1 (edited by Kun1 2015-07-28 09:52:23)

Topic: Filament winder is jerky, causing filament vibration.

Hi guys

This is my first post here. Hope this is the right place to post this question :-)

I just set up my extruder together with a filawinder and am starting to extrude filament. I am really struggling getting a reliable filament diameter - the filawinder is pulling at the filament realy erratically and sometimes the filament starts vibrating a lot (sometimes it jumps up and down with an amplitude of 2-3 cm (this is also why I increased the length of the sensor guide bars). This vibration in turn (I think) causes a lot of variation in the filament diameter.

The setup: I can't mount the extruder and winder on a wall (which I think would be the best option). So instead I have put them both onto a small table (each) on top of my lab-bench, and have put the sensor on the lab-bench below. The extruder is also angled a bit. The filament forms a relatively smooth loop from the extruder, through the sensor area and then to the winder. I am attaching some pictures to show what I mean. I don't think I'm allowed to post a video, but let me know and I can send you the video I made during extrusion, showing the vibration.

http://soliforum.com/i/?GCyS2Zd.jpg
http://soliforum.com/i/?TdgLXAn.jpg


I think I need to 'hack' the filawinder software to add some damping. I've downloaded the software, and have looked a the code - there seem to be a lot of different settings that I could tweak - the pulling speed, the amount of sensor readings per second, and the amount of time in between readings. 

So, my question to you guys: have any of you experienced similar vibration while extruding and using a filawinder? And how did you solve it? Which of the filawinder settings should I adjust first and what are some good settings to aim for?

Thanks in advance for your advice - really appreciate any help that you can give me.

2

Re: Filament winder is jerky, causing filament vibration.

Here is a later version of the firmware- https://www.dropbox.com/s/1mg6rnx1rhndo … 1.zip?dl=0

This one has damping.  You can adjust it by changing the number in this bit at line 295

  if ( (puller_speed - puller_speed_old) > 25) puller_speed = puller_speed_old + 25;
  if ( (puller_speed - puller_speed_old) < -25) puller_speed =  puller_speed_old - 25;

For vertical, you could get another pair of Lack tables, screw the Filabot and Filawinder to the surface and then set them on their sides on top of the ones you already have.  You could print a bracket or large clip of some kind to hold the down the legs of the sideways tables so they don't fall over.

3

Re: Filament winder is jerky, causing filament vibration.

How does the Filabot do without a winder? It is possible the filament diameter variance isn't from the Filawinder, but rather the material (if not dried properly) or the Filabot itself.

That also looks like a *very* tight PTFE loop, which may be overloading the winder.

EDIT: Actually, is that even a Filawinder? It looks more like Noztek's winder.

4

Re: Filament winder is jerky, causing filament vibration.

Awesome, thanks so much IanJohnson. Will try the damping function tomorrow. Hope that solves it! :-)

5

Re: Filament winder is jerky, causing filament vibration.

@Elmoret - I think that the extruder does well without the winder. If I don't use the winder and just let the filament coil to the floor I get quite decent filament. But obviously, I'd prefer to have the auto-winding feature working...spooling loose filament onto a reel by hand isn't all that much fun!

Hmm, good question about whether that is a Noztek winder. I have to check (I bought 2 different assembled winders online, don't remember which one I am using right now). But, I thought that the Noztek winder was simply an assembled Filawinder? The design for the Filawinder is opensource, to my knowledge. The guys at Noztek simply took this design, assembled a few winders and are now selling it to people (like me) who don't have the time to assemble their own winder (or are too lazy to do so ;-)). I could be wrong though...still learning a lot every day about this stuff...

About the tightness of the PTFE loop. You are right, it IS very (!) tight. I thought it was supposed to be like that....am I wrong? I can always replace the loop with a tube that has a wider opening. Would that help? I thought that the purpose of the loop was to constrict the filament and 'give the winder something to pull on'.....? Curious about your thoughts.

Thanks again for your advice!

6

Re: Filament winder is jerky, causing filament vibration.

I'd test the extrusion to the floor to be certain, that eliminates variables. What variance do you see with the winder? How much without the winder?

Noztek does not compensate Ian for use of his design - they don't even properly credit him (proper accreditation is a link to the source). Sure, it is open source - but I'm just saying you shouldn't expect support as if you were actually using a Filawinder - nor should the title of this post include the word Filawinder, for the same reasons. Many things are different, for example the laser guide looks different. There's no guarantee the motor or firmware are the same either.

Yes, the PTFE tube's purpose is to provide tension, but that looks like quite a lot of tension. The motor may be struggling. Too much friction will also cause the jerking you describe. No need to get different tube, just make the loop larger.

Finally, yes you can post a video, just put a space in the link to get around anti spam.

7

Re: Filament winder is jerky, causing filament vibration.

Hi elmoret....thanks very much for explaining the issue with Noztek and the fact that they don't provide proper accreditation. I definitely think it's not cool of them to take the Filawinder design without giving proper credit (I was not aware of that, so apologies for my mistake). I'll have a look at what winder I am using and make changes to the post title if necessary. Thanks for your help in any case :-)

8

Re: Filament winder is jerky, causing filament vibration.

I uploaded the extrusion video to dropbox (in case it's interesting/useful :-))

https://www.dropbox.com/sh/o4xww16jkxm4 … zXWWa?dl=0

9

Re: Filament winder is jerky, causing filament vibration.

Yeah that's pretty extreme. Looks to me to be excessive friction in the PTFE tube and possibly some subpar firmware. Ideally the motor would spin at a more consistent speed.

10

Re: Filament winder is jerky, causing filament vibration.

I wasn't sure if it was Noztek from the photo since the laser mount was acrylic, but I guess they only did sheet metal for the main unit.  They do credit me, they just don't have a link to the Thingiverse page.  They are however, obligated to post the sources for their derivative.  I don't really care if they do however.  They made it prettier, but far more expensive.

I think the jerking is a combination of the loop and firmware.  They will likely be using a version of the firmware without dampening.  Also that loop is very tight for 3mm.  Start with using the full length of the PTFE in the loop and tighten it from there if you find that the filament wants to pull away from the spool. 

Actually what is most likely happening is that they are using a motor that is too fast.  If the PID outputs a motor speed of 125 (out of 256), that is going to be much faster for a 100 rpm motor than a 33rpm motor and give you a lot of jerky motion.  The damping will help that.  It limits changes in motor speed to increments of 25 so it takes a little more time to rise or fall to the demanded speed.  Lowering that number will increase the damping since it will take more program loops to reach the new motor speed.

11

Re: Filament winder is jerky, causing filament vibration.

The PTFE loop looks pretty tight, a larger loop should unload the motor and make it run smoother.

12

Re: Filament winder is jerky, causing filament vibration.

Hi guys

I wanted to update you on my progress - your advice was extremely helpful. By increasing the loop size and uploading the firmware from IanJohnson (and decreasing the motor speed slightly) I could get the winding process under control. As you can see from the pictures below, the filament now moves smoothly through the sensor area and through the loop.

http://soliforum.com/i/?ExDNkF3.jpg
http://soliforum.com/i/?chlNqV1.jpg

I have also uploaded a video of the new process, in case of interest :-)

https://www.dropbox.com/s/6c7dpxhowgurx … g.MOV?dl=0

Thanks again to everyone for your help.