumped up to 64 while glue settles). It's in a cooler room of the house, and doesn't have an enclosure so I'm afraid the cool temp is affecting it. Thanks Edit, last maybe significant (or maybe not) details: printing on glass with glue stick applied to it - been doing it for weeks and works quite well for the most part. Also, printing on a raft. Edit, here are a few pics. Both are from the back of the printer looking forward. I have one cam almost exactly level with the glass so I can see the hot-end extrude filament and another one slightly above it. Sorry for the bad lighting. Also included screenshots of my slicer settings for this print. Btw, I've since resumed printing after glueing it down, so far so good - but as you can see in the first pic it may have some possible structural defects and still has a slight curl:
Putting a large brim on it can help, I've seen people rescue prints with larger brims by literally adding weights to the brim around the areas that are peeling up. I've done it by taping brim edges down carefully as the print goes.
More glue to hold it down and lower in-fill percentage will reduce the warping. Or adding more cut-outs to the design like you have further up the shaft.
What defines the speed at which a motor is retracting and how can it be changed? Given a Marlin firmware and a line of G-code such as the following: G1 F100 X50 Y50 Z0 E-10 What defines the speed at which the stepper motor associated with the E-value is retracting? It is my understanding that the Feed Rate defines the speed of the movement (in this case 100mm/m) but I am not clear how I could accelerate a retraction? The reason I am asking is that I am not seeing a swift removal of material as i retract. Could the slow feed rate be the issue? I am using a pellet printer (WASP 3MT) and generating G-code from polylines on Silkworm.
You instruct the printer to move from a certain X-Y position instructed by the previous move, to X=50 and Y=50. While moving at a feedrate of 100 mm/min, it will also retract 10 mm of filament (if the previous extruder distance was 0) during that move. If the movement distance is large, the retraction is slow. If you started from X,Y = 49.99,49.99 it would be very fast. If you want a fast retraction, first move to a position, and than retract fast, so in separate commands. Do note that we usually do it the other way around: first retract fast and then move, this way there is less oozing of the nozzle. To sum up, in your G-code command, the speed of retraction depends on the path of travel (the length and speed defined by the feed rate F). If it is fast retraction you are after, you should split the command into two separate commands.
It seems like you are particularly talking about your extruder, please correct me if I have misread. In the command G1 F100 X50 Y50 Z0 E-10: G1 - move linearly F100 - Use a feed rate of 100 mm/minute X50 Y50 Z0 - tells those axes to move to (50, 50, 0) (absolute positioning) E-10 - tells the extruder to retract 10mm (relative positioning) If you are not experiencing high enough retraction speed: Try increasing the retraction speed in your slicer. Try increasing the max acceleration for the E axis using M201 (e.g. M201 E10000 sets to 10,000 mm/s). Try increasing the max feed rate for the E axis using M203 (e.g. M203 E25 sets to 25 mm/s). Notes: If you're able to connect to your printer over USB, and you aren't already using an interface to control it, something like Pronterface might make it easier to test retraction speeds. If this is the correct firmware for your printer, it looks like it is able to save changed values to the EEPROM. They may be using a modified Marlin firmware. If so, EEPROM can be saved with M500, loaded with M501, and reset with M502. If your firmware limits what you can set with M201 and M203, you may want to download the firmware and try to edit its maximum accel/feedrate before flashing.
How can I determine whether my borosilicate glass is fake or not? Is there actually a way to determine whether the sheet of glass that one has received from a Chinese supplier on eBay is actually a sheet of Borosilicate glass, and not just a piece of normal glass intended for glazing purposes? Are there any laminations, or markings/features, visible under a certain light or if the sheet is viewed at a particular angle? I ask because most of the 200 x 300 mm sheets on eBay are priced at around £16 - £26, but there are a few priced at just £10, which still claim to be toughened glass, although they do not mention the word Borosilicate, such as this one, 300x200x3mm 3D Printer Heated Bed Toughened Glass Clear Build Plate UK STOCK. To paraphrase, "Usually if things seem to good to be true, then they are probably fake." However, I wondered whether it was worth a punt as it is only £10, and then if I would be able to verify its veracity once it was in my hands.
Google turned up this thread: http://arstechnica.com/civis/viewtopic.php?t=96214 There don't seem to be any easy, definitive tests. You have two main options, both of which seem to require a fairly high level of laboratory skill (but nothing fancy in terms of equipment). The refractive index of Borosilicate glass is very similar to vegetable oil (implying no reflection from a boundary if you have a pool of oil on the glass). The density of the glass can be measured too, using a water bath. This should also be able to give a fairly accurate answer. Stress testing seems the most reliable way to determine if you have a 'good' sample, but may be more expensive.
"toughened" glass, Lead, cadmium and barium would have that effect. It goes without saying that you wouldn't want that kind of glass anywhere near your neighborhood. Even those items that mention borosilicate glass in the product description may have never seen any boron. Some studies have found that more than 90% of items sourced from China are fraudulently labeled. There is no enforcement. While stats say nothing about any particular sample, I'd certainly keep them in mind.
What stepper motor to use in heated chamber I want to build a 3D printer with a heating chamber of around 90 °C with build area 200x200x200 mm. I have never build a CoreXY system, so my design is currently an XY system with moving X motor (mounted on Y). Since it has a heating chamber I can't use normal stepper motor (there's a way, but I have to provide forced air cooling like NASA did, or water cooling). Extruder is Bowden type. I have already sourced almost all components, but I'm stuck at choosing the motor. I could find high temperature stepper motor in India (that's where I'm from), but it cost too much. I found one at the Visionminer website, they're the dealers for Intamsys printers, which has a chamber of 90 °C and they are providing replacement stepper motors as well. Comparing the cost, the motor I found in India costs three times as above. Even with shipping I will save a lot. But one issue is they're not providing any details about torque and current rating. There's one image in the website and it says, MOONS STEPPING MOTOR TYPE 17HDB001-11N 60904162 18/04/12 I thought it might be a MOONS motor, so I contacted them, no reply so far. I tried to find the motor by part number, but failed. I tried mailing Visionminer as well. Anyone have any idea which motor is this? or know any high temperature motors? Also they use Gates belts, which is rated for 85 °C. How reliable will it be in 90 °C chamber?
"Since it has a heating chamber I can't use normal stepper motor" Sure you can, the interior doesn't get all that warm unless you really seal it up tight, and that's not really needed. I have an enclosure around my 200x200x200 mm MIGBOT (early Prusa clone with direct drive extruder), printing PLA with 60 °C bed, the interior only gets a few degrees warmer. The motors can take a lot more heat than you think they can. I have a couple pictures in this thread. The front & back panels are 18x24 inch polycarbonate from Home Depot, I 3D printed the corner brackets, and added a couple of pieces of wood for some stiffness. The entire front hinges up. The top is 24x24 inch, and the back 6" hinges up to access the SD card that is on the display/control panel.
You don't need to worry about the stepper for heating chamber since the direct drive uses a fan for cooling the motor area. When I started to make my own printer I had the same question but in order to make me feel good and peace. I prefer to use a bowder extruder. This bowden extruder comes in different sizes: Normal as picture shows above, small, and mini like the other that shows pre assembled below. However the question should be different like, Can I print inside an oven? for this will address the question to other possibilities: 1.- Cover or shield the motor with some foil to avoid the heating 2.- Add a water cooling like CPU, so the water flows from outside to the motor to keep a low temperature. 3.- Add cooling fans, this ones should take te air from outside and tha air can be directed with a corrugated tube for the Extruder motor and the radiator. For the X, Y and Z motors can be a rigid tube. This cooling fans won't affect the internal chamber temperature due the cooling process is punctual. 4.- Many electronics components are designed to work at 105°, so won't be affected in short terms, however the life of circutry will decrease a lot, since designs cover until 5 years at normal conditions so your printer can last up to 1.5 years. Recommendations: I don't see a real reason to keep the printer isolated to high temperatures while the porpuse of this is to keep temperature variations from clime like winter and summer. In my case the print room has a normal temperature of 38°C on summers and -2°C on winter, so how can I print with the same quality on winter if the printer is so cold? ah, I need a chamber to keep that temperature of summer. then I made the chamber to acheive 38°C not the whole temperature of the bed print. If I need to print ABS so I set the bed temperature to 80°C so the parts won´t get warped, due the temperature for adhesive for ABS is the correct; also this temperature won´t over heat the chamber at least near to 60°C, but can it be reduced extracting the heat with other fan. For this case is just only one or two fans. Note: The Idea to have the whole printer inside an oven will help to keep that 80°-95°C under control is good, but some times is hard to implement it due materials and its purposes are different and serviseable life will be too low. So your chamber should include the printing area only or follow the recommendation as the picture above. Those photo was taken from the site industrial RepRap and also exposes som e features as i'm suggesting.
Bad PEI plate adhesion to heated bed I'm (surprisingly) having a problem getting my PEI substrate to stick to the heated bed surface. The ABS item being printed stuck great to the PEI surface. I've not seen anyone else post on this problem, but perhaps someone has seen it. Note, this is the PRINT SURFACE warping, not the item being printed. That part went great with no delamination even on a large flat bottom. I'm using 3M 468 adhesive to attach a PEI plastic sheet directly to the heated surface of my Flashforge Creator Pro printer. The heated bed is anodized black and the adhesive is a sheet cut to the requisite 6" X 9". It seemed fine and looked pretty good with few air bubbles when first attached, but the first print caused the edges of the PEI plastic to warp up from the corners of the bed. It appears that the weak link is that the PEI detached from the adhesive. While this was not uniform, removal of the PEI sheet showed that 95% of the adhesive was still attached to the print bed. The PEI has one glossy side and one matte side, and I chose to apply the adhesive to the matte side to get the glossy surface. Does this make a difference? I applied the adhesive to the PEI surface just as it was after removing the protective plastic coating. I figured this was a clean as it was ever going to get. I'd definitely appreciate some ideas on this.
Perhaps this is due to the aluminum build plate warping/crowning as it heats? I use a PEI print surface on my Monoprice Maker Select (Wanhao Di3), which has an 8.5"x8.5" build plate, and I've had no issues with the PEI coming unstuck. However, instead of attaching mine directly to the aluminum heated build plate, I attached it (using 3M 468) to a piece of Borosilicate glass, which is then attached to my aluminum build plate using silicone thermal heatsink pads. The rational for using a Borosilicate glass print surface is that a heated metal build surface is going to warp or move some, as the heat is coming from one side; the glass will not warp and so you have a completely flat surface on which to print (and the heatsink pads help make up the difference in surface geometry). I used this 3dprinterwiki article as the basis for my glass bed mod, and applied the PEI/adhesive using another source (which I've forgotten, possibly the RepRap wiki). While the first link is Wanhao Di3 specific, it covers the idea. Make sure you tweak your z-endstops if needed on your particular machine, as adding the glass plat raises the bed height by several mm. Please note that I'm theorizing - I did not ever try applying my PEI directly to my aluminum build plate, so I have no direct comparison. But the use of glass to eliminate build plate warping/crowning is fairly well documented. See also this 3DPSE Answer.
I never used one, but the thing that makes two parts stick together is the atmospheric pressure. The glue is just a thing that, even if it does not make any chemical bond, is keeping the air out. A pure example of that is if you put water in a glass and then put another one on it. Although the water is not glue you can't easily pull it up and if the surface is big you can't at all. It is the atmospheric pressure that keeps them together and the water keeps the air out ;). I suggest try the other side and see I'm just guessing . Good luck!!!
My Hatchbox PLA filament only allows me to use rafts When I first got my 3D printer (a FlashForge Adventurer 3), it came with a sample pack of filament. With this filament, I was able to use skirts for my first layer. When the sample filament ran out, I switched to Hatchbox PLA filament. For some reason, I cannot use skirts with the Hatchbox filament. Now, whenever I try to print something with a skirt, the print moves around, ruining it. The only first-layer that works now are rafts, which I do not like, as they use up more filament and are more of a pain to remove. Is anyone else having this problem? If so, what are some workarounds to this issue? Here are the failed prints. I terminated them mid-way, as they started to shift on the build plate.
It is unlikely this is a filament material issue since many of us have used Hatchbox PLA without this issue. This is a first layer adhesion issue. Your bed may not be clean, or the bed may have had an adhesion layer you cleaned off. Your nozzle may be too high on the first layer. Your bed my not be level. Glue sticks can help adhesion. Glue sticks usually don't post the composition on the packaging, but Elmer's glue sticks work. Elmer's washable makes it easy to remove the old layer before adding a new one. There are also glue sticks specified for 3D-printing. Here's a discussion on glue sticks: Are all glue sticks PVA-based? How to find out?
Please try one (or more) of the following: Change the temperature of your heated bed (50 - 60 °C) Check your nozzle height (0.15 - 0.25 mm is what I use) Make sure your bed is level (Maybe use a leaving stick (or whatever those bubble things are called)) Use adhesion. Start with painter's tape (least damaging, easy to remove), then use a glue stick or even hair spray if necessary. Be sure the bed is clean (I recommend glass cleaner 1-3 squirts and then a light scrub of a paper towel)
How to switch motor outputs and use E1 as X, in Marlin firmware? I broke up my electronics and now the output for X is not working. The stepper is OK. There is any simple solution to remap the output pins? I want the E1 output to act as the X output.
When using Marlin firmware you could easily change the pin layout of the extra extruder (E1) with the broken X stepper pins by changing the pins_RAMPS.h file. Download the firmware and open the firmware project in Arduino IDE. Navigate to the "Steppers" section of the pins_RAMPS.h file and replace: #define X_STEP_PIN 54 #define X_DIR_PIN 55 #define X_ENABLE_PIN 38 #define X_CS_PIN 53 for: #define X_STEP_PIN 36 #define X_DIR_PIN 34 #define X_ENABLE_PIN 30 #define X_CS_PIN 44 and also change: #define E1_STEP_PIN 36 #define E1_DIR_PIN 34 #define E1_ENABLE_PIN 30 #define E1_CS_PIN 44 to: #define E1_STEP_PIN 54 #define E1_DIR_PIN 55 #define E1_ENABLE_PIN 38 #define E1_CS_PIN 53 When the file is changed an saved, build and upload code to your board and plug the connector of the X stepper into the E1 header.
you can find pin.h in marlin directory and in pin.h file change # define pinout of z motor and extruder motor
RF100 Settings after reinstalling firmware I have an RF100 v.1 printer with stock software (again) onboard. After my attempt of changing firmware, the X-axis motor decided to stray printhead away out of the borders of the printer and somehow motor wires broke out of the adapter. I fixed the wires and reverted to stock firmware, though now whenever I auto-home or press "Home" button on Ultimaker Cura, printhead goes wild top and left reaching borders and starts buzzing like crazy. current condition: Head is TOP LEFT with X=100 and Y=100 on the table being under head Z=0; screenshot Even with these beautiful tiny settings, I cannot print anything because printhead goes too far TOP RIGHT by itself. Ideas?
Everything was fixed by thorough googling and flipping the whole wire-set of the motor. It started to twirl in a right direction. Sorry for the time
Everything was fixed by thorough googling and flipping the whole wire-set of the motor. It started to twirl in a right direction. Sorry for the time
Heated Bed vs Heated enclosure for large scale printer I'm /relatively/ new to 3d printing (I'm getting pretty good prints from my Wanhao di3 plus, but haven't done any DIY kits or anything) and materials engineering is probably the furthest thing from my area of expertise so I thought I would pose this to more experienced makers: If I'm building a large scale printer (probably a similar size to substation33 - sub33D's 1200x1200 printer) that I'm only going to print in PLA and MAYBE ABS occasionally, can I substitute the heated bed for simply heating the entire enclosure? The idea would be to have a thermistor measuring the ambient air temp inside the enclosure with a heat gun or two to hear the entire enclosure as necessary. I want to do this to try and reduce the cost (significantly as far as I can tell) as it seems 400*400 silicone heat pads tend to go for about 80 bucks a pop. An alternate idea I had was to use two or three heat pads and space them evenly under a glass bed, although I feel that this won't work as well because the heating won't be homogenous... Any input is appreciated :)
I tend to agree with Davo, that you might want both. But I'd probably try the enclosure first. My printer is about 1200 by 400. I looked at options, and found some difficult trade-offs: First, the big heatbed approach: as you know, getting a single heatbed that big will be expensive (and if it ever breaks or fails, you have to replace it all). a big heatbed also draws far more power than a RAMPS board can switch, so you'll have to use the on-board control to control a power relay (solid state or mechanical). a big heatbed will also waste a great deal of power when you're doing small things. Second, the heated enclosure approach: more costly to heat up at the beginning, but if insulated it may be cheaper for long prints (with a bed that large, your prints might take really long, too) because it can retain the heat better. if you keep the air circulating inside, you'll get much more uniform heat, rather than hot first layers and cooler higher layers. consider the effect of the heat on all the other components: motors, electronics, pre-heating the fibers, thinning any lubricants,.... you could save some heating cost and time by providing a way to shrink the space to be heated: perhaps just a partition you can insert when doing smaller prints. It wouldn't have to be nearly airtight to make a big difference. A third option is several small heatbeds: this lets you turn on just the ones you need for any given print still expensive, but cheaper than one big one, especially when any of them fails. there will be uneven heat at the seams, but if you place the boards tight together I doubt it's enough to matter. You could also carefully trim the boards' edges to get the spacing closer to uniform. temperature regulation will be tricky. If you want to support a temperature sensor for each board, you'll have to start hacking at the control software, because (as far as I can tell) there's no provision for multiple heaters. Probably easier to create a completely separate unit with a big power supply, and a simple thermostat for each of the beds, that you just set manually before starting a print. Overall, I think the heated enclosure may be best. I like the hair dryer idea or heat gun idea, of course you'll need a little extra circuitry to switch it, but you can use the usual software for temperature regulation by installing thermistor or thermocouple in the case (just be careful about circulating the air well enough to avoid "hot spots"). You may want to add an extra cutoff to protect against runaway heating -- I had that happen once when a thermistor literally fell out of the mounting hole in the hotend. Something like https://www.grainger.com/product/SUPCO-Thermostat-407L11. Let us know how it turns out.
If it is yours first DIY 3D printer try building smaller version first with just one silicone heat pad. 400^2 mm^2 is good enough. From my experience: I used 500W silicon heater with SSR (AC mains) and it heats as fast as hotend (on DC 12V). I also recommend tooling plate (CNC machined Aluminum plate), it is better for heat transmission (Al) and it is more flat (machined to +-50um). You might consider thermal insulation under the silicone heater and also some foil on build plate itself - to minimize thermal losses. Cork might be useful for bottom. For top I use blue foil (in which it was originally wrapped). Enclosure I think it is also necessary, combined with heater, but then you must also have tubes for cooling of printed part, motors, fridge on hot end, etc. Word of caution: You must realize that you will be probably using 1.0mm nozzle or even more, since on 1200x1200mm you will just wait for weeks for part to complete. For reference: I print on 340mm diameter (delta) with 0.3mm nozzle and works just fine. But if I use just 0.1mm instead of 0.2mm layers it takes more time, but at the end it looks fantastic. So in your case even layer height must be big to reduce printing time. So you can increase speed, but then there could emerge some artifacts on printed object and also consider heater on hot end, which might be lagging behind. There are a lot of problems when you go this big and all are not even closely to be considered solved, but if you have time..
How to prevent bend (or warping) with M3D printer? Hello is there a way to prevent bend on print with M3D printer?
You can to print a brim, a thin layer on the bottom connected to the model. This will help hold it in place. Since it is thin (one or two layers) it will not warp itself. The brim is not the same thing as a raft. A raft is under the model. The brim is on the same layer as the models bottom layer but outside the model. It looks something like this: I assume that you use a heated bed if you have one? Also, it is imperative that you get a good first layer. Calibrate your machine carefully.
Try using an adhesive before you print. This could be http://airwolf3d.com/shop/wolfbite-prevents-3d-printed-parts-from-warping ... What is the temperature of the room you are printing in like?
Nozzle moves very slowly from home to start position I´m currently working on a custom delta printer running Marlin 1.0.2 firmware. To control it I´m using the Repetier Host software with the onboard Cura software to slice my objects. The printing works fine but it takes the nozzle nearly 5 minutes to move from the homing position at 250 mm to the starting position of the print near z=0. I already tried to increase the travel speed but nothing changed. Can you guys please tell me how to increase the speed of my nozzle to move down to my starting point of the print?
I had the same problem, and I solved it by changing the Z-Axis Feed Rate to a much higher value (1000 mm/min.) in Repetier Host via Config -> Printer Settings -> Printer.
Try changing the travel speed in you r slicer, if that doesn't work , try changing out the motors for newer ones.
How to debond cyanoacrylate glue from pla I have used cyanoacrylate glue aka superglue to bond PLA. I have created several electronics enclosures. (Definitely the most time-consuming part of the project.) Now my question is which debonder/solvent can I use to separate the pieces again without destroying the PLA parts? Wikipedia proposes the following: nitromethane dimethyl sulfoxide, methylene chloride, gamma-Butyrolactone.
Acetone Acetone will dissolve cynoacrylate (superglue) and should weaken it enough to be able to separate the parts. A readily available cheap source of acetone is nail varnish remover (just make sure you don't buy the acetone free version!). Give the pieces a soak in nail varnish remover for 10-20 minutes and they should come apart with some prying. Acetone does not dissolve PLA, so the PLA parts should be undamaged. If you were to try this on ABS parts however, they would begin to dissolve.
Gamma butyrolactone is by far the best product to remove cyanoacrylate and also great to dissolve PLA.
Are the Makerbot 5th generation's initial issues resolved I am considering buying a 3D printer for work (scientific research). On paper, the Makerbot 5th generation seems to be the best option, because the price is right for my budget, and also because I'm generally pretty busy, so I want something that's as close to "plug and play" as reasonably possible. This will be my first 3D printer and I have no prior experience with the technology. However, any search for "Makerbot" brings up a raft of negative reviews from around the time of the 5th generation's first release, which mostly focus on issues with the smart extruder constantly failing and needing to be replaced under warranty. Many of these reviews point out that this may be an initial "teething" issue which might be fixed in later versions of the model, but now, one year later, I'm unable to find any information about whether this was indeed the case. So: are the initial issues with the Makerbot 5th generation's smart extruder generally considered resolved, or is it an underlying flaw of the model that won't be fixed until the next generation? I imagine that people might want to suggest alternative models in the same price range. That would be welcome, but I have an extra constraint, which is that I can only buy models that are available in Japan without a long delivery time - this probably limits my options quite a bit.
Makerbot claims the problems are fixed. I have heard from a number of resellers that the problems are fixed. Unfortunately, both of those are somewhat biased sources. It's surprisingly hard to get good info on the subject -- very few credible people are talking about recent experiences with the product line. Issue #1: The main surviving user forum (https://groups.google.com/forum/#!forum/makerbot-users) has had the torrent of people reporting 5th gen issues more or less dry up. There ARE still people posting issues, but at a fairly low level that is not far outside what you would expect for an average hobbyist printer. What we DON'T know is whether the flood of complaints has slowed because they're working better, or because very few people are actually buying them any more. Issue #2: The power-users and community leaders that typically evaluate and review 3d printers are all avoiding the 5th gen line like the plague. Makerbot burned up a lot of community good-will by going closed source with the Replicator 2, and lost more good-will through a series of misunderstandings over patent applications and the Thingiverse terms of service (Takerbot scandals), and put the nail in the coffin by knowingly releasing a non-functional 5th gen product line. Very few credible people are willing to give them a chance at this point, so there is a severe shortage of unbiased reviews. Issue #3: Makerbot has a proven history of buying off journalists and reviewers to get positive 5th gen press. Some verifiable examples: Hundreds of fake 5-star Amazon reviews from paid review accounts. A relevant analysis: http://www.amazon.com/review/R2JI8LRRXZYNX1/ (Not verified, but related: a widely-believed rumor states that the German Amazon site actually wiped all of Makerbot's 5th Gen reviews because of blatant tampering: http://www.amazon.de/MakerBot-MP05825-Replicator-5th-H%C3%B6he/dp/B0183TP806/) The CES awards initially given to the 5th Gen line were given before Makerbot had functional firmware. None of the 5th Gens at the CES launch were functional. All demo prints shown were made on Replicator 2s. How could a non-functional product win awards? Not hard to figure that one out. Historical positive press in the 3DP journalism media (3ders, Make Magazine, etc) has been directly proportional to the volume of ad-buys from Makerbot or the parent company Stratasys. I could keep going, but you get the idea. So it's hard to evaluate the reality of the situation. But even IF the Smart Extruder issues are truly all fixed (jams, leveling issues, thermocouple connection, filament encoder failures, etc), there are still meaningful problems with the product line. The print quality is not impressive. The motion mechanics, which have not been appreciably changed to my knowledge, are not great. The Smart Extruder is basically a giant pendulum wobbling off the side of a non-optimal gantry selection of an H-bot architecture. Makerbot has addressed the floppy construction via firmware tweaks to significantly slow down the machine to give marginally-acceptable print quality. According to most reports I have seen, a Replicator 5th Gen will print significantly slower than a Replicator 2, for example. Expect in the neighborhood of perhaps 30-40% longer print times than comparable printers. The price tag is roughly double or even triple the current market price for the size and print quality output of the machine. There are so many great printers on the market now for significantly less money that it's kind of nuts to drop the cash on a Makerbot. It is marketed as a PLA-only machine. That's fine if you're printing art and trinkets, but it's not a great option for mechanical parts. While you CAN print other materials, this is not technically supported. The support plan structure has quite frankly become abusive. Makerbot used to have really helpful phone tech support, but the crushingly massive volume of 5th gen troubleshooting requests forced them into a paid tech support model. Around the same time, Makerbot shut down their user community forum and deleted links to external technical resources off their website. So unless you know the right places to go, support is scarce. For official tech support you must buy "Makercare" or pay for each help ticket. This is completely out of line with industry norms for a hobbyist/consumer 3d printer. They essentially took their biggest liability -- unreliable printers -- and tried to twist it into a profit center. I personally think this is a significant reason to avoid the company entirely. Is it possible to get good results from a Replicator 5th Gen? Sure. But it's a poor value for the cost, from a company that has spent the last few years systematically driving away its former loyal user base. I would recommend staying away until there's some significant change in the product line at minimum.
I own a 5th gen printer and after teaching myself how to dismantle and clear the smart extruder of jammed filament I can attest to the machine working better than any other 3D printer I have used. I run the 3D printing section at Tampa Hakerspace and the MakerBot is much more reliable than the prusa or wanhao which we use at the space. I had problems with the machine initially but now it works like a champ
Under extrusion when starting outer wall See the pictures below. I have a severe under extrusion when the printer starts the outer wall, which is resolved by the time it finishes the outer wall. It starts the layer in the same place every time, so it results in this vertical line, on one side of which is fine (where it finishes the layer) and the other side has bad gaps and the wall is much thinner. In this picture, the problem is on the outer wall (see red outer line, the print head is moving counter clockwise. And here is the print showing the issue. Just fine on one side, but terrible on the other, precisely where it starts the layer. Strangely, this only occurs on the layers with infill. The top layers seem fine (despite starting in the same place). I have disabled retraction with no effect. Here you can see it start at the tip and get gradually better as it progresses. Here you can see that by the time it finishes, it's just fine, and also what a contrast is between the start and end. That should be a flush edge, there. Another view Is there some setting that I should be tweaking? I've exhausted my own ideas of different tweaks to no avail. The printer is a Monoprice Maker Select V2. I'm using Ultimaker Cura 4.1.0. Material: PLA Layer height: 0.24 mm (in the pictures, but replicated with 0.16 mm also) Temp: 205 °C, here (but tweaking this hasn't had any effect) Retraction: Disabled This doesn't happen on layer changes. Although it does line up with the Z seam, you can see from the G-code visualization below that the outer wall is the very last thing it does. The issue occurs at the start of the outer wall (still on the same layer) that it has the issue, but by the time it ends the wall (just before switching layers) it is fine. I've tweaked a few other settings, one by one, and seeing if any have any effect. So far, not really: Outer Wall Wipe: 0.2 (default), 0.8, 0.0 Jerk Control - Wall Jerk max velocity change, 5 mm/s Outer Wall before inner (Yes instead of No) Wall Line Count (3 instead of 2): This improves it some, but I suspect just by making it a little more difficult to see What have I done since... I reset all settings in Cura to a default "Draft" setting and then set layer height to 0.24 mm, and turned off Infill. Then I have tried prints with different settings for "Retract Before Outer Wall" and printing temperature. Here are those results: These pics seem to suggest a very clear lag in extrusion. 1 and 2 are different temps. 2 and 3 are different retraction. 205 °C, 0 % Infill, Retract Before Outer Wall: Off 195 °C, 0 % Infill, Retract Before Outer Wall: Off 195 °C, 0 % Infill, Retract Before Outer Wall: On Note: the retraction setting resulted in a noticeable pause before printing the outer wall. Retraction distance is 6.5 mm, and this is not a Bowden fed device. There doesn't seem to be anything strange about the G-code, either. Here are the G0 travels just before the outer wall followed by the wall. ... G0 F7200 X106.319 Y93.413 G0 X106.26 Y93.909 G0 X107.213 Y93.658 G0 X107.8 Y92.542 G0 X107.286 Y90.844 G0 X107.509 Y90.394 ; (outer wall of outside) G1 F1328 X107.985 Y90.707 E116.98713 G1 X108.38 Y91.128 E117.01098 G1 X108.658 Y91.623 E117.03444 G1 X108.813 Y92.18 E117.05833 G1 X108.832 Y92.751 E117.08193 G1 X108.713 Y93.315 E117.10575 G1 X108.463 Y93.837 E117.12966 ; (first curve complete, on to straightaway) G1 X99.631 Y107.716 E117.80936 G1 X98.912 Y108.59 E117.85612 ... Even more done... These are retraction off, and 205 °C If I set the "print speed" in Cura to 20 mm/s (normally 60 mm/s), the outer wall speed is reduced from 30 mm/s to 10 mm/s. The result is quite good. If I leave the "print speed" at 60 mm/s and adjust only the outer wall to 10 mm/s, it's still quite good. So it seems like an acceleration thing. If I can figure out how to get it to slow down in just the right spots or compensate in some way, then perhaps I can make this problem go away with minimal sacrifice in total speed.
Based on other comments, answers, and question edits so far, in addition to your original question, I believe there are possibly two things going on here: incorrect retraction settings, including a misunderstanding of which settings are relevant and what they do, and issues related to slow acceleration. Both relate to misdepositing/loss of material. First, some basics. When the filament is advanced to the point needed to extrude material and print at the intended volumetric rate, it's under significant pressure, compressed between the extruder gear and the nozzle. My understanding is that your printer has a direct drive extruder, not a bowden, so there's far less compression than with a bowden setup but it's still there. This means that, if you try to stop extruding, it's material will continue to come out of the nozzle, just at a decreasing rate, until the pressure dissipates. This effect is reduced but still present if the nozzle is held-against/moving-over already printed material, and heavy if moving over empty space, even moreso if moving across sparsely-filled space like infill where it will bond with the already-deposited material and get "stretched"/"pulled" out. The idea of retraction is to pull the filament back when the print head is moving to a new location without trying to deposit material, to relieve this pressure and prevent unwanted misdepositing/loss of material, and to reverse the process, putting the filament back exactly where it was when the last printed line ended, the next time it starts trying to deposit material. The relevant options in Cura are: Enable Retraction - must be on Retraction Distance - should be at least 5-6 mm for bowden setups, probably more like 0.5-2 mm for direct drive. Retraction Minimum Travel - should be 0 Combing Mode - try different settings. Off is probably the best relative to your issues, but hurts your print time a lot for certain models, and can hurt quality in other ways. Everything else related to retraction is fairly irrelevant, especially "Retract at Layer Change" is a niche option and not typically useful. As I understand it, turning just "Retract at Layer Change" on does not mean retraction is on. Now, your other issue may be acceleration. Extrusion works best as acceleration speed approaches infinity, because the extrusion rate and pressure needed to extrude will be fairly constant for the entire line/curve. If acceleration is very slow, pressure will be wrong during the start and end of lines. It's likely that, due to high pressure, excess material will get deposited at the end of one line while slowing down, then after moving to start the next line, even if you retract the filament, you'll have insufficient pressure at the nozzle after reversing the retraction to start the next line. A jerk limit of 5 mm/s is really low. I'm used to more like 20-30 mm/s. You don't say what your acceleration limit is, but it's probably also low. Slow acceleration has minimal impact if your max speed is slow, because you quickly reach the max speed and most of the print speed (and thus extrusion rate) is steady. But if you want to print at high speeds, you need high acceleration. Try and see if you can increase it. Or accept printing at slow speeds. Another option, if you're open to hacking on your printer, is replacing the stock firmware with a recent version of Marlin with the Linear Advance feature. It does the math to model the filament pressure as a spring, with a spring constant you can tweak, so that it can compensate for varying print speed and end lines with approximately no pressure remaining.
try retract before outer wall off.
FlashForge Finder turns off mid print and is unable to reset or update firmware I am new to 3D printing and am trying to get started with a FlashForge Finder. Unfortunately, the printer keeps shutting off mid print, and I am unable to find a cause. Here's what I have tried thus far:  System Settings Factory Reset - When I attempt a factory reset, nothing happens (the menu closes, but usage stats & wifi do not get reset)  Firmware Update - Updating the firmware from the touchscreen immediately fails every time Printing I can start a new print, but the machine turns off after about ~10-15 minutes at temperature The transformer on the power cable remains lit green even after the machine fails and turns off I have tried printing multiple files from Thingiverse with no edits (an elephant and an owl) Temperature is set to 220 °C Observations One of the times that the print failed, I noticed that the temperature had dropped to ~204 °C prior to failing. I don't know if that's normal behavior or not
Same problem occurred to me, now twice. How much is the Finder's memory able/allowed to register? After erasing all but the print job, things go smooth.
Try unplugging the USB lead as soon as the printer starts to print (I had similar issue and this worked for me).