YouMagine

Information Comments Technical Files

Comments

You need to be signed in before commenting.

dintid

about 8 years

Sprocket Teeth? Not a good tactics if you want help... The files are on his site, if you read the descriptiopn.

Alessandro Romagnoli

about 8 years

Hello sprocket teeth! I'm sorry but I'm interested in this that I feel better! you may not have the STL of this ??? Thank you ! Alex Elvis3D

dintid

about 8 years

Alessandro: they are on his website, but the design is redacted. New model: https://www.youmagine.com/designs/ninjaflex-geared-ultimaker-2-feeder

Alessandro Romagnoli

about 8 years

Hello ! Where are the STL file to download I do not see them ?? Alex Elvis3D

dintid

almost 9 years

Any plans on making a version with a fitting that non-US people can buy? Also wondering at the use of MXL betls instead of GT2 which are somewhat cheaper? First one is a pressing one, while the second question is just nice to know.

Anonymous

almost 9 years

The problem of enhanced friction could be reduced by using Igldur filament http://www.igus.de/wpck/12019/iglidur_TriboFilament at least partly - however it's expensive compared to ABS ... however it would come close to a bearing.

stabilo60

about 9 years

<p>Well, I'm sorry, I found the files on his web site :-) Thanks :-)</p>

stabilo60

about 9 years

<p>I can't download :-( No button for that :-(</p>

tetnum

about 9 years

<p>where are the files?</p>

david ganon

about 9 years

<p>cant download !!!!! no button !!!!</p>

Stu_le_brew

almost 10 years

I recently printed this feeder and will be trying it soon, I am convinced it will work well particularly with flexible filament. I was able to source the belts via alfa-tech3D.com who are based in Denmark so much easier for shipment than McMaster they provided very good service for me

Spirit

almost 10 years

<p>The measurements that you listed for the belts matches the ones that I am currently using. I'm not sure what else could be causing the problem if all the dimensions are correct. Have you tried measuring the distance between the belts when the extruder is fully assembled?</p>

KorneelB

almost 10 years

sure, i've done my measurements in MM. the belt fits on "tight" on the gears, so i don't think that's the problem. it fits well and turns smoothly. the diameter is from tooth-outside 1,13MM and from between-tooth to outside 0.68 MM.

Spirit

almost 10 years

@KorneelB Can you measure the diameter of the belt gears along with the thickness of your belts? If either one is off it would affect the spacing. <div><br></div><div> @Ovidio Any design with a large number of retractions with retractions always enabled (ie turn combing off) should show off the strength of the design. Most of my current prints have close to 20 retractions per mm of filament (impossible to print on the original extruder). Regardless of how many retractions, the belts should prevent the filament from stripping. As the filament shouldn't ever partially strip, you should see much better extrusion consistency as well. <div><br></div><div>There is an issue with the currently posted design (rev. E) however that may cause underextrusion if the intermediate gear is not tightened down enough. I've fixed this in my new revision which I'll update here in the near future. I've described the problem more extensively in my blog post on the newer revision.</div></div>

Ovidiu

almost 10 years

I reprinted all the parts in PLA except the motor gear which is still in ABS. The PLA helical gears fit much tighter on the bearings, which have to be pushed slightly now to be able to get in. This seems to solve the loose belt problem I had previously. I will mount the new mechanism and report back with how its working.<div><br></div><div>Spirit, is there a piece that can be printed that shows the strength of your design?</div><div><br></div>

KorneelB

almost 10 years

Hi Spirit; understood. i don't think it's the coating on the outside, since the belts don't actually touch the material that well.. it seems the belt might not be thick enough..

Spirit

almost 10 years

Its possible but I would like to avoid modifying the design files for each individual user. I would suggest checking the belt gear diameter (should be ~15.1mm) and/or removing more of the coating on the outer surface of the belt for a better grip.

KorneelB

almost 10 years

spirit; in this revision, is there a way to make the belts a bit closer to eachother? it seems that they are not getting enough grip on my filament.

Spirit

almost 10 years

The design files are in my blog post linked in the information section. A new revision will be posted shortly.

fabi580

almost 10 years

I would be very interested in the Inventor Files to adjust the design for metric belts. Is there a possibility?

Spirit

almost 10 years

This design should extrude no better than |Robert|'s as it uses the same underlying components. What it does do however, is improve reliability and consistency of prints. The extrusion test, while decent for benchmarking maximum extruder speed, doesn't really show the strength of this design. If you tried printing something with a large number of retractions (ie, turn off combing), you'll see that this extruder (properly assembled) will never fail part way through a print. <div><br></div><div>For reference, I've gone >200 hours now without a single underextrusion occurrence.</div>

EldRick

almost 10 years

I got adventurous and printed the herringbone-geared version. On the positive side, dimensions were spot-on, requiring little cleanup, and the design is a very nice piece of work. It lapped itself in quickly and ran smoothly. <br> Unfortunately when I installed it I saw no advantage over |Robert|s feeder on the UM2 extrusion test, so I reinstalled that and proceeded to produce the best UM robot my printer has ever done. <br>I'm hoping for better results from the Flex3Drive direct-drive mod.

KorneelB

almost 10 years

I just received a bunch of 40MXL (or 400 MXL according to my supplier) belts and they fit perfectly. I'm still waiting for the ball bearings but these seem to be perfect. I think this confirms that Ovidu might have bought \received the wrong belts.

Spirit

almost 10 years

The 26.5mm distance between the gears (vertically) is correct, but I measured the diameters of the belt gears on my prints to be slightly wider at 15.1mm. Your slightly smaller gears will probably affect the performance a bit as the spacing of the belts that grip the filament will be slightly larger. The (horizontal) spacing between gears is 19mm, so a diameter of 15.1mm with a belt thickness of 0.5mm leads to a gap of 2.9mm. Your gear diameter of 14.6mm would result in a gap of 3.4mm, larger than the diameter of the filament itself. <div><br></div><div>The small diameter difference along with the extra 4.36mm from the belt is certainly enough to account for the extra slack visible in your pictures. I would definitely contact McMaster to see if they can get you a properly sized one. </div><div><br></div><div>You may also want to try increasing the extrusion multiplier, decreasing the print speed, and lowering the layer height to see if you can get the gears printed to the dimensions that I gave above. Unfortunately I cant account for this sort of variation between printers in the design. </div>

KorneelB

almost 10 years

I just printed mine as well, extra carefull (don't have my belts yet but ordered somewhere else since they could not deliver in NL for a good price). center of the holes for the gears (outside to outside) is 29.8MM center of the holes for the gears (inside to inside) is 23,7 mm center of the holes for the gears is 26.5 MM measured frmo the center. I'm thinking your belts are the problem here. also, if you buy 4 inch belts, they should not be 4.1715... i would contact mcmaster..

Ovidiu

almost 10 years

The outside circumference of the belts is 4.1715 in, obviously longer than what advertised. The distance between the centers of the gears moving a belt is about 26.5 mm. The diameter of the gears moving the belt varies between 14.52 and 14.59 mm, surprisingly good for plastic. According to my calculations, the inside length of an ideal belt using the gears I have would be approx 98.7mm. With the outside belt at 105.96mm, there's a lot of slack. If the belts would be exactly 4 in, the outside length would decrease by 4.36mm, which reduces the slack by 60%!

Spirit

almost 10 years

<p>The belts that I linked to on my post is the exact belt that I used. Can you try measuring the outside diameter of the one that you have? If you could also measure the outer diameter of the gears where the belts touch as well as the spacing in between the center of the gears, that would be helpful as well.</p><p>The issue may also be with plastic shrinking or warping when cooling. Looking at your pictures it seems that some of the pieces are a bit warped.</p>

Ovidiu

almost 10 years

The belt is exactly the one you indicated, I got it from McMaster. They must have changed the belts :(

Spirit

almost 10 years

The slight offset is not an issue. If you look at my assembly you'll see that the same offset is there. The offset should not be affecting the performance in any way, as there should be enough contact (~3mm) between the teeth of the gears for adequate torque transfer. The motor is most likely skipping not because of the gears, but because there is too much rotational friction when fully assembled. Your belts are also longer than 4". I can't say that it has a definite effect on the performance but I would imagine that the extra slack would make a difference when the motor reverses during retractions.

Ovidiu

almost 10 years

Here are some pictures: https://www.flickr.com/photos/ovidiup/sets/72157646514961767/ You can see the motor gear being slightly offset from the intermediate gear. I put a washer in between the motor and the gear to maintain the distance. Unfortunately I cannot pull out the stock drive gear, the screw that fixes it to the shaft is already at the tip of the axle. The offset between the drive gear and the intermediate gear causes skipping to occur when the filament is moving at high speeds (e.g. when it's being loaded).

Spirit

almost 10 years

1. 2mm short? Are you talking about the knurled pieced that sits on-top of the motor gear? The knurled piece has no choice but to stick out a bit due to the thickness of the bottom piece. The motor gear isn't suppose to sit flush against the face of the stepper motor. A picture would be helpful here. <div><br></div><div>2. Tension on the belts is a non-issue. The gap between the belts is far more important in terms of how well it can grip the filament. </div><div><br></div><div>3. The knurled portion increases the accuracy factor of the design. If you read my development history in my detailed post I do give reasons for using the knurled piece in addition to the belts.  </div><div><br></div><div>4. If you cant easily turn the gears by hand, they are too tight. Try adding some lubrication or print at 0.3mm nozzle size to increase the gap between moving parts.</div>

Ovidiu

almost 10 years

<p>I just printed and installed the extruder on my Ultimaker 2. There are few problems:</p><p>1. The motor gear is about 2 mm short.</p><p>2. The belts are not tensioned sufficiently to be able to pull on the filament.</p><p>3. The knurled portion of the original gear mechanism that sits on the motor's axle is still pushing on the filament, which doesn't solve the original problem you were trying to solve.</p><p>4. The new motor gear skips steps. The latter might be caused by too much tightening of the screws holding the 4 gears in place, I'll loosen them more.</p><p>It'd be great if you could publish a newer version of the motor gear that's 2 mm longer at the bottom. Any thoughts on how can the belts be tensioned more in the current design?</p>

KorneelB

almost 10 years

thanks! I have tried finding a shipper in the US to ship this to me, and they ended up charging me up to a 100 USD to ship me 9 USD of belts.. ended up with AMerican Metric, those guys actually make those belts :) http://www.ametric.com/p-47881-400-mxl-025.aspx shipping is 8 USDm and the belts are 1.85 USD a piece.. nowhere cheaper :)

Spirit

almost 10 years

<p>The only two I could think of would be to 1. get someone to ship you a belt from the states or 2. find a belt of similar length and modify the design a bit to use it.</p>

KorneelB

almost 10 years

I'm looking high and low, it seems I cannot find the 40MXL belt in THe Netherlands.. any suggestions?

Charles_design

almost 10 years

What speed have you been printing at?

Spirit

almost 10 years

I've been printing retraction heavy prints with this design for over 30 hours now and I have yet to see a single underextrusion occurrence. The ratio is 1:1 so nothing needs to be changed in whatever software you're using for slicing. Link to the parts and design files can be found in my more informative post at the very bottom (there is a link to it on the information tab).

Charles_design

almost 10 years

That's genius! Can you confirm you still haven't had any under-extrusion? As I understand it the filament is in contact with both the belt and bobbed shaft so the ratio remains 1:1? Could you also provide me with a link to purchase the parts? Thanks!<div><br></div><div>Edit: Are these the belts used in this design?</div><div>Would you be able to send the SW assembly file over? :P</div>

Spirit

almost 10 years

2.85mm should be nearly identical to 3.00mm when it comes to filament diameters. ~0.15mm should be within the elasticity range of the belts, it should just fit a bit tighter.

clews

almost 10 years

Did you try it with 3.00mm? I use both with my Ultimaker2... The Design of this extruder is really interesting, but is there enough room for 2.85 and 3.00mm filament?

Spirit

almost 10 years

The design only works with 2.85mm filaments.

Vumelyan

almost 10 years

Since this was deigned for the ultimaker it should also work with 2.85mm filament?

Spirit

almost 10 years

Should certainly be possible as they're both use bowden tubes. The design can be modified for 1.75mm but it would take a bit of work.

Vumelyan

almost 10 years

Can this be used with the Kossel/Rostock style 3D pinters...looks like it just needs a plate to mount to and can it be modified for the 1.75mm filament?