Datasets:
title_body
stringlengths 20
149
| upvoted_answer
stringlengths 32
13.1k
|
---|---|
"Tac Tac" sound when printing with ABS | @Ecnerwal is right: that noise you hear is the extruder not being able to push the filament, and the stepper can't push any harder. When the extruder tries to push harder than it can, it gives up, and the "spring" tension it created in the filament forces it to go backwards a tiny bit. Then it tries again.
Possible causes/fixes:
Temperature too low -- this makes the filament not liquid enough to push through the nozzle easily. For ABS, you should be in the 230-240 range.
Clogged nozzle -- take the nozzle off (while hot) and try heating it with a torch to burn out anything that might be in there.
Bad filament -- If the filament has contamination in it, or is too large to fit through the hot end in places (I.E. it gets up to 1.9mm instead of 1.75mm)
Stepper current too low -- I'm not sure if you can adjust the current that is sent to your stepper motors, but if it is too low, the stepper can not provide enough torque to push the filament through. I don't see the stepper drivers on the site, so I don't know if you can adjust them or not. |
Strange stringing-like activity when printing face shields | This isn't stringing. It's failure to adhere to the previous layer, so tension in the newly extruded material just makes it form a straight line until it bonds to something again. Are you used to printing PETG? You didn't mention fan settings, but having any cooling fan at all will cause failure of layers to bond. Your print speed is borderline for PETG, especially if you're using thick layers like 0.3 mm which you also didn't mention but looks like might be the case. |
Can I use a 12 V single channel optocoupler module in a 24 V machine? | You can safely use the module with 24V.
The input side shows a red LED, optocoupler and 1k resistor in series. The LED and optocoupler probably have a voltage drop in the neighbourhood of 3.1-3.5 V put together, so for a 12 V input you will get a current of approximately 9 mA-.
For a 24 V input voltage the increased current will cause a slightly higher voltage drop, but even if the voltage drop remains as low as 3.1 V the current will still only be 21 mA. This is well within the rating of the optocoupler (similar optocouplers are often rated for 60 mA) and slightly pushing the rating of the LED (similar LEDs are usually rated for 20 mA) but it will probably be fine.
For extra peace of mind you could connect an additional resistor in series with the input. The "ideal" value (that is, to keep the current identical to that at 12 V) would be 1.3 kΩ, though any small value resistor (above 100 Ω) would be fine. |
Materials testing of 3D printed parts | As with all AM related simulation existing FEM codes are not well suited to solving the material addition issue.
It is very hard for the codes you mentioned to simulate layer-by-layer and toolpath-by-toolpath material addition.
There has been significant research in AM simulation, most of it focuses on the metal additive manufacturing processes. The simulation of the entire build process with complex geometry and material addition almost never occurs due to computational cost. For example this work is one of the most sophiticated simulations developed and it is so computationally expensive that only a single laser scan was performed.
If what you actually want to achieve is stress/deformation of a FDM part then perform a structual analysis using anisotropic material properties that match that of your printed parts. I would assume that by now data for material properties of FDM parts with respect to machine parameters would be known at least for commonly implemented materials such as ABS and PC. |
Attach handle on a loose screw | You are facing 2 problems in this case:
Material choice
Design
Material choice
As far as I can see, this is an oven that can heat its interior to roundabout 250 °C, which means that the door will radiate a lot of heat. This needs special high-temperature plastics to withstand - PLA and ABS are not an option for this!
I would suggest looking for PEEK (stable at ~260°C) or a high-temperature resin print in this case, as there are some resins which can withstand much higher temperatures after curing. Note that PEEK is traded usually for above 100 €/kg and needs a specialty printer. Even the less stable PEI filament (works under 170 °C) needs a specialty printer: All metal hotend and 350-380 °C extruder are needed!
Design
The plastic screws you show are self-cutting plastic screws. if their head is accessible, they get screwed into holes that have a bore large enough for the center.
If the screw head isn't accessible, for turning but can be pressed down, then one could opt to use high-temperature resin to cure inside oversized holes over the screw. This is not an option if the screw can't be pressed into the part though, as then the handle always will rattle.
Repair?!
Note that you might be able to access the screws with a little work, if you have the right tools and manual. |
How should I test filament samples? | Select an STL that exhibits characteristics you want to test (bridging, resistance to various pressures) and print it with each material. Preferably, run the same exact gcode, only changing temperatures, so that you get an apples-to-apples comparison. |
Does a teflon insert in an E3D V6 clone limit it's useable temperature so it can't print nylon? | It depends how deep it goes. If the teflon goes into the hotend then yes, it will limit the temperature end.
But: the teflon may just be something that ends somewhere in the cold side and sticks out so there is something to put into the extruder or higher up into the connector.
I cam currently setting up a Slice Mosquito for a Bondtech DDX. The Mosquito is full metal, but there is a (actually printed nylon) adapter for the DDS. In this adapter you put a teflon/capricorn tube, that ALSO sticks out only around 5 mm. Here is the point though: it never goes into the even cold side and is only there so the connection to the extruder on top has a width limitation.
So, it really depends how long this tube is (and no-one here will know because a v6 clone may be different internally from the original). I would suggest pulling it out (it should move out easily) and then seeing how deep it goes. As long as it stays on the cold side before the heatbreak, it never gets in touch with anything that is hot. |
OpenSCAD 2018 Command line | You can specify variable values from command line using:
openscad ...\
... \
[ -D var=val [..] ] \
... \
... \
filename
See the OpenSCAD Manual. |
G28 ignoring Z-Probe X/Y Offsets | You need to enable the constant Z_SAFE_HOMING (like: #define Z_SAFE_HOMING) in your printer configuration file (if you're using Marlin firmware that is). This will move the nozzle to the middle of the plate prior to lowering the nozzle by default:
#if ENABLED(Z_SAFE_HOMING)
#define Z_SAFE_HOMING_X_POINT ((X_BED_SIZE) / 2) // X point for Z homing when homing all axes (G28).
#define Z_SAFE_HOMING_Y_POINT ((Y_BED_SIZE) / 2) // Y point for Z homing when homing all axes (G28).
#endif |
Anet A8 Z-Probe Bed Positioning going outside of bounds | If you have done major revisions to the printhead by changing to another carriage, it could be that the nozzle in the carriage is not at the exact position as in the previous carriage.
So, have you made sure that the new carriage has the appropriate values so that the nozzle is at (0, 0) when commanded to that position? Alternatively, in different wordings, this translates to: "Is the printer printing in the center?/Are the prints centered on the build plate?". See e.g. "How to center my prints on the build platform? (Re-calibrate homing offset)".
From the Thingiverse link van be read that this carriage requires an offset. The carriage design mentions a shift of the X axis:
#define X_MIN_POS -10
Is that working for you?
General remarks regarding the firmware update:
Have you also loaded all values from firmware using M502 and then M500? (if M500 commands are enabled). Also, Skynet3D is obsolete!, use Marlin instead! |
What are the pros and cons of collecting parts yourself, versus getting a DIY kit and then modifying it? | From a general point of view, there are a few things to consider.
If you buy a kit:
Pros:
You get some insurance that you have all the parts that you need to get a functional printer - all the electronics, structure, bolts, nuts, screws, washers, wires and so on.
Most likely, all the parts you get are made to fit together.
You will (usually) get a manual, often a community that can help you out, and sometimes even technical support.
Sometimes, it can be cheaper than buying each part separately (but it can also be more expensive)
Cons:
You have limited/no options to customize your printer to your own preferences without purchasing additional parts.
Some kits can be difficult to upgrade later or may be locked to some configuration or software.
My opinion:
The way I look at it, the better option for you depends on how you want to spend your time. That is:
If you get a kit, you can spend more time building.
If you collect all the parts yourself, you will have to spend time planning, ordering parts (possibly multiple times) in addition to actually building the printer. A possible lack of manuals could also increase the building difficulty.
If you don't already own a 3D printer, I would recommend getting a kit, simply because struggling with trivial things like parts not fitting together can take away the fun for many people. |
Why is my print getting messed up mid-print? | For me, this looks like a cooling problem. Try to print this part 2-4 times in one print and get sure your fan is cooling all the time.
It is a known problem, if the layers get smaller, that they do not have time to cool down. So you're printing on a wobbling bunch of still soft layers, which result in what you showed on the photo.
In my opinion, this has nothing to do with the filamen. |
Ender 3 Won’t Start - flashing screen and blue LED on main board | Just after I posted this I tried swapping various components with my other Ender 3.
It turns out a faulty power supply will produce this effect. When I swapped out the power supply it started normally. |
Calculating gear ratio for leadscrew | Your lead screw nut advances 8 mm every complete rotation of 200 full steps, so a single full step would change the height by 8/200 = 0.04 mm. This implies you need to set the layer height a multiple of 0.04 mm.
The gear ratio of 4:1 (a reduction) is required to get to your desired 0.01 mm layer height, but that is a very uncommon and a too small layer height to use. |
Marlin: possible to set ZMin limit when ZProbing? | You have not stated the version of Marlin you are using. I will assume we are discussing the latest Marlin 1.1 RC8.
There is no longer any Z_PROBE_DEPLOY_HEIGHT but in earlier Marlin versions it did not function as you think; it was used to signify the amount of Z travel to execute prior to Z probe deployement. This is to ensure available space for servo-mounted, sled, or other types of "stowed" probes. This was not a limit to the Z travel, but the amount of Z travel to always execute before begining homing. If a printer without Z_MAX_ENDSTOP is left at maximum travel position, using non-zero Z_PROBE_DEPLOY_HEIGHT could crash a bot into the Z maximum end position.
In these previous versions of Marlin, the homing height was computed by a combination of the above constant adding and subtracting to/from some others, which has since been replaced by the simpler and singular Z_HOMEING_HEIGHT (which works indpendant from the new Z_CLEARANCE_DEPLOY_PROBE):
//#define Z_HOMING_HEIGHT 4 // (in mm) Minimal z height before homing
(G28) for Z clearance above the bed, clamps, ...
#define Z_CLEARANCE_DEPLOY_PROBE 10 // Z Clearance for Deploy/Stow
The absolute value of the maximum bed positions are used in Marlin as sanity check to ensure no axis moves greater than this amount in one movement in either positive or negative direction:
#define Z_MAX_POS 200
If your probe does not function correctly, no ficticious Z_MIN_WHILE_PROBING will prevent a faulty probe from causing a head crash into the bed, so it is not implemented, allowing the Z_MAX_POS sanity check to prevent the Z axis stepper from continuing to run indefinitely during a fauly probe condition.
If your Z_MIN_WHILE_PROBING were implemented, if the printer were powered off (or Marlin crashed, etc) with the head at any Z distance greater than Z_MIN_WHILE_PROBING there would be no mechanism to begin a print on the next poweron, since Marlin will never move in the negative Z axis except during G28 probing and after probing completes successfully. The only way to recover in this case would be for someone to continually attempt to both home the printer then power-cycle, moving the head Z_MIN_WHILE_PROBING closer to the bed each iteration. This would be an unacceptable user exerience.
Further, if Z_MIN_WHILE_PROBING were implemented, the only percieved safety measure it would add is that during probe failure, the Z stepper would crash into the bed and continue to run for only Z_MIN_WHILE_PROBING stepper rotations instead of Z_MAX_POS rotations. Regardless, the bed would be impacted so there is hardly any additonal safety added and a stepper driver should not overheat or cause any more damage (other than what was already done to the bed) in a single Z_MAX_POS length of rotations.
For more piece of mind during homing, you may use the Z_MIN_PROBE_ENDSTOP feature with a normally-closed limit switch wired to an available pin on your control board:
// Enable Z_MIN_PROBE_ENDSTOP to use _both_ a Z Probe and a Z-min-endstop on the same machine.
// With this option the Z_MIN_PROBE_PIN will only be used for probing, never for homing.
//
//#define Z_MIN_PROBE_ENDSTOP
Then use the Z_MIN_WHILE_PROBING feature to move the head above the Z_MIN_PROBE_ENDSTOP home position to perform probing. This will ensure that the printer will crash for only Z_PROBE_OFFSET_FROM_EXTRUDER Z stepper rotations if the probe malfunctions:
#define Z_PROBE_OFFSET_FROM_EXTRUDER 0 // Z offset: -below +above [the nozzle]
In summary, to prevent the head from crashing into the bed, you must ensure your Z probe is functioning correctly :) |
How to line up (x,y) print area between hardware and software? | The problem you are experiencing is because the position where the y endstop is triggered does not correspond to y = 0, but perhaps corresponds to y = 15 (replace 15 by the offset you're seeing). You can perhaps solve this by adjusting the endstop to trigger at the correct point, but you can also adjust this behavior in software: In your start G-code, after the homing (G28) command, insert a G92 Y15 to tell the printer that the current position (reached after homing) is actually y = 15.
Another option is to use the M206 command to permanently store the offset in EEPROM (rather than needing to provide it in the start code each time).
If your printer moves towards max rather than min, the same applies, but consider that the offset may be caused by the bed size defined in your firmware not corresponding to the bed size set in your slicer. |
Ender 3 X-Axis Layer Shifting | OK I found the answer. I removed the extruder cover and found that the 2 screws attaching the extruder assembly were loose. A very simple fix to a very irritating problem. |
Slicing directly from non-mesh data? | No, not natively
To the current point, all slicers in frequent use do use some kind of 3D model with explicit surfaces to cut up into slices and then solve the path functions to create the G-code. The model can be in STL or OBJ or some other format, depending on the slicer, but at this point (November 2019), no slicing program I know about supports direct math as input.
Probably make it yourself?
However, you have a way to design the models by solving a mathematical formula. You could probably use the program that solves the formulae to also act as a slicer of some sorts.
One software that might form a base could be Cura, which allows writing plugins, so there might be a way to have Cura automatically solve the surface formula and then plug that into the slicing without storing the intermediate data as an STL.
Slic3r might also work as a base since the whole source code is open. It would be a similar endeavor as modifying Cura. |
Apparently systematic nozzle clogging | The Micro Swiss hotend uses an all metal hotend. These type of hotends are more difficult to operate considering they do not have a Teflon liner that shield the filament of heat creep. From this article:
Jams and clogs are often from a combination of excessive heat and
non-optimal material flow. This effect is worsened by poorly cooled
all-metal hot ends, high torque extruder gears, small nozzles/layers,
slow printing speeds, too thin first layer, and excessive retraction.
The bold faced text in the quote sums up what is causing this. A smaller gear requires more force/torque as the arm i.e. the radius is smaller.
The article describes what steps you could do to alleviate the problem. Of all the suggestions, "Minimize retraction", seems a possible candidate for you to look into considering the posted print settings. As this is a heat related problem it is advised to also increase your printing speeds, these are pretty low (30 mm/s for slow and 60 mm/s for normal printing) and also check the cooling of the "cold end" (the fan that cools the radiator fins). Also reduce the printing temperature, 210 °C is pretty high for PLA filament, personally I don't go over 200 °C (note that this depends on your filament, but most PLA brands can be printed in the 185 - 195 °C range).
You have a pretty large retraction specified. The Ultimaker default is 6.5 mm is considered to be large, but works perfectly for Ultimaker machines (read Bowden tube setup). In my Ultimaker 3E which uses all metal hotends, or, in my custom HyperCube Evolution, also Bowden, but with a lined hotend, 6.5 mm retraction works perfectly.
Please look into this answer and this answer. Both describe that the retraction performance is worse with all metal hotends. My experiences are exactly the same with metal hotends, at least the cheaper production ones (I tested cheap all metal hotends, but also ran into problems because of production and design errors, I have not tried the better quality heat breaks/throats yet).
Please lower the retraction setting considerably to see if it has an effect. The Monoprice Maker Select uses a direct drive. Direct drive extruders do not need a large retraction length setting. If the filament is hot in the throat (as there is no PTFE lining that in fact acts as an insulator), too large of a retraction may not be reversed when the filament cooles during the retraction.
I think you might be experiencing what is described in this question: "Extruder prints fine up until further down the print". This answer describes issues of the metal heat breaks.
To comment on your statement in comments above, I am not suggesting you should use a liner in your current extruder. I'm pointing out the differences. Metal hotends are just more tricky to operate regarding retraction and heat management. |
How can I generate and export the support structure for an object as a mesh? | You could try Meshmixer its free and the supports generated with it are embedded into the 3D model.
Here you can find a reference on Meshmixer supports. Meshmixer is well known for making custom supports for complex 3D models. |
PETG Sticking to Nozzle | It could be that you're having issues now due to the filament absorbing moisture. I had a big problem with it doing that when I had waited for a few months after opening my filament. Dehydrated it using just my heated plate and a foil lined box over the top and it fixed it right up. If that's something you haven't tried yet perhaps that is what is causing you some grief. |
Replacing ceramic tape on Wanhao Duplicator i3 | I have a wanhao dup i3 too. It does not matter a lot which ceramic wadding and / or kapton you use. The more you seal off, the better it works. If you have a full metal hotend and a powerful cooling blower and need temps above 250 celcius, you might want to do the best you can with a good padding of ceramic and kapton. For normal operation, a half-done job will do just fine.
See this question too: Efficient and easy way to thermally insulate the heat block of the hotend? |
Would appreciate advice on unclogging hot end [Makerbot Replicator v2] | For molten filament to jam at the inlet to the thermal barrier tube like shown here, there must be something wrong with the cooling in your setup. The cold zone needs to be cool for the extruder to work right. It looks like the cold zone is not staying below the glass point of the plastic, so the filament softens and mashes into a jam.
Here are some common causes of jamming in Replicator 2 extruders:
Printer's environment is too hot. If it's in a hot garage, move it indoors. If you have it inside an enclosure or cabinet, open it up to let it breathe. All-metal extruders with PLA are very sensitive to ambient temp.
Not having good contact between the heatsink and the cooling bar. Normally, just bolting them together is fine, but a small amount of of thermal grease (like Arctic Silver) smeared over the contact surface can help too.
Not having the fan pointed the right way. You want the fan to blow ONTO the heatsink: this provides more turbulent flow and better cooling. This means you should not see the fan sticker when everything is assembled.
Fan dying and and spinning too slow. A new fan can be a good idea. (24v fans only!) Or you can try gently peeling off the fan sticker and adding a single drop of light oil (like 3-in-one) to the bearing, then re-attaching the sticker.
Using non-stock thermal barrier tubes that do not have the correct internal geometry. Lots of people sell bad tubes that don't have the "secret sauce" (internal diameter step). The four good vendors I'm aware of for this part are Makerbot, Carl Raffle, Micro Swiss, and Performance 3D.
"Floating" the thermal barrier threads by tightening jam nuts on both top and bottom of the cooling bar. All-metal extruder designs require good contact pressure between the thermal barrier tube threads and the cooling bar, to ensure maximum heat transfer. The best approach is to put one jam nut below the bar to forcibly push the thermal barrier tube threads into the cooling bar, and no nut on top. (Using just one nut on top is better than two nuts, but worse than one nut below.)
It's also possible that there is some material partially blocking the nozzle, which is contributing to molten plastic backup up and jamming.
What I would recommend right now is removing both the nozzle and thermal barrier tube from the printer, and either replacing them or fully cleaning them out. There are multiple options. You can preheat the hot end with the cooling bar in a vise or on a metal surface with no fans attached, to heat up the plastic in the thermal barrier tube and allow you to push the blockage through with a small allen key. Then do something to REALLY clean it out, like nylon cold-pulls, or cooking the parts in a blue flame such as a torch or gas stove. Then check for internal debris with something like a toothpick.
Once you can see through the nozzle orifice and there's no junk inside anything, reassemble everything properly and try again. |
FreeCAD: Scene doesn't look 3D | FreeCAD, like many other 3D programs (including blender) does use Isometric/Orthographic display as the default displaying option as it allows to see on one view if some lengths or angles are "equal".
In many cases, a perspective view can be toggled on to allow visualization in a room, especially for architectural one.
In the case of FreeCAD with all its options, this option gets swamped out of the user's field of view sometimes, and the documentation is not that easy to navigate without a proper keyword, making the right thing slightly hidden in the documentation. But once you know it, it should be obvious:
View > Perspective View toggles to perspective view.
View > Orthographic View toggles back to isometric. |
Ender 3 Pro Weak Prints after Cura upgrade | Total user error on my end. With the new version of this app somehow I did not catch that material selected on the UI dropdown was set to some custom material and not Generic PLA. Once I selected the proper material, all worked as expected. |
How to add tolerances/gaps in 123d Design? | Unfortunately, 123D Design doesn't have such a feature.
You could select multiple surfaces and push/pull them all at the same time though.
Remember, your 3D Printer's slicer should have an option to undersize or oversize holes and walls, to help reduce/remove the effect you talk about.
It's called size correction, I think? Simplify3D Has it, atleast.
You might have to tweak this for every filament type, though. |
Z-Axis raising and dropping substantially during print knocking print off | Can you send a link to a short video of the problem occuring? It might be because of multiple factors but we will be able to identify if there's a video of it.
Possible reasons:
Z axis nut might be too tight. This would cause the nut to bind with the leadscrew and so the motor will not be able to lift the z-axis further while printing.
Not enough retraction. This would cause filament to ooze from the nozzle during travel moves and this filament would collide with the printed object, causing it to be knocked off. A possible solution for this would be to enable Z-hop in the slider, have the hop distance set to at least 1.5x the layer height you are currently printing with.
Too small of a contact area. This usually happens more often for tall prints. A solution would be to increase brim in the slicer settings (at least 10~15 lines) to increase the adhesion of the print and the bed.
Dirty nozzle. Old prints filament may be stuck on the outer face of the nozzle and might be catching onto the printed part while printing. |
Connecting Sain Smart Mechanical End Stops to Ramps 1.4? | The multiple connections are redundant. Your picture indicates the wires labelled as "SIGNAL, GND, GND, VCC". This is correct if the board is a "standard" mechanical endstop v1.2. The two middle pins are both ground, but you only need to use one of them.
You can simply connect signal to signal and ground to ground. That is sufficient for the endstop to work. You only need to connect VVC to VVC if you want the LED that's on the board to function. |
Could you use ultrasonic vibrations instead of a roller with an SLS/SLM printer? | No
The problem is twofold. Resonance and Granular convection
Resonance
Let's start with an empty box. We toss in some powder to create the first layer and use an ultrasonic to create a first layer. What happens? The bed starts to resonate depending on the sound you send into it in patterns - and the powder starts to form valleys and ridges along them as one can see in this video.
Granular Convection
What happens if one shakes a box of fine granulate that contains larger items? Granular convection happens! All items raise simultaneously and the small items start to fall first, resulting in them getting under the larger ones, so as a result end up pushing the large items up.
Because of both effects, there won't be an even layer and it would raise the items printed, even if we managed to get good layers. |
E3D Thermistor Alternative | I like the mounting method on my Mk9 extruder: there's a small hole drilled into the heat block, parallel to the heater hole. B3 Innovations sells a thermistor packaged into a small spring. The whole assembly goes in the hole, and a setscrew (gently!) secures it. You could just as well put a bare thermistor into a spring or something similar; it's mainly to protect the thermistor from the setscrew, and from excessive bending.
From your photo, it looks like the same thing would work fine for your hotend.
However you mount it, adding some thermal grease can help conduction from the surroundings. |
After print failure cant extrude PLA | If the hotend is okay, consider yourself lucky and consider the nozzle a loss. Aside from specialty items like ruby tip ones, nozzles are a consumable anyway. Even if you can clean it out well enough to get it working again, it's unlikely to extrude as well as a new clean nozzle. Acetone is probably not going to help; PLA does not dissolve in acetone, and while the pigments/additives might be affected by it enough to weaken and deform PLA in a way that could get it to detach, what's clogging your nozzle is likely burnt/carbonized PLA that's unlikely to be removable by anything that won't also destroy the brass.
If you don't already have replacements on hand, get yourself a pack of 10-20 for $10 or so and be ready to replace when needed. You can also go for a mixed-size pack if you want to try printing fine details with a smaller (0.1-0.3 mm) nozzle or super-fast but rough with a large (0.8-1.0 mm) nozzle. |
FDM, ABS Post-processing methods | Your objective is quite a challenge, considering the limitations within which you have to operate. Small diameter of an appreciable length is daunting indeed.
I would consider to use a regulated heat source such as a soldering station, one with removable tips. After electing an appropriate tip to handle the dimensions, a hole drilled into the tip to accept the pin diameter and length would be the next step.
There would be consideration needed for the plastic material pushed ahead of the soldering tip, as well as angular stability while pushing the soldering iron onto the pins.
To improve these circumstances, a sharp edge ground at the end of the soldering iron tip would give a better shearing of the excess plastic.
The temperature control would be critical to prevent destruction of the pin within the tip and to provide optimum removal via the cutting end.
Directly related to the above, but without the heat would be the equivalent tool to be used in a rotary grinder such as a Dremel. The bore of the custom-made cutter would match the pin and the sharp cutting end could also have serrations such as those found on hole-saws. Low speeds and a steady hand are required!
If either of the above suggestions do not remove sufficient plastic roughness, one would make the pins of a larger diameter and ensure that excess is removed to specifications.
EDIT: Additional thoughts. Hobby stores and online equivalents will sell very small diameter tubing. It might be an easy matter to find 2 mm inside diameter tubing of brass or even stainless steel. A few passes on a piece of sandpaper while the tubing is chucked in an electric hand drill and you have a sharp cutting edge. |
How to connect filament end-to-end? | You'd have to ensure that the joining portion of the two filaments do not "bloom" or increase in diameter, which would happen if unconstrained at the melting and joining time. Alignment is also critical, otherwise you have a varying diameter from one color to the next at the point of join.
There's an item on ebay which is precision drilled and has precision machined mating surfaces to ensure alignment of the filament. The filament is extended through the two components, heated (in the case of the video, with a match) and the two metal parts are pushed together. The joining of the metal parts also cuts away the excess bulge of the melted filament, ensuring correct diameter.
In the case of the above item, the joining devices have to be threaded with the filament prior to the joining process and then have to be threaded off the filament by pushing them the entire length of the filament.
This would be okay if one were joining short lengths only. Consider how much fun it isn't to have to slide these two pieces over an entire full spool of filament.
A split device which would enable one to un-latch or otherwise open a clamshell to release the filament would require much more challenging machining to achieve the necessary precision, which is probably why we don't see such a product. |
Printing Plastic Replacement Parts | The easiest way is as you currently do: model the pieces by hand, using (digital) calipers to measure them.
Scanning technology isn't very good, and the models are not of printable quality. Usually, fixing a scan is more work than modeling an item from scratch. |
What does Marlin's G30 code do? | I believe G30 is a carry-over from CNC (G-code originated for CNC not printers)
I believe it is for going to a secondary reference (home) position and includes an optional by-way-of address that can be included in the command.
Looking at Marlin 1.1.0-1 (latest release), it seems to do what you said:
Move to the requested position (if selected in command, else N/C)
Deploy probe
Go home
Stowe probe
Report the requested position and probed Z position
Report the current position (home?)
It appears that for Marlin, there is only one reference address (home); so, it would seem a G30 would be the same as a G28 (go to primary reference); but, not so.
It looks like G28 is a home of a different color. It looks like it homes the axis one-at-a-time and does not support a by-way-of location. Note that you can select which axis to home by adding the letters 'X' 'Y' and/or 'Z' to the command.
I am not sure what benefit this command has for a 3D Printer other than allowing you to alter the printer's path to home.
Note: Unfortunately I do not have Marlin code up and running on my printer now so I cannot confirm what I am seeing in the code. |
Optimum ratio of line width to layer thickness for overhang performance? | Cura high-lights overhangs in red if the printer would end up printing completely into thin air.
If your overhang is angled instead of being a purely vertical-to-horizontal transition, a much thinner layer height can increase the chances that enough of the prior layer exists to support the next layer being printed.
Basically, the thinner the layer (within reason), the greater the allowable angle of overhang that is safe to print. |
Is having 4 bed leveling adjustment points rather than 3 problematic? | Similar reasoning as for question "3 vs 4 bearings for y axis travel"" holds. If you introduce a fourth point, it is more difficult to make a flat plane.
Provided that your bed is stiff, e.g. a flat piece of glass, or a thick metal plate is used, you do not need more than 3 screws. Both my custom built printers use pieces of glass on aluminium heated beds that use three screws for levelling.
If your bed is thin (e.g. thin heated bed with tape, so no glass) and flexible or warped, an extra screw might be handy to deform the bed to a more flat plane. |
Mechanics for modelling a push button latch spring | A commonly seen method to provide spring action in a 3D printed model is to use a series of curved and straight segments. Some of the designs use the segment assembly in compression, aligned with the plane of the print, while others use the segment assembly for springiness perpendicular to the plane of the print.
An example of the latter is a squishy turtle from Thingiverse resulting in four legs that provide some bounce to the turtle if dropped from a small height. This video shows the movement of the legs more clearly, as there were no suitable still photos available on the site.
The image below shows another spring type print from Thingiverse in which the forces are in line with the print plane.
The S-curves cannot be too thick or movement will be severely limited, and too thin means impossible to print and possibly reduced lifespan.
Many designs abound with a search of the 'net for "3D printed springs." |
Printing starts well but then it breaks down, Anet A8 | The answer by fred_dot_u is fine for small prints with relative low cost or batch producing, where a single part is needed several times. But if you print something big or unique it's not cost effective. Slowing down the complete print is also not very time efficient.
In Cura there an option called Mininum layer time, which addresses this problem:
Cooling -> Minimum Layer Time / Minimum Speed
This means that it will only slow down when the print distance for a particular layer is short. Giving the material some extra time to harden. Other layers will be printed at normal speed. Increased print time will be limited this way.
For your situation I would advise you to increase the part cooling, if possible, in conjunction with configuring the Minimum Layer Time option mentioned above. |
Show the time index in Cura Layer View playback | These estimates tend to be very approximate, even if Cura has the accurate acceleration values for your firmware. An error of 100% is not unusual.
What you probably want is an alarm at a specific layer (a few before the critical ones). You might be able to add this to Octoprint fairly easily - it does support plugins which can provide (for example) pushbullet notifications.
I'm not sure that 'critical' points are much more likely to fail than other less predictable things (like bed adheesion failure, extruder jams, filament breaks) - unless you're testing features (and then hopefully you can print only a slice of the part). |
Connecting (Anet A8) 2004 display to MKS GEN L | I've got the Anet A8 display working on a RAMPS 1.6 (which basically works exactly the same as a RAMPS 1.4 board). Note that at first connection I got exactly the same display when I connected the LCD to the EXP1 and EXP 2 headers using the "smart adapter".
To get the Anet A8 display working on a MKS GEN L (which is basically a RAMPS board as the pin layout of the RAMPS configuration are used in the firmware of the MKS GEN L) you need to forget about the EXP1 and EXP2 headers and the "smart adapter" for the Anet A8 display. Instead you need to connect the Anet A8 display to the AUX2 header. The only thing you would have to do (from the hardware side) is to switch the cables that are connected to the pin 1 and pin 2. I used Dupont connectors and cables. Pin 1 on AUX2 is VCC and pin 2 on AUX2 is GND, this is reversed on the Anet A8 display.
From the software side you need to set the following constants:
#define ULTRA_LCD // Character based
and
#define ZONESTAR_LCD
Please note that in the photo you still see the "smart adapter" with flat ribbon cables, but they are not connected to a screen, the reason for them being there is I also tested the full graphic discount controller in a previous session. The Anet A8 display is connected to the flat ribbon cable on AUX2 just above the top-left most stepper driver.
UPDATE:
At the time of writing and the use of an older Marlin version, this required to set #define ULTRA_LCD. In Marlin 1.1.9 you do not need to set #define ULTRA_LCD explicitly, this is now intrinsically set by #define ZONESTAR_LCD. To elaborate on this, the #define ZONESTAR_LCD sets the constant #define ULTIPANEL which sets constant #define ULTRA_LCD in Conditionals_LCD.h. |
Raft hard to remove? | A couple things to consider:
Ensure that your build plate is flat and level. An un-parallel HBP could result in the object "welding" to the raft.
Turn down your nozzle temperature. It is likely that the material is hotter than it needs as it is extruding. This results in a slower "cool-down rate". So, if it takes longer for the filament to cool between the raft and the first layers of the object. Therefore, cooling together in a manner that somewhat binds them.
Personally, 266C seems VERY high to me. I've primarily only used ABS on my MakerBot and have successfully printed with 225C +-5C nozzle temperature and 110C +-2C HBP temperature.
Typically you want to extrude slightly above the melting point. You don't want to liquefy the material, but make it pliable enough to bond it to other layers of material (or a BP). |
Can I print directly onto the bed plate? (Ender 3 Pro) | I've printed for several years direct onto the 3 mm heated bed of the Anet A8 I used; worked perfectly! I did use a specific 3D print spray for adhesion, see e.g. this answer on "Should you use hairspray on a metal bed 3D printer?". |
BLTouch with TH3D Ender 3 pro | After setting up the BLTouch in firmware, you need to activate the leveling function for every print by inserting the leveling G-code 'G29' in the start G-code script of your slicer. The code should be placed directly after command G28.
Note that you can test the working of the sensor from the printer display: sensor can be deployed, stowed, reset and tested. |
How to calculate the approximate volume of material used in 3d print? | Possible way to calculate the volume of material used is to multiply the filament length (calculated by CURA after slicing and displayed in lower right corner) and surface of filament tip.
Volume = Filament_length * ( Filament_diameter / 2 )^2 * PI
Of course do not forget to convert all values to the same order of magnitude (e.g. cm³). |
How do I keep the .stl files from being resized while exporting? | .stl Basics
The .stl format has no inherent sense of which units you use. items are to scale to an ambiguous 1, which could be 1 meter, one millimeter, one lightyear or one inch. To a .stl, only the relative sizing matters. All these faces you see are compared to a line with the length of 1-unit that is
Slicer-Modeling Software interaction
The most common graphic design programs export in millimeters, but some US ones just assume inches, which is a factor of 1"=25.4mm.
Cura, Netfabb, and Slic3r expect that the 1-unit line is one millimeter long - but if it is an inch instead, then the model is shrunk by 1/25.4 or to about 4% of the right size. Scaling up by 2540% one would return to the millimeter scale.
But then there are other programs that use other choices of scales. Blender for example assumes a scene is in meters by default.
Inventor
Inventor can export .stl in a variety of scales, which all just serve as how the length of the inherent but invisible 1-unit line is drawn. The default choice is centimeters, so a scaling factor of 1cm=10mm, which would explain the models being only 1/10th of the expected size in Slic3r. to change the scaling, follow the manual:
upon exporting a .stl, click Options
under Units, choose mm |
German RepRap NEO only heats to 130°C | A few possiblitites.
You wire is too small. If your wire is HOT that is a fire hazard.
Your thermistor is bad. Check with a high temp heat probe or try replacing thermistor.
Your heating element is bad (rare).
Last it could be a limit in your firmware. But that would surprise me.
Any chance you have the bed and the hotend reversed? If you had the Bed as Hotend, then it would max out around 100. This last one I would say is most likely.. |
Can I order a 3D print of a 1 ml transparent container? | Yes. You'll probably want to use SLA or Polyjet printers with transparent resin. For example, here's Shapeways' page on transparent SLA and their page on Polyjet (which says you need to phone them for transparent Polyjet parts as their online order system can't handle it).
FDM printing with transparent materials doesn't usually result in parts that look like transparent injection-moulded parts, because the lines of material laid down by the printer are visible. There are some techniques to make this better, but a printing bureau is less likely to offer this kind of special handling.
In any case, you should discuss your requirements in more detail with suppliers, and they'll be able to advise whether they offer any manufacturing processes that meet your needs. In particular, if you need your containers to be food-safe, you should mention that at the start, as it'll rule out a lot of possible suppliers, machines, and materials. |
Ender 5 Underextruding | If you pulled the Bowden tube out of the hotend and then got this, you almost surely installed it wrong, leaving a gap between the tube and the nozzle for molten filament to fill and jam in. Remove it again with the hotend hot, and if there's a mess inside, look for guides on cleaning it. Then, to reassemble, loosen the fitting from the heat sink by 3/4 to one full turn, press the tube in as far as it will go, then tighten the fitting down to compress the tube against the nozzle. |
What is the easiest way to modify/resize an object inside a STL file? | Many resources are available for modification using 3D CAD (including learning a tool such as sketchup, which is VERY beginner friendly). The easiest and fastest solution to your particular problem may be the reddit community "3D Print My Thing" which was created for EXACTLY this type of situation (help with modelling parts which will be 3D printed.) Another potential useful reddit community is "3D Modelling" which will have many people who are able to help you with this quadcopter/Go-Pro attachment 3D model.
Third solution: Thingiverse has a customize option for 3D models (including this one). have you explored using this interface to edit the model? This is most likely the easiest DIY solution. |
How to use a multimeter to test how many amps RAMPS is pulling? | To measure amps (current), the meter has to be wired in series with the item to be measured (for this reason, ammeters are designed to have very low resistance).
This has the down-side that you have to disconnect the component to put the meter in line with it. That makes it hard to do the "check a component at a time" method you mentioned.
An ammeter measures actual current flow, so you really can't test a component for it in isolation. Components can have wildly different "current draw" depending on the situation. For example, motor current varies with torque and speed; current through a resistor varies with the voltage across it; and so on.
There are special "clamp-on" current meters that just clamp around a conductor and report the current by using induction. Very nice if you have one.
If you just want the total current the entire RAMPS board is pulling, put the ammeter between the power supply and the RAMPS power input connection(s). Be very sure not to have the meter set to read volts or ohms when you do this (it might or might not survive). |
Help with Repetier firmware (probably locked and licensed) | The firmware is not really locked by the OEM1, you just need to upload (or, as you say burn) a new version. This is not unusual2.
Repetier is just like any other firmware that uses Arduino boards. While it may seem intimidating at first, it really is quite easy to regularly upload new firmware to the Arduino Mega 2560 board3. If you are not familiar with Arduino programming at all, then there are a good number of Repetier oriented guides out there, for example, from Repetier's own web site, Repetier-Firmware Documentation.
If are you familiar with Arduino programming, then it really is the same as uploading the Blink sketches, but the base code is just bigger, that is all.
There is too much to detail in a simple answer here and, as I stated above, there are a lot of online guides. However, en bref:
Download and install the Arduino IDE to your Mac/Wintel/Linux PC;
Download the latest Repetier firmware to your Mac/Wintel/Linux PC4;
Connect the Arduino Mega 2560 to your PC using a USB cable;
Change the settings, i.e. Boards and Port settings, in the Arduino IDE, so that your PC can "see" the Arduino Mega 2560;
Open the Reptier firmware, that you have just downloaded, in the Arduino IDE;
Configure the Repetier firmware source code, i.e. make some edits in the Configuration.h file5, with the tweaks that you require (which you mention in your question)6;
Compile the firmware using the Arduino IDE. and (finally);
Upload the firmware to the Arduino Mega 2560.
That is it. Once you have done it a few times, it will become second nature.
Watch a few online youtube videos, as well, before you start, so that you have a better idea of what to do. For example:
Repetier 0 92 8 Firmware Configration. or;
Kossel Mini 3D Printer Repetier Firmware and Calibration.
As an aside, for gaining a basic understanding of the configuration file of Marlin, not Repetier, I found this video #18: Calibration extremely useful. If you feel like changing the firmware completely, this may come in handy.
With respect to the Slicer's profile settings... I am not sure about that at all.
The reason that you have no RAMPS board is that, on your MKS board, the RAMPS and the Arduino are merged into one. Any new controller would work, as they all run the same firmware. There are many to choose from. As you are just experimenting (and reverse engineering) at this stage, I personally would buy a cheap Chinese Arduino Mega 2560 (€6) and RAMPS (€4) board on eBay, for a total of only €10 (if you buy them separately - don't get a combination "deal" as they are generally more expensive - although compare the prices first, you may find a bargain), to test your settings. Once you are happy with those settings, then you can spend more money on a better board, should you feel that you need it.
Footnotes
1 If you have to get around the licensing, then it sounds like you could/should recompile the firmware, or use some other firmware, i.e. Marlin. However, wait to see what other people suggest.
2 There really isn't much risk in uploading new firmware. The only risk, which I guess Mark is referring to, is that you may/will not have the correct firmware settings for your make of printer, and therefore the printer may not work correctly, until you get the firmware settings correct. So, in that respect, it is a good idea to buy a secondary Arduino Mega and RAMPS1.4 board to play with, for just $10 (Chinese clones work just fine). But, as you say that your board already has bad settings, and you do not wish to pay for more licensing, then you haven't much to lose by uploading new firmware to it, except the warranty. So, if I were in your position, and I did not want to pay the licensing, I would set aside the supplied controller (for possible future use/reference), and use a Arduino Mega 2560/RAMPS 1.4 combination to control the printer. That said, if your printer is still under warranty, can't you ask the manufacturer to supply you a new controller with the correct settings?
3 There is no way of extracting the code from an Arduino and then modifying it. You need the original source code files. See Is there any way to download a sketch from an Arduino? OK, you can obtain the machine code, but unless you are an expert, or guru, modifying it is impossible.
4 The installation order, of the Arduino IDE and Repetier, will not matter. You probably need to rebuild the firmware, I would guess, but I am not 100% sure. It might be a good idea to contact the vendor to see what they recommend, before overwriting their code (assuming that the control board came pre-programmed)
5 I found this video #18: Calibration extremely useful, for configuring Marlin, and understanding the firmware settings in the Configuration.h file.
6 You may find this link useful for your "tweaks": Repetier-Firmware configuration tool for version 0.92.9 version |
Is it legal to make a fire weapon with a 3D printer? | Sort of.
Fully plastic guns are banned in the United States by the Undetectable Firearms Act, because they can pass unseen through metal detectors - a huge problem. One way around this is to insert a slip of metal into the gun, thereby making it detectable by metal detectors. 3D-printed guns made of metal - were that possible - would not violate the law.
The UFA was originally enacted in 1988, extended from 2003 to 2013, and then again until 2023. It may seem a bit unnecessary, as all-plastic guns, like the famous Liberator have structural issues that make them extremely difficult to fire when compared to a normal all-metal gun. However, then can be used.
In Europe, gun laws vary widely by country. The European Firearms Directive is the main international agreement on gun laws, but its application varies, and it does not directly address 3D-printed guns.
I'm not aware of a country that has an explicit law banning 3D-printed guns. The only countries that ban them either unintentionally cover them or ban guns altogether. |
Has anyone experienced scorching or burning of cork heatbed insulation? | The real fire risks with 3d printers are electrical in nature. Lets say your wires come undone and happen to come next to the other lead. If it archs and happened to be in just the right position. Pretty unlikely. Once I did a bad solder job and when I was working on my printer. The wires literally burst into flame in my lap. No damage done (other than needing a new board). That said we often see people who have their boards catch fire in the flashforge owner groups. Weak solder joints, and over all bad quality.
Do you really have something to worry about with Cork? No. A series of unfortunate events would have to happen, and more likely your board will cause a fire. That said I would look into adding a layer of aluminum and reflect some of that heat back up.
If you are that worried I would just remove the sheet. Unless you are having issues calibrating you heated build plate / PIDs there is no reason to use insulation. I only use it on printers that have a hard time hitting ABS temps of 100c. Even then I only use tin foil and try to trap the air underneath.
Just a side note. If you are having heated build plate issues maybe what you really need is a heat chamber. |
Removing Supports Inside Cone | That model didn’t need supports. Most of the time you can break the connections with the supports to the model by inserting a knife or flat object into the cavity. In this case you could use a probe to loosen the joins then wedge it out. If that doesn’t work you can use needle nose pliers to remove prices.
Your best and cleanest option would be to reprint the mouthpiece without supports at all. Print it so that the wide section(the section touching your mouth) is on the buildplate. |
3D printing extruder not warming up | I really hope the extruder doesn't warm up as this would be a problem with your stepper motor :)
I'm sure you mean the hotend doesn't heat up. This could be a number of things but i would start to double check the pinout in your code vs the pins on your mainboard and measure if it gives 12v/24v (depending on your PSU) output.
There's a lot more info required to properly help you.
What board are you using?
What OS are you using? Marlin or something else?
Share the relevant code of the OS for the hotend config/pinout
potentially share pictures of your wiring |
Differences between moving build plate vs. extruder | Without going into too much detail, since this is a very exhaustive topic, I'll write some pro's of each down from the top of my head:
Cartesian XZ hotend, Y bed (eg. Prusa Mendel):
easy to build (relatively)
easy to maintain
easy to modify
understandable kinematics
with the right frame, no x-y-z orthogonality (90 degree angles) needs to be adjusted
affordable
bad for timelapse recordings
print quality will theoretically always be inferior at the same speeds and accelerations to kinematics that have less mass to move (heavy printbeds will lead to ghosting)
z-wobble is only existent in this approach
big build-plates are no option for this design (last feasible size might be 20x30 cm)
Cartesian XY hotend, Z bed (core-XY, sparkcube, Ultimaker, Makerbot)
less mass to be moved -> faster print speeds possible
almost no size limitaions
construction is easy to enclose in most models due to the cubic frame
looks almost always professional
enclosure can be hard to modify due to constraints in space
XYZ hotend (Delta bots)
master of circles
less mass to be moved -> faster print speeds possible
impressive to watch
more load on the processing unit due to more complicated kinematics (32 bit needed for fast print speeds and responsive control with display)
kinematics not easily understandable
error-cause search can be very complicated
more accurate in the center than on the outer limits due to the kinematic approach
The list is for sure not complete, and as a major disclaimer: print quality will always, with every approach, depend more on the setup and calibration of the printer than on the model. There are people around that produce great prints from an acrylic frame cartesian printer and lots of people that produce mediocre results with expensive printers in fancy designs.
I will add some links to the list items when I find the time, for now you have to believe me. I am highly appreciating corrections and additions! |
FlashForge Finder turns off mid print and is unable to reset or update firmware | 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. |
Increasing hotend temperature to compensate for increased filament throughput | I think I see what you're asking, but I think you may be thinking about it incorrectly. It's really all about heat being added to the system at the same rate that it's leaving. The heat block is there as a heat reservoir from which the filament draws heat for the glass transition. The heat in that reservoir is maintained by cycling the heating coil to add energy (more heat) to the systems as it's lost.
In the very local vicinity of the nozzle, the temperature will decrease slightly as it's being transferred to the filament, but because the heat block is massive in comparison to that drain, and because the heat block is a good thermal conductor that temperature reduction is very small.
I do not know what tolerance and hysteresis are built into the temp controller, but think the variation is likely small. The difference in additional heat required (more energy into the system) for any practical difference in feed rates (40 instead of 60) is thus likely to be very small compared to the filament cooling experienced immediately after it leaves the nozzle.
Bottom line: the adjustment you would want to make is not to increase the temp, but increase the duty cycle of the heating element to maintain the desired temperature. |
Installing E3D v6 extruder electronics on Ultimaker Original+ | I'm not certain exactly what you're asking, but I'll give this a try. If I've misunderstood, please give more details.
It sounds like you're upgrading to the E3D v6. If so, the heater, thermistor, and fans should connect to the same places as the previous ones did. The exception would be if you had thermocouples instead of thermistors before; that's a more complicated change.
If the connectors are different, you'll need to get matching connectors from a parts supplier (post a picture if you're having trouble identifying them). Fortunately each of the items you mentioned has 2 wires, and for the heater and thermistor polarity doesn't matter.
For fans, polarity does matter, so match up red vs. black, or just try it one way, and if the fan doesn't turn, then swap the wires. Some fans have a third wire, which could be used for measuring the actual rotation speed (see [http://pcbheaven.com/wikipages/How_PC_Fans_Work]); but it's not necessary. |
Marlin - slowing down print speed | G-code commands are not in the firmware. You're trying to find something in the code that simply isn't there. Your slicer is responsible for generating a g-code file, which you can then send to your printer (either by printing from an SD-card) or using a host program like repetier or cura. The commands in the G-code file are then executed by the firmware, but they're not part of it.
How the commands are actually processed is not trivial, and it happens in multiple places.
The commands are processed in the process_commands of Marlin_main.cpp. Upon reading a G1 command, it calls get_coordinates to parse the X/Y/Z/feedrate values from the command. Next, the prepare_move function is called, which applies the necessary transformations (clamping to endstops, and the reverse kinematics if you have a delta) before calling plan_buffer_line (which is in planner.cpp) which does some more preprocessing and computes a "block" which is pushed onto the buffer. Next, planner_recalculate is called, which iterates over the blocks to find feasible acceleration patterns for them. Finally, an interrupt (in stepper.cpp, ISR(TIMER1_COMPA_vect)) periodically reads blocks from the buffer and actually sends the signals to the steppers. |
Can I print an object with thin walls? | The test print in your question used an SLS printer. The other common alternative is FDM.
For narrow outer walls with FDM printers, you want to think in exact multiples of the nozzle size. Furthermore, unless the printer is well-tuned, you usually want at least two rows for the outer wall. Otherwise any flaw at all in the outer wall leaves gaps or tears in the final print.
Given most printers use a 0.4 mm nozzle, this means the typical minimum wall thickness is 0.8 mm. I use this number fairly often with good success.
If you want 0.5 mm you could step up to a larger 0.5 mm nozzle and give it a shot with only one row. But again, this will end up with flawed prints unless the printer is very well tuned and assembled, and if strength is one of your concerns it will be significantly weaker. Or, you could step down to a 0.3 mm or 0.2 mm nozzle and use multiple rows. Neither of those sizes will give you exactly 0.5 mm, but they can get you a smaller thickness than the 0.4 mm nozzle, at the cost of longer print times. |
How can I solve the problem of line distortions on my prints? | As towe noted in a comment, that's Z seam. The mechanism by which it happens is mostly the same as bulging corners. The print head is moving considerably slower just before and just after the layer change, decelerating to a stop then accelerating back up to speed in order to perform the slow Z-axis move, but the rate of extrusion, which is dependent on pressure built up (in the compression of the filament, and possibly expansion of the bowden tube) between the extruder gear and the nozzle, not directly on the motion of the extruder gear, remains the same.
In my opinion, the best fix for this is upgrading to firmware that compensates for it. In Marlin firmware (the stock Ender 3 firmware is based on Marlin 1.0), this feature is called Linear Advance. You tune a spring constant for your material's compression, and after setting it, extrusion is roughly proportional (linear) with respect to logical E-axis moves, regardless of the speed the print head is moving. However, upgrading firmware is nontrivial to do, and has a lot you might have to tune/tweak to get it working the way you want.
Short of that, there are a lot of ways of mitigating the problem:
Select "Outer Before Inner Walls" option (in Cura, or equivalent in other slicers) and "Infill Before Walls" when slicing. This will ensure the layer change happens inside the model (on an inner wall) not on the outside surface, because outer surface is neither the first nor last thing printed on the layer.
Increase X/Y acceleration and/or jerk limits so that motion can start/stop a lot quicker, so there's a shorter period for the disproportionate extrusion to happen in. But if you do this too much, you'll get layer shifts (failed prints) or at least bad vibrations that show up in print surface. My Ender 3 can handle accelerations up to at least 3000 mm/s² in X and Y and 10 mm/s "jerk" (a misnomer, thus the weird unit) and you can likely go higher if you experiment.
Tell slicer to retract at layer change. This will prevent extrusion from continuing during the Z move, but won't help with the extrusion during the time you're decelerating/accelerating in the XY plane.
Increase Z acceleration/jerk. This will reduce the time spent during the Z move, and might even allow the head to keep moving without coming to a full stop during layer change (but that can't happen if you turned on retract at layer change).
Print at lower speed. If the max speed is lower, it takes less time to decelerate/accelerate from/to it, and nozzle pressure will be lower (so extrusion is naturally closer to linear).
Enable coasting in your slicer. This is a hack that really poorly approximates what Linear Advance would do, but it can lead to underextruded walls and brittle prints. I always had bad luck with it and don't recommend it, but it's an option.
I would recommend starting with items 1 and 2. #1 is good for a lot of other things too, and the only tradeoff is that overhangs might print worse (in my experience that's not really the case and they sometimes print better even). And #2 will get you faster prints (sometimes at lot faster) which is nice. |
Will I brick my Ender 3 if I try flashing without a bootloader? | My understanding is that if there is no bootloader, you can't flash over USB; you need an ISP programmer attached to the board. So if you're able to flash over USB, that means there is a bootloader, and updating via the bootloader should leave the bootloader in place and just overwrite the rest of the firmware. I'm not sure how strongly this is enforced, though, so you may want to wait for supporting comments/answers from others.
Of course it's always possible to damage your printer in ways that require replacement of the board or at least some parts on it if you flash bad firmware, so you need to be cautious anyway. |
Good lubricants for moving parts? | When dealing with lubrication of plastics, any solvent or reactive substance is to be avoided. Petroleum is risky and Vaseline™ is a brand name for petroleum jelly.
I've had quite good results using inert lubrication such as PTFE and silicone based lubes. PTFE is the generic term for Teflon™ and is quite a good lubricant. There are both silicone and PTFE greases for higher viscosity applications.
From the Teflon™ link:
Teflon's amazing properties are down to its structure. Like most
polymers, Teflon has a carbon-based chain. However, instead of
reactive C-H bonds which occur in most polymers, Teflon has all its
hydrogens replaced by fluorines. These strong C-F bonds are extremely
resistant to attack by any other reagents, making Teflon very inert.
This means that no other molecules will react with or stick to Teflon.
The exception is Teflon itself, which will stick to itself quite
readily, forming thick layers or solid blocks. With a friction
coefficient of <0.1, Teflon has the second lowest friction coefficient
(surpassed only by diamond-like carbon), which makes it perfect for
non-stick items e.g. pans. DuPont invented the non-stick pan coated
with Teflon in 1956 and have manufactured it ever since. Teflon
coatings are so slippery that they are the only material that a gecko
cannot stick to.
Who knew that gecko testing was a thing?
Wikipedia for silicone grease:
Although silicones are normally assumed to be chemically inert,
several historically significant compounds have resulted from
unintended reactions with silicones.
Powdered graphite is also a good lubricant if one can tolerate loose powder in some constructions.
I've read of others using lithium grease, but not for plastic lubrication. |
Are all glue sticks PVA-based? How to find out? | Not all glue sticks work! The working ingredient of a glue stick is Polyvinylpyrrolidone; a more elaborate answer is found here on question: "Why does hairspray work as an adhesive for ABS?".
There are very good alternatives to glue sticks and hair spray nowadays. Specific adhesion sprays exist for several years now (e.g. Dimafix, 3DLAC, Magigoo for PA, Plasticz, PrintaFix, Dr.Mat, etc.; my personal experience is with the first 2 mentioned, both work for nylon: Dimafix has more tack at higher temperatures > 80 °C, up to 80 °C 3DLAC works perfect).
E.g. Dimafix has a higher temperature application range than e.g. 3DLAC. From the manufacturer can be seen that:
Source: http://www.dima3d.com/en/home/dimafix/
This image shows that the spray has its maximum tack/adhesion at about 120 °C and holds this adhesion level at least up to about 145 °C according to the image.
However, not all glues get stronger with temperature! PVA (also called PVAc) glues soften very quickly, making them good only at low temperature. See "Influence of temperature on the strength of bonded joints" which discloses this graph for PVAc Rhenocoll 3W, 4B Plus: |
Is possible print this level of detail? | It is definitively possible to do what you want, but your questions are samewhat problematic:
So, I need to know if it's possible to print that cylinder hard enough to work as an axis.
"hard enough" is a mysterious quantity. What is the intended application? The load of the axis, the rotation speed, the medium in which the part will be in, its operating temperature... they all affect the answer.
And what should be the gap size between the cylinder and the counter part's hole to rotate properly?
Reading at the question and the comments, I think you may have the wrong representation model in your mind. There are four different concepts at work here:
Accuracy is the maximum dimensional variation between parts.
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.
So: what you want to achieve for the object to rotate is to have at least some clearance once you have the parts printed. Therefore, you want to design your part with an allowance which is at least as much as the accuracy.
Note that a machine cannot produce parts with a tighter tolerance than its accuracy. So you must design your part with a tolerance equal or greater than your printer accuracy.
The correct number will therefore be entirely dependant from the specific printer you will be using. You can find out the specific accuracy of a printer by printing a tolerance test (I know, I know... why isn't it called "accuracy test"?)
See this unrelated answer - from wich I took the above definitions - for learning more about the above and a concrete example.
If it's is like 0.05mm, can I print that level of detail with a 3D printer too?
I hope it is now clear why this question makes no sense: clearance is a variable which depends from accuracy (and the application), not the other way around.
I can't add so much gap because I have really limited space
This comment too is incorrect: the "gap" (clearance) can be very very small. You have to have the correct allowance in your design, and allowance will not intrinsically make a part larger.
What hardware and material should I use to do this?
Again: this is entirely dependent from your application (load, operating temperature, orientation, speed...)
A consumer-grade FDM printer (easy accessible, cheap and cheap to operate) will allow you to print a rotating part, a SLA/DLP printer (less common, toxic resins, more expensive to operate) will allow to print the same part with different materials and tighter tolerances...
I don't worry about breaking, but it cannot be flexible
Again: without an explanaton of the intended use (or the numbers associated to it) it's impossible to answer this comment conclusively. Resins tend to harden to more rigid solids, but you have thrown around tolerances as small as 0.05mm in your writing, and over 12mm of axis, that is a deviation of less than 0.5% from "perfectly straight". I'm hard pressed to think you will find a printable material with such a rigidity. |
Why can't powder-based 3D printing techniques create enclosed or hollow structures? | It's not hollow or enclosed structures that are a problem. It's structures that are hollow and enclosed. Think about it. The machine lays down a thin layer of powder, and then a laser fuses some of that powder together to make a shape. Then, it repeats the process for the next layer. If you try to build, for example, a hollow ball, the ball gets created in the middle of the powder, meaning that there will be powder around the outside of the ball and in the inside of the ball, and with the ball completely enclosed, there is no way for the powder that ends up inside the ball to get out. That's why hollow objects printed in this manner must have a hole somewhere to let out the powder contained in the object.
This is not only true of metal powder, but any 3D printing process that uses a bed of powder or a volume of resin and fuses some of it into a shape. Fused deposition modelling (FDM), the process used by most consumer-level 3D printers, build models in thin air, so only air is trapped inside the printed objects, making the printing of hollow, enclosed objects less of a problem. Of course, one has to deal with gravity. A hollow object might collapse before the filament hardens enough to support its weight. |
How to minimize damage when removing an ABS print from a heated glass print bed? | I moved to a plain glass heated bed with a brush applied acetone and ABS mixture. Using an old emptied nail polish bottle with brush, I added some acetone and then threw in ABS pieces until it reached a brush-able consistency. I then brush it on the glass build plate where I believe the print will occur, and it works very well. On removal of the part the coating comes with it.
I just found previously that ABS would adhere to my kapton taped heated bed too strongly to use, and so while this involves a little work before each print, it's overall better than kapton for me.
I did experiment with sheet metal beds coated with kapton, but they curl during printing due to the ABS thermal stress, allowing my parts to be concave on the bottom side. Easy to remove from the plate, though, since it flexed. There may be a good middle ground material but I didn't experiment further. |
Simplified 3d will not run extruder | To use Mach3 to send printing you need to define the extruder as A instead E, since Mach3 only drives X,Y,Z,A and B motors. you can see an explanation here on is possible to send printing with mach3? Question.
I haven't tried to see if is possible to change this letters directly on 3D simplify, because i'm using ramps 1.4 instead the CNC driver control. But there is an ADDON to slice your gcode with Mach3 called gnexlab, try to find it on google search. I've downloded few years ago.
Added Link
gnexlab Addon |
Short circuit on RAMPS 1.6 board? | Your multimeter showing continuity doesn't necessarily mean there is a short.
All your multimeter is doing is applying a small voltage and then, if the current that flows is over a certain threshold, reporting that there is continuity.
The components (stepper drivers, microprocessor) on the board, draw current. That's normal. The current draw might be enough for your multimeter to report continuity. Because the multimeter is only using a small voltage to test (and not the required 12V), the current draw may be intermittent (not enough voltage for the microprocessor to actually start working), and capacitors getting (dis-)charged may also affect things.
The multimeter not reporting continuity is a guarantee of no short. However, the multimeter reporting continuity does not mean there is a short. The only way to find out is to apply 12V. If you use a current-limited power supply the possibility of damage if there is a short circuit is limited. |
How does a 3D printer print itself given the huge size? | 3D printers obviously do not print themselves. 3D printers are sometimes used to make parts for other 3D printers, and this is what is meant by "self-replicating" 3D printers (which is a bit of a misnomer, considering that all designs require a substantial amount of non-printed "vitamin" parts).
A printer that is able to produce many of its own parts is the RepRap snappy, which (as the name suggests) consists of plastic parts that - while they each fit within the build volume - snap together to make up the larger parts for the printer. |
Insulating bed material? | I’m pretty sure it can be brought separately but usually comes with a lot of heat-beds; It is a type of foam that has adhesive on one side and aluminum foil on the other. Here is an example of what I mean: |
Issue with weird heated bed temperature fluctuations | It really looks like either a bad thermistor or bad electrical joints. Are the connections to the thermistor itself crimped, or soldered? Are there connectors near the thermistor that can get heated up by the bed?
Electrical connections that are mechanical in nature (such as crimped, or using a connector) can degrade quickly in heated situations, particularly when the heating is intermittent. Consider replacing crimped connections with soldered joints where possible, and where not possible use gold plated connectors and crimps to avoid oxidation issues. Oxidation and poor mechanical connections will change resistance based on movement, giving you apparent increases or decreases in sensor readings even when the sensor output is unchanged.
Chances are good that the thermistor itself is fine, but that the connection nearest the thermistor, between the wires and the thermistor, is faulty. If it's crimped, you may be able to simply add flux, then solder it to increase the bonding between the wires and the thermistor and decrease future issues with oxidation.
Note that soldering is really only useful for the thermistor on the bed. The head thermistor must be crimped, as the head temperatures can exceed solder temperatures depending on the target temperature and the solder alloy. |
Raised lips on sharp vertical corners | If the amount of time spent by the head near the corner is significantly longer than what's spent per unit time elsewhere, you will get more extrusion at the corners and thus this effect. High print speed combined with low acceleration/jerk limits give you this effect.
Using a lower print speed should make it go away because it takes very little time to accelerate up to (or decelerate down from) a low nominal print speed from (to) the max cornering speed allowed by acceleration/jerk settings.
Using (much!) higher acceleration/jerk limits can also make it go away, but can cause other print quality problems and even layer shifts.
The Right Fix for this problem is upgrading your firmware to a version of Marlin with the Linear Advance feature and calibrating it. Then the printer automatically compensates for this issue by varying the pressure at the nozzle to get consistent extrusion rate regardless of changes in speed. I've had very good luck with this on my Ender 3, with a calibration constant K=0.6 for PLA and significantly larger ones for flexible filaments (the constant is essentially the spring constant of the filament for the length of filament path between the extruder gear and the nozzle).
If you don't want to replace the firmware, though, increasing acceleration and jerk limits and lowering speed should work ok. Your machine should be able to handle 3000 mm²/s acceleration on X and Y axis, with 10 mm/s "jerk" (the units of jerk are weird because it's not actually jerk; see my question on the topic). |
PLA 3D Print longevity after being printed | I've had white Hatchbox PLA parts in use outside in direct or partial sun for 3+ years without issue. It may help if you paint the parts for outdoor use, but I suspect the environment you describe would be fine for several years. |
Problem with 3D printer not accepting new print gcodes | As you found out, sometimes the SD cards need reformating.
And if they came with the printer they could be fake (Fake capacity), when you try to upload more than its real capacity they will just corrupt.
My Anet A8 came with an 8GB SD card and it turned out to be 128 mb. With RMPrepUSB you can do a capacity test and format it to the real capacity. |
How come firmware isn't uploaded? | Are you sure that the firmware is not being uploaded?
You can debug this changing the STRING_VERSION on Configurations.h.
If the version is not being changed, I would suggest trying to use another computer to upload the firmware or even another board. |
My Anet A8's filament isn't extruding properly | The most likely problem is that you have a jammed nozzle. To fix this heat the hot end up to about 140°C then insert some PLA filament push it through and then pull it out forcefully. That should solve your problem. |
What is the meaning of extruder's knocking sound? | The extruder motor may be skipping steps because of insufficient current. Typically, the stepper driver will have a small potentiometer that lets you increase or decrease the current. Increasing the current will give the motor more torque/force, but also cause it (and the driver) to heat up more. |
12v 25A 300x300 Heater bed Mosfet | I have looked over the 20 or so most expensive MOSFETs in that shop, and none are suitable.
It might be easier to go with a DC Solid State Relay rather than a bare MOSFET.
If you do want to go with a bare MOSFET instead, a TO-220 package MOSFET can dissipate around 1W without a heatsink, and around 3-4W with one. For 25A of current, that means the resistance of the MOSFET can be at most 1.6mΩ (without heatsink) or 5.6mΩ (with heatsink).
The IRLB3034 might be suitable. It is readily available on ebay (for under $1 shipped). 195A, 40V and an Rds(on) of only 1.4mΩ. A caveat is that this resistance is specified at a gate voltage of 10V, whereas it is slightly higher at lower gate voltages (1.6mΩ-2.0mΩ at 4.5V). This is on the upper edge of what is acceptable without a heatsink, but should still work (junction to ambient resistance is 62C/W, so in the worst case of 2.0mΩ the temperature would get to 77C above ambient - the MOSFET is rated for operation up to 175C). With a heatsink it will definitely work well.
Note that the gate charge is much higher (10x) than that of the MOSFET used on RAMPs so you should use a low switching frequency. |
Problem with swapping E0 for E1 (TEVO TARANTULA MARLIN 1.1.x) | At first it was unclear from where the snippet you posted is taken from as it was not stated in the question (this has now been addressed by a moderator edit).
Depending on the value of EXTRUDER_USE_E1 (and where and how it is set) the underlying code of the if statement will be carried out. Albeit said, swapping lines will not work, if you want to use the E1 connector of your motherboard, you have to make the printer think that it is using the E0 while it is redirecting to E1! This implies that you need to assign the pins of the E1 to the E0 extruder (so swap the pins, not the lines). This has been explained before in this topic by editing the correct pin layout file of the Marlin firmware.
EDIT : Further investigation shows that you have a custom Marlin for the TEVO Tarantula and are using the fork of Marlin maintained by JimBrown (this is essential information for your question). I have looked into the files, the only thing you would need to do is define the constant EXTRUDER_USE_E1 in your configuration.h file:
//#define EXTRUDER_USE_E1
to:
#define EXTRUDER_USE_E1
So do not swap anything. Once this constant is defined, the pin re-allocation is done for you automatically! (see the pins_RAMPS.h file)
Basically, this is exactly the same as is explained in topic How to change E0 to E1 on Marlin 1.1? and hence a duplicate. ;) |
Diamond hotend with single extruder motor | You might save yourself a lot of engineering by copying the Prusa MMU2. Here is a github project for someone doing exactly that: https://github.com/cskozlowski/mmu2
If you look on Prusa's web site, you can find printable parts for the MMU2. |
What's the best way to pull out fillament from an all metal print head | Cold/Luke-Warm pulling can cause damage and wear to the hot end, and should only be used in cases where you suspect there's a jam in the nozzle.
Pulling the filament out at a few degrees above its normal printing temperature and pulling quickly is what I use, and what I've seen other's also say, and the procedure most machines use. |
Anet A8 mainboard dimensions and hole offsets | Why don't you download one of the many casings that are found on Thingiverse that house the Anet electronics board and take measurements from those cases.
I measured it from my own casing I designed a while ago:
Hole distances are 86.3 mm and 91.3 mm (center to center) |
3d Print pops off before complete | The issue you are having is with bed adhesion. Basically your print is not sticking well enough to the printer bed for it to resist the mechanical forces the part is being subject to during print.
This the most common issue in FDM 3D printing and you can find a lot of material online on how to mitigate the problem. Just as a "teaser":
You need to ensure your printing bed is dead flat and at the right distance from your nozzle. Search online for: "3D printer bed leveling": the procedure is normally the same for most printers, but if your model is a common one, chances are you will find a video/how-to/blog post for it.
You should use print settings that facilitate bonding of the plastic to the bed (typically: slow speed, no cooling fan, a bit of overextrusion...)
You can tell the slicer to add a brim or a raft to your part, to increase the amount of material holding down the print.
You have the option to cover your printing bed with some material that increases the bonding of the plastic (this is type-of-plastic specific but typically: PVA glue or painter's tape work well for PLA, and kapton tape or ABS sludge for ABS).
If you have a heated bed, use it (each filament brand has their own recommendations, but typically 50-60°C works well for PLA).
As for why your problem is happening towards the end of the print, there are a series of factors contributing to it:
The tallest the print, the more mechanical advantage (leverage) the nozzle has over the adhesion surface.
It looks like your top layers are progressively small, so they print very fast and this in turn the does not allow for the plastic to completely cool down before going to the following layer, so when the nozzle prints on top of it the still-warm layer is "pulled up" by the action of the nozzle, creates a blob that sticks to the nozzle and that only makes situation worse. Snowball effect. Most slicers have an option "minimum layer time" that delays the print of a layer until X seconds have passed from the start of the previous one.
It may be possible that the Z axis is inaccurate and the nozzle catches on a previous layer. Most printer have some kind of accuracy issue with the Z axis the furthest up the head is, especially if the Z-axis has only one threaded rod (as for example in the Creality CR-10). Another source of inaccuracy (see comments below) could be the movement multiplier for the Z being slightly off, causing the nozzle to dig into the print if there's overextrusion and enough cumulative error. Z axis inaccuracies are however unlikely to be the primary source of troubles for you, but it's a "nice to know" thing to know about 3D FDM printers, nevertheless. |
Is it necessary to replace the SMD fuses in RAMPS 1.5 or greater, for use with 24 volts? | Without knowing the exact part numbers used for F1 and F2 it is impossible to say whether the fuses need to be replaced or not. However, based on the manufacturer provided schematic and BOM we can make a pretty good guess.
Looking at the PDF you linked, it states that F1 is rated for 16V. Looking at the BOM spreadsheet it also says F1 is 16V and 30A.
Based on the fact that the only two reference documents available say 16V, I would strongly recommend replacing this component for 24V operation. |
How do I use MEGA pin number designations rather than RAMPS pin numbers in marlin | The pin mapping that Marlin uses is defined in the file fastio_1280.h:
/**
* Pin mapping for the 1280 and 2560
*
* 1280 22 23 24 25 26 27 28 29 53 52 51 50 10 11 12 13 37 36 35 34 33 32 31 30 21 20 19 18 81 82 83 38 00 01 78 05 02 03 79 80 54 55 56 57 58 59 60 61 41 40 39 71 70 04 17 16 84 06 07 08 09 85 15 14 72 73 75 76 77 74 62 63 64 65 66 67 68 69 49 48 47 46 45 44 43 42
* Port A0 A1 A2 A3 A4 A5 A6 A7 B0 B1 B2 B3 B4 B5 B6 B7 C0 C1 C2 C3 C4 C5 C6 C7 D0 D1 D2 D3 D4 D5 D6 D7 E0 E1 E2 E3 E4 E5 E6 E7 F0 F1 F2 F3 F4 F5 F6 F7 G0 G1 G2 G3 G4 G5 H0 H1 H2 H3 H4 H5 H6 H7 J0 J1 J2 J3 J4 J5 J6 J7 K0 K1 K2 K3 K4 K5 K6 K7 L0 L1 L2 L3 L4 L5 L6 L7
* Marlin 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85
*/
Confusingly, the line labeled "1280" refers to the pin numbers as used by Marlin. The line labeled with "port" gives the corresponding port/pin combination as it applies to the AtMega2560 chip itself. To translate between the port numbers given above, and the Arduino Mega pin numbers, you can use the following pinout diagram:
Licensed under CC-BY-SA, www.bq.com
For instance, according to the pin mapping in fastio_1280.h, Marlin pin 34 is "port" C3. This corresponds to PC3 in the pinout diagram, which is Arduino Mega A11. Conversely, if we want to find out what PWM pin 9 is in Marlin, we can look at the diagram to find out it corresponds to PH6, which is Marlin pin 09.
If you want to change this numbering, then unfortunately, you have to make quite a few modifications to fastio_1280.h. For instance, the "reason" pin 5 corresponds to the pin it does, is the following set of definitions:
#define DIO5_PIN PINE3
#define DIO5_RPORT PINE
#define DIO5_WPORT PORTE
#define DIO5_DDR DDRE
#define DIO5_PWM &OCR3AL
If you wanted to point pin 5 towards another pin, you'd need to change all of these lines to point to the correct registers and pin numbers. |
Anet A8 Calibrated but print sizes are wrong | If you have a stock printer, your calibration values should be stock - i.e. 100 steps/mm for X/Y. Also, as an extra hint, X and Y steps should be identical since the mechanics are identical (unless you swapped out one of the motors or drive gears).
It sounds like you have Y correct, but something wrong with X. This could be the belt being damaged, or slipping (or having slipped during your initial cal, and not later). Try increasing/decreasing print speed, this might show up some dynamic problems with the movement.
If you print something bigger (like this) you don't need to worry so much about measurement precision and under/over-extrusion (which adds to the error, but doesn't scale with size).
You don't make it clear if you've started by calibrating your extruder. This is the most important first step, and ensures that you extrude the right length of filament during the print. All this requires is that you mark out 10cm of filament and extrude it into free space. |
Why not use GT2 en lieu of a Z-axis threaded rod or leadscrew? | A lead screw can be fitted with a anti-backlash nut to get past your concern about backlash. Generally, backlash is not that of a problem considering the weight of the carriage (e.g. Prusa), or whole bed (e.g. CoreXY or Ultimaker). Certainly if you're not using Z-hop, the Z screw will only advance in a single direction!
GT2 belts are always reinforced with fibres either glass, steel or some other fibre, all with their own (dis)advantages.
As far as I have learned is that lead screws are far better capable of transporting higher loads at the cost of speed. Also, a leadscrew will have more resistance so that if you unpower the steppers, the gantry will not plummet to the bed. |
Can toothbrush bristles be printed? | Actually last year a group did use a normal FDM printer to 3d print hair, brushed, etc. See the press release from Carnegie Mellon University
https://www.engadget.com/2015/10/29/3d-printing-hair-is-as-easy-as-using-a-hot-glue-gun/
http://technabob.com/blog/wp-content/uploads/2015/11/3d_printed_hair_by_Gierad_Laput_Xiang_Chen_Chris_Harrison_1.jpg
That said as far as I know you will not have access to this process, and is probably under a mountain of patents and other innovation killers.
Now how to do this outside of fancy software.
For a FDM printer the smallest nozzle I ever got was 0.1mm, it jammed instantly. One could print rows at this precision.
Now we have to move to something more advanced, such as DLP. Not the materials you want, but closer to the size you want. a formlabs can print at 25 microns. Which as a hair is 17 microns, you are "close enough" .. but a resin would be brittle and break. They do have other materials such as flexibles, but I am not familiar with them enough.
Also just going to mention. Tiny slivers of plastic is more likely to cut you than comb you. |
default axis steps per unit Makerbot replicator 2 | The Replicator 2 and 2x use 18 tooth GT2 pulleys, 1/16 microstepping, and 200 steps/rev steppers. That makes the proper steps/mm value 88.888889.
Note that Makerbot used ~88.56 steps/mm in their defaults, which is the value you get if you calculate from the belt+pulley pitch diameter from the Gates GT2 specs. But this is the value you use for calculating belt length required in a closed loop, not for steps/mm. The 88.88... number is more accurate. |
Re-homing Z: trouble resuming a print, what if I can't re-home? | Was able to make it happen, though I fear that I may have misjudged exactly where zero was, but we'll see in a few hours.
Anyway, for those who may come along later, here's what I figured out:
I already knew what my X, Y, and Z positions should be.
I kept the heat on for the bed so the print wouldn't shrink and break loose.
With the plate lowered and the print well out of harm's way, I homed X and Y. Then moved X10 Y0, a convenient place to work.
I set relative positioning on, G91, and moved Z up toward the print head until a business card was firmly clamped between the head and the plate.
I then set Z at zero, G92 Z0. Apparently Marlin (or Octoprint? idk) will let you print without homing Z as long as it knows where zero is.
I moved to print head to my desired X, Y, and Z positions and checked it against the model, which seemed to be right. I fear that I may have not zero'd down enough, so there will be a noticeable gap/weak spot, but I won't know that until I get more layers on than there are at this moment. I suppose if that's true, then I could just shave off the new layers and reset zero a couple of layers tighter.
The rest of the steps for resuming a print (editing the G-Code and what-not) are readily available in a dozen or more tutorials, so I won't cover that here.
Follow-up: yep, left too much room. I shaved off the new layers (having a kiridashi really helped there) and re-zero'd, but this time, after doing so I entered my initial Z value so that the nozzle touched the model and adjusted from there. Note that aesthetics aren't an issue: these are mechanical parts to all they need is to not snap in the middle. Having a seam there isn't a problem. |
Creating a heated chamber from a very large heated bed | Cardboard is fundamentally paper. Paper ignites - just as Ray Bradbury claims - at around 233 °C or 451 °F.
A thin heater wire usually glows red hot under operation. Not dark red, not blood cherry, not dark cherry, it's usually medium cherry to dark orange in operation, as one can easily tell by looking into a trusty toaster oven, which also shows us how powerful of a heating element this is.
In a typical toaster, heater wires are mounted in a metal frame and backed by some non-flammable material that looks like paper but is not. In a 2004 patent for such a method, they use a sheet of mica. But how hot is our wire? Well, under mains power, the wire of a toaster reaches 700-900 °C within seconds, heating the inner chamber of the toaster despite the slots to about 200-250 °C so quick that the bread roasts deliciously... until the heat controlling bimetal strip releases it up.
We operate these wires for at best a few minutes. In an electric forge, pretty much the same wires with a little more thickness to them are run at lemon glow for hours... and they are embedded in firebrick material to prevent them from igniting stuff touching them by accident!
How we know the temperature? Incandescence charts of course:
Conclusion
Paper or cardboard backing on heater wires is a bad idea. Mounting the heater wire to a metal frame that is as nonconductive as possible and without a chance that something accidentally touches it is needed. Fiberglass or rock material would serve as a good carrier, an aluminium block with radiators would give a good heating-to-surface mix, especially is precautions are taken to make the wire surface nonconductive.
Even using wood close to such a heater is a gamble I would not want to take. It's also best to mount the heating element into an enclosure that can't touch the wooden outer chamber and that stays put if the enclosure is opened or removed. |
How specific can I get setting layer heights? | Layer hight can be theoretically anything as long as it fits into these ranges:
it needs to be at least one step for the Z-Axis motor to be physically possible
it should be at max 3/4th of your nozzle diameter to create an adhesion surface
In praxis, the lowest setting for layer height due to physical limitations of the extrusion systems is around 0.05 mm.
Also, not any number is possible, it is dependant on the Z-axis system: since one or one-half step is the smallest rotation that the stepper can achieve, the raise that belongs to this partial rotation is how much one and the next layer height can be together. The step limitation usually is of no concern though, unless one has a very steep lead screw.
As a rule of thumb: doubling layer thickness results in almost half the print time. 0.1, 0.2 and 0.3 mm are common because they allow easy checking the printed accuracy in Z-axis with calipers. |
commandline remeshing and simplifying tools | Netfabb came up in a number of suggestions that I found. Take a look at 3D Printing Under Arch Linux, which offers a number of solutions for Linux, including:
OpenSCAD
FreeCAD
Blender
NetFabb
Slic3r
Netfabb
Taking the section on Netfabb:
netfabb Private will provide you with a fully-loaded tool-box to do all this and much more:
fix complicated file errors
make late design changes
combine parts into one
hollow out solid parts into a shell
smooth the mesh to improve surfaces
reduce file size
Installing it:
netfabb is not at official repository so you must install it from AUR
using yaourt
$ yaourt netfabb-basic
$ netfabb-basic
However, it is unclear as to whether there is a cmd line interface. It could be worth reading Problem in Netfabb Command Line execution and in particular the documentation on Automating Netfabb - although this seems to suggest that command line functionality is only available to users with Ultimate subscription and not Basic:
From Reducing Size of STL file to upload to shapeways, this post (#10) states:
You can also decimate the model in Netfabb easily - Right click > Level of Detail > Choose something like 50%... or more or less depending on the level of detail needed.
Although, this isn't available in Basic subscription, or so says post #12.
Meshmixer
I also came across this, non-command line option.
From How to Reduce the File Size of .STL and .OBJ 3D Models
For example, you can give the free 3D modeling program Meshmixer a try. Choose the ‘Select’ tool from the menu on the right, then double-click on your model (this will select your entire mesh). Then click on the ‘Edit…’ option and select ‘Reduce’.
You can now reduce the polygon count using several options. The best working choice is to use the percentage tool. It will immediately tell you by how much percent your file size (and polygon count) has been reduced.
Available for Windows and OSX.
From Reducing Size of STL file to upload to shapeways, this post (#8) states:
The meshlab tutorial is exactly what you want to do - the options described help preserve the overall geometry of the model while reducing the poly count. Start with a target of 150,000 polys and work up until the desired detail is retained, bearing in mind the printability I mentioned above.
Also, by way of Quora, How do I reduce the size of an STL file? Does it affect the CAD coordinates for 3D printing?, I also found this Meshmixer tutorial: Easy Guide on How to Repair Your .STL Files. Under the section Problem 4: File Size is too Large for 3D Printing are the following steps:
The last problem that you may stumble upon, particularly as your
objects get larger and their complexity increases, is that the
resulting .stl file size is too large for your 3D printer or 3D
printing service of choice. Third party printing services like
i.materialise or Shapeways have upload limits so you may have to
change the size of your file (100 MB limit at imaterialise and 64 MB for
Shapeways). It’s important to reduce the mesh without sacrificing the
quality, which I should say at the outset is best done at the initial
export from your CAD software to minimise any distortion. My number one quick
tip is to make sure in the export settings that your saving in
‘Binary’ format, not ‘ASCII’ – it’s a significantly more compact file
format for .stl’s without losing quality). However if it’s not an
option, Meshmixer can be really useful to quickly reduce file size and
visualise how this will affect the quality of your surfaces.
For this example I’m going to return to the leaf from Problem 1, which
as a .stl file is 30.757 MB – pretty large for such a small design and
definitely over the limit for someone’s email if I wanted to send it.
So let’s look at how we can get this under 10 MB.
Select the entire model by pressing Ctrl + A on your keyboard to select all.
Go to Edit>Reduce to bring up the options to reduce the mesh. Essentially the file size of a .stl is directly related to the number
of polygons (triangles) that make up all of the surfaces. We will use
the Reduce tool to reduce the quantity of polygons, therefore reducing
file size.
The menu will give you a range of options to reduce the size of your file. The simplest is to use the default Percentage option,
allowing you to determine the final file size you want rather than
needing to know exactly how many triangles will be used to make up the
file. In this case I have calculated that in order to get my original
.stl file below 10 MB, I need a reduction of 70 % or more. As you
adjust the values, you will get a preview of what your model will look
like, allowing you to control how much reduction you employ before the
model becomes too distorted and ‘pixelated.’
Once you’re happy and have accepted the changes, you can click on the Export button or go to File>Export to save this new reduced .stl
file. As you can see in the comparison below, there is almost no
perceptible difference in the quality of the 30.757 MB file compared
with the final reduced version at 9.362 MB. I have simply increased the
size of the triangles, minimising the quantity needed to create the
same volume. Your regular 3D printer, or even a highly accurate SLS
printer will never even know the difference in a change like this.
It’s only if you really go crazy with the file reduction that you
might start to notice them in your final print (and in fact many of
the low-poly models which are popular on Pinshape can be produced
using this technique – start with a detailed .stl of the object you
want, and then just keep reducing the number of triangles!).
Blender
From Reducing Size of STL file to upload to shapeways, this post (#8) states:
Another option is to import the model into Blender, then use the Remesh or Decimate modifiers. Blender can be tough to get into initially but there are plenty of introduction tutorials on YouTube to help.
Blender does offer a Command Line Interface (CLI), see Command Line Arguments:
Usage: blender [args …] [file] [args …]
However, there doesn't appear to be an option for Remesh or Decimate. Although take a look at Is it possible to decimate or remesh via the CLI?:
You can specify a Python script to run, see the Python Options section of the documentation you linked. In this script, you can import the model, add the modifiers, apply them and export the result.
Related - Is it possible to automatically simplify geometry in blender without losing object shape?, which has this comment:
Is there a way to execute this modifier from command line and export the result?
which then leads to Can modifiers be computed and applied from the command line?
You'd have to use bpy.ops.object.modifier_apply, e.g.
import bpy
for obj in bpy.context.scene.objects:
bpy.context.scene.objects.active = obj
count = 1
length = len(obj.modifiers)
while obj.modifiers:
name = obj.modifiers[0].name
print("%s: Applying %s (%d/%d)" % (obj, name, count, length))
bpy.ops.object.modifier_apply(modifier=name)
count += 1
print("All done.")
For info on how to to run the above, see here.
Meshlab
Seeing as you are already using Meshlab, it might be worth reading Executing meshlab from commandline reduce faces of a mesh iteratively (awesomebytes/reduce_faces.py)
#!/usr/bin/env python
import sys
import os
import subprocess
# Script taken from doing the needed operation
# (Filters > Remeshing, Simplification and Reconstruction >
# Quadric Edge Collapse Decimation, with parameters:
# 0.9 percentage reduction (10%), 0.3 Quality threshold (70%)
# Target number of faces is ignored with those parameters
# conserving face normals, planar simplification and
# post-simplimfication cleaning)
# And going to Filter > Show current filter script
filter_script_mlx = """<!DOCTYPE FilterScript>
<FilterScript>
<filter name="Quadric Edge Collapse Decimation">
<Param type="RichInt" value="1448" name="TargetFaceNum"/>
<Param type="RichFloat" value="0.9" name="TargetPerc"/>
<Param type="RichFloat" value="0.3" name="QualityThr"/>
<Param type="RichBool" value="false" name="PreserveBoundary"/>
<Param type="RichFloat" value="1" name="BoundaryWeight"/>
<Param type="RichBool" value="true" name="PreserveNormal"/>
<Param type="RichBool" value="false" name="PreserveTopology"/>
<Param type="RichBool" value="false" name="OptimalPlacement"/>
<Param type="RichBool" value="true" name="PlanarQuadric"/>
<Param type="RichBool" value="false" name="QualityWeight"/>
<Param type="RichBool" value="true" name="AutoClean"/>
<Param type="RichBool" value="false" name="Selected"/>
</filter>
</FilterScript>
"""
def create_tmp_filter_file(filename='filter_file_tmp.mlx'):
with open('/tmp/' + filename, 'w') as f:
f.write(filter_script_mlx)
return '/tmp/' + filename
def reduce_faces(in_file, out_file,
filter_script_path=create_tmp_filter_file()):
# Add input mesh
command = "meshlabserver -i " + in_file
# Add the filter script
command += " -s " + filter_script_path
# Add the output filename and output flags
command += " -o " + out_file + " -om vn fn"
# Execute command
print "Going to execute: " + command
output = subprocess.check_output(command, shell=True)
last_line = output.splitlines()[-1]
print
print "Done:"
print in_file + " > " + out_file + ": " + last_line
if __name__ == '__main__':
if len(sys.argv) < 3:
print "Usage:"
print sys.argv[0] + " /path/to/input_mesh num_iterations"
print "For example, reduce 10 times:"
print sys.argv[0] + " /home/myuser/mymesh.dae 10"
exit(0)
in_mesh = sys.argv[1]
filename = in_mesh.split('/')[-1]
num_iterations = int(sys.argv[2])
folder_name = filename.replace('.', '_')
tmp_folder_name = '/tmp/' + folder_name + '_meshes/'
print "Input mesh: " + in_mesh + " (filename: " + filename + ")"
print "Num iterations: " + str(num_iterations)
print "Output folder: " + tmp_folder_name
try:
os.mkdir(tmp_folder_name)
except OSError as e:
print >> sys.stderr, "Exception creating folder for meshes: " + str(e)
exit(0)
for it in range(1, num_iterations):
if it == 1:
out_mesh = tmp_folder_name + folder_name + "_it" + str(it) + ".dae"
reduce_faces(in_mesh, out_mesh)
else:
out_mesh = tmp_folder_name + folder_name + "_it" + str(it) + ".dae"
reduce_faces(last_out_mesh, out_mesh)
last_out_mesh = out_mesh
print
print "Done reducing, find the files at: " + tmp_folder_name
Also
here are some updates for python 3.6.2: https://gist.github.com/tylerlindell/7435ca2261e7c404ccc1241f18e483aa |
"configuration files" issue when trying to run Ultimaker Cura on RPI 4 | The regular version of Cura has system requirements that the Raspberry Pi cannot meet, including the GPU required to run it.
However, it looks like you can try an experimental version of Cura here. |
How do I calibrate temperature control in a Repetier firmware? | When applying a voltage directly to the heater element, it should heat up very fast. You could measure the resistance of the heater element. A 12 V heater element of 40 Watt should have a resistance of about 3.6 Ω; you could test that with a multitester. If the resistance is a lot higher, the heater cartridge could be a 24 V heater element. Operating such element on 12 V would imply that the power output would only be 10 W. Heating of the hotend is much slower at low power output.
Incorrect PID settings can indeed prevent the heater to heat up fast enough to trigger errors during tuning. The trick is to heat up the element a little prior to performing the tuning, or upload new PID settings to the firmware prior to tuning.
PID settings can be set not only by uploading a modified firmware version (configuration file), but also by sending commands to the printer over serial terminal connection (e.g. Repetier host, OctoPrint, Pronterface, etc.) or stored in a G-code file and executing it ("printing"). The G-code command to set the PID values is M301. The relevant parameters are:
Hnnn heater number Pnnn proportional (Kp) Innn integral (Ki) Dnnn
derivative (Kd) |
3D printed mold techniques for long and short term usage | Yes, this is very broad. That said...
For high detail you want SLA. i.e. jewelry. If you just want a prototype of a mold, you can do a standard FDM style printer (95% of printers are FDM, and that number is a guess)
Really, you should be asking what material you need for your mold, but you can open a second question for that.
Do more research on injection molding. There is a great deal of information on how molds are made, i.e. How It's Made Plastic injection molds.
You will see there is a vast difference between a plastic, or silicon, mold and an injection molding machine. You are thinking that injection molding as a single mold, when it is really it is a system composing of several pieces of heavy duty machinery that can pump out hundreds of items a day automatically. However, it usually starts at 20k USD for the tooling for injection molding. Your costs could be a fraction of that or could be several times that. This is just a generality. So, if you are making 100 units you won't want to go down that route. For 10,000 units, on the other hand, it would be acceptable. |
Second layer of ASA print 'runs' | I've now worked out the reason for the plastic 'running' on the second layer. It is because of the way that I prepared the G-Code file, using the option in Slic3r which lets you alter the layer height for certain layers in the print. The problem with this appears to be that there are other settings which are not altered at the same time for the thicker layers. So the first half of the file ends up printed using the settings that work for a 0.05 mm layer height, but with the layer height set to 0.2 mm.
I have prepared a new file by rendering two G-Code files, one using the Prusa settings for a 0.2 mm layer height, and the other using a 0.05 mm layer height. I then merged these manually by looking for the layer change at 2.8 mm in both files and appending the end of the 0.05 mm file to the start of the 0.2 mm file. (I found the layer changes by using Slic3r's custom G-Code option, and adding custom comments into the G-Code which included the layer number and height on every layer change). Then I inserted an M600 'Change filament' instruction at a height of 3.05 mm, where the writing begins.
When I print this file, the problem with the plastic running together on the second layer disappears. |
Dataset Card Creation Guide
Dataset Summary
We automatically extracted question and answer (Q&A) pairs from Stack Exchange network. Stack Exchange gather many Q&A communities across 50 online plateform, including the well known Stack Overflow and other technical sites. 100 millon developpers consult Stack Exchange every month. The dataset is a parallel corpus with each question mapped to the top rated answer. The dataset is split given communities which cover a variety of domains from 3d printing, economics, raspberry pi or emacs. An exhaustive list of all communities is available here.
Languages
Stack Exchange mainly consist of english language (en).
Dataset Structure
Data Instances
Each data samples is presented as follow:
{'title_body': "Is there a Stack Exchange icon available? StackAuth /sites route provides all the site's icons except for the one of the Stack Exchange master site.\nCould you please provide it in some way (a static SVG would be good)?",
'upvoted_answer': 'Here it is!\n\nDead link: SVG version here\nNote: the same restrictions on this trademarked icon that apply here, also apply to the icon above.',
'downvoted_answer': 'No, the /sites route is not the right place for that.\n\n/sites enumerates all websites that expose API end-points. StackExchange.com does not expose such an endpoint, so it does not (and will not) appear in the results.'}
This particular exampe corresponds to the following page
Data Fields
The fields present in the dataset contain the following informations:
title_body
: This is the concatenation of the title and body from the questionupvoted_answer
: This is the body from the most upvoted answer
Data Splits
We provide multiple splits for this dataset, which each refers to a given community channel. We detail the number of pail for each split below:
Number of pairs | |
---|---|
gaming | 82,887 |
dba | 71,449 |
codereview | 41,748 |
gis | 100,254 |
english | 100,640 |
mathoverflow | 85,289 |
askubuntu | 267,135 |
electronics | 129,494 |
apple | 92,487 |
diy | 52,896 |
magento | 79,241 |
gamedev | 40,154 |
mathematica | 59,895 |
ell | 77,892 |
judaism | 26,085 |
drupal | 67,817 |
blender | 54,153 |
biology | 19,277 |
android | 38,077 |
crypto | 19,404 |
christianity | 11,498 |
cs | 30,010 |
academia | 32,137 |
chemistry | 27,061 |
aviation | 18,755 |
history | 10,766 |
japanese | 20,948 |
cooking | 22,641 |
law | 16,133 |
hermeneutics | 9,516 |
hinduism | 8,999 |
graphicdesign | 28,083 |
dsp | 17,430 |
bicycles | 15,708 |
ethereum | 26,124 |
ja | 17,376 |
arduino | 16,281 |
bitcoin | 22,474 |
islam | 10,052 |
datascience | 20,503 |
german | 13,733 |
codegolf | 8,211 |
boardgames | 11,805 |
economics | 8,844 |
emacs | 16,830 |
buddhism | 6,787 |
gardening | 13,246 |
astronomy | 9,086 |
anime | 10,131 |
fitness | 8,297 |
cstheory | 7,742 |
engineering | 8,649 |
chinese | 8,646 |
linguistics | 6,843 |
cogsci | 5,101 |
french | 10,578 |
literature | 3,539 |
ai | 5,763 |
craftcms | 11,236 |
health | 4,494 |
chess | 6,392 |
interpersonal | 3,398 |
expressionengine | 10,742 |
earthscience | 4,396 |
civicrm | 10,648 |
joomla | 5,887 |
homebrew | 5,608 |
latin | 3,969 |
ham | 3,501 |
hsm | 2,517 |
avp | 6,450 |
expatriates | 4,913 |
matheducators | 2,706 |
genealogy | 2,895 |
3dprinting | 3,488 |
devops | 3,462 |
bioinformatics | 3,135 |
computergraphics | 2,306 |
elementaryos | 5,917 |
martialarts | 1,737 |
hardwarerecs | 2,050 |
lifehacks | 2,576 |
crafts | 1,659 |
italian | 3,101 |
freelancing | 1,663 |
materials | 1,101 |
bricks | 3,530 |
cseducators | 902 |
eosio | 1,940 |
iot | 1,359 |
languagelearning | 948 |
beer | 1,012 |
ebooks | 1,107 |
coffee | 1,188 |
esperanto | 1,466 |
korean | 1,406 |
cardano | 248 |
conlang | 334 |
drones | 496 |
iota | 775 |
salesforce | 87,272 |
wordpress | 83,621 |
rpg | 40,435 |
scifi | 54,805 |
stats | 115,679 |
serverfault | 238,507 |
physics | 141,230 |
sharepoint | 80,420 |
security | 51,355 |
worldbuilding | 26,210 |
softwareengineering | 51,326 |
superuser | 352,610 |
meta | 1,000 |
money | 29,404 |
travel | 36,533 |
photo | 23,204 |
webmasters | 30,370 |
workplace | 24,012 |
ux | 28,901 |
philosophy | 13,114 |
music | 19,936 |
politics | 11,047 |
movies | 18,243 |
space | 12,893 |
skeptics | 8,145 |
raspberrypi | 24,143 |
rus | 16,528 |
puzzling | 17,448 |
webapps | 24,867 |
mechanics | 18,613 |
writers | 9,867 |
networkengineering | 12,590 |
parenting | 5,998 |
softwarerecs | 11,761 |
quant | 12,933 |
spanish | 7,675 |
scicomp | 7,036 |
pets | 6,156 |
sqa | 9,256 |
sitecore | 7,838 |
vi | 9,000 |
outdoors | 5,278 |
sound | 8,303 |
pm | 5,435 |
reverseengineering | 5,817 |
retrocomputing | 3,907 |
tridion | 5,907 |
quantumcomputing | 4,320 |
sports | 4,707 |
robotics | 4,648 |
russian | 3,937 |
opensource | 3,221 |
woodworking | 2,955 |
ukrainian | 1,767 |
opendata | 3,842 |
patents | 3,573 |
mythology | 1,595 |
portuguese | 1,964 |
tor | 4,167 |
monero | 3,508 |
sustainability | 1,674 |
musicfans | 2,431 |
poker | 1,665 |
or | 1,490 |
windowsphone | 2,807 |
stackapps | 1,518 |
moderators | 504 |
vegetarianism | 585 |
tezos | 1,169 |
stellar | 1,078 |
pt | 103,277 |
unix | 155,414 |
tex | 171,628 |
ru | 253,289 |
total | 4,750,619 |
Dataset Creation
Curation Rationale
We primary designed this dataset for sentence embeddings training. Indeed sentence embeddings may be trained using a contrastive learning setup for which the model is trained to associate each sentence with its corresponding pair out of multiple proposition. Such models require many examples to be efficient and thus the dataset creation may be tedious. Community networks such as Stack Exchange allow us to build many examples semi-automatically.
Source Data
The source data are dumps from Stack Exchange
Initial Data Collection and Normalization
We collected the data from the math community.
We filtered out questions which title or body length is bellow 20 characters and questions for which body length is above 4096 characters.
Who are the source language producers?
Questions and answers are written by the community developpers of Stack Exchange.
Additional Information
Licensing Information
Please see the license information at: https://archive.org/details/stackexchange
Citation Information
@misc{StackExchangeDataset,
author = {Flax Sentence Embeddings Team},
title = {Stack Exchange question pairs},
year = {2021},
howpublished = {https://huggingface.co/datasets/flax-sentence-embeddings/},
}
Contributions
Thanks to the Flax Sentence Embeddings team for adding this dataset.
- Downloads last month
- 504