126

Re: Filament Winder

Lots of variables. Temperature, cooling, height, etc. If you need bigger, just go up a size - since it's drilled already, its no problem to drill up a size or two by hand. There's no one-size-fits-all number.

127

Re: Filament Winder

Hi Ian,
Just wondering if you got that PCB board back or any other developments.
Thanks

128

Re: Filament Winder

I did, but it didn't work very well.  I incorporated an arduino by adapting an arduino-on-a-breadboard tutorial to a pcb.  The arduino boards must do something different with power, because the guide servo kept freaking out, and nothing else would work unless the servo was disabled in firmware.

I'm doing another version using an Arduino Nano.  The Nano is an arduino on a tiny board that plugs into a couple of rows of female headers kind of like the stepper drivers on a Sanguinololu.  Then all I have to do is route traces around to connections for the motor, sensors, etc.

I changed the sensor to 4 photo cells in a row, and added a library from Pololu that has a line following function.  It can detect the position of the filament and output it as a number from 0 to 3000.  I feed that to a PID which controls the motor speed and tries to keep the filament at 1500, in the center.  This should be more steady than bouncing up and down between sensors.

I'm also trying to get rid of the puller motor.  I'm letting the sensor control the spool speed, and running the filament through a pinch roller in place of the puller motor.  Tension is adjusted by controlling how easily the pinch roller turns.  I'm still getting the parts tweaked and put together, but if it works well enough I can drop 1 motor and the tension control arm.  That knocks about $15 off the cost and makes everything a little simpler. 

Getting rid of the puller does make it a little harder to use.  With the puller motor, once you get the end of the filament into the puller you can fumble around with getting the filament through the calipers, the guide and onto the spool without messing up the loop.  You can also let it run until it is stable before putting it onto the spool, since the first foot or so is usually not very usable.  Without the puller you might need to start and stop the extruder a few times to keep it from getting ahead of you.  The Extrusion Bot, if it is running at 4ft/minute will definitely need starting and stopping.  Even at 9"/min I can feel like I am racing the clock to keep the filament from backing up while I guide the end through the winder.  It might be a good idea to run a marker on the first foot or so of filament on the spool to note that it might not be to tolerance,  kind of like the stripe that tells you a spool of receipt paper is almost empty.

129

Re: Filament Winder

Hi Ian,
Thanks for the updates, good to see your still actively on this.
Cheers
James
...who now has an ever increasing pile of filament gathering on his floor.

130

Re: Filament Winder

Hey Ian - found this on youtube:

http://www.youtube.com/watch?v=Q9KxA28inCo

131

Re: Filament Winder

Ian,
   Thanks for all your work on this.  I just ordered the parts needed to replicate your puller configuration.  I am most interested in extruding 1.75mm PLA which apparently is going to be quite a challenge.  One modification that I was planning on trying was to have a thermal mass pulley (probably aluminum) right in front of the extruder nozzle.  The idea is that it will support the much more flexible PLA and help to more rapidly cool the filament before getting into the apparatus that you have configured.  I was wondering if you have tried such a setup or have any thought on doing something similar.

Thanks,
Mike
P.S.
I will be documenting my experience on my blog 3dprintmd.com/print

132

Re: Filament Winder

Setting up the pulley is a little more complicated, and I never had much luck getting filament to turn a roller as it extrudes.  It might work if the roller is powered.  However I think the best result would be extruding downward because then you are working with gravity rather than against it.  For cooling some kind of blower would probably work better.  In industry they use an air blade which is a stream of air concentrated through a narrow slot.  You could probably work out some kind of duct that would blow air at two sides from a blower like this one - https://www.sparkfun.com/products/11270 (powerful) or this one - http://www.ebay.com/itm/DC-5V-50MM-x-15 … 258082ea78.

Anything that touches the filament between the nozzle and the winder is bound to cause problems because the hardened part of the filament will partially support or push around the filament coming out of the nozzle.  It needs to be free to move around a little at the nozzle.  If it must move through a set pathway, it is bound to get jammed up if something is less than perfect at the other end.

133

Re: Filament Winder

frHere is a preview of v2, on its way to becoming a kit.  Still needs testing and a couple of updates of printed parts, like the holder for the new filament sensor board-

https://lh6.googleusercontent.com/-WdicqjVWIJI/Ugc2XApK8JI/AAAAAAAACxI/CfJQkYU-U-Y/w901-h676-no/20130810_234934.jpg

https://lh5.googleusercontent.com/-L1uca1KzGZU/Ugc2RrVjrHI/AAAAAAAACxo/CknkTZ2UxC4/w901-h676-no/20130810_235138.jpg

https://lh4.googleusercontent.com/-twCIxa--iJ0/Ugc2LvV0OyI/AAAAAAAACws/2idt6RYMcfI/w507-h676-no/20130810_235215.jpg

https://lh6.googleusercontent.com/-9pL_V_6NSBc/Ugc2HxIF2WI/AAAAAAAACwk/js5WtawtkvY/w507-h676-no/20130810_235240.jpg

134

Re: Filament Winder

Interesting Ian, keep going... keep going.

135

Re: Filament Winder

reesy wrote:

Interesting Ian, keep going... keep going.

+1!!!

Great work Ian.  I just waiting for this as I'm running out of space, so can't let my filament fall on the floor, I have to spool it directly!!!

Masterbatch, ABS and PLA Pellets available for UK and Europe.
http://www.emakershop.com/Seller=1324

136

Re: Filament Winder

Awesome work.

137

Re: Filament Winder

This is a great update, Ian, I love seeing your work.

138

Re: Filament Winder

Is there an email list or something that will let you notify us when the v2 winder kit will be available?

139

Re: Filament Winder

If you are going to open source the project and allow people to make their own kits it would be a great idea to set up a donation address. I'd be more than happy to send over funds for your hard work.

140

Re: Filament Winder

Ian,
Any updates on kit status? Need any testers that can self fund?  I have four filastruders set up and willing to test with.... 2 ABS 2 PLA... Very interested in this for the PLA testing as well as im only managing +- .07 without a winder... ABS im managing +- .02 without winder. would love to see how tight i can get those numbers with a winder.

141

Re: Filament Winder

Ian,

What were you using for your light source on the filament sensor module?  The answer may be buried somewhere in this thread, but I cannot find it.

Thanks!

142

Re: Filament Winder

A line laser, afaik.

143

Re: Filament Winder

The winder is awesome. Ian put a lot of thought into it and once he's ready to offer kits everyone will want one for their extruder. High quality parts, good design. I'll be ordering three more myself to go with the beta as soon as he's ready.

144

Re: Filament Winder

Pics of your setup? Filament tolerance data?

145

Re: Filament Winder

Unable to post anything or give specs right now. Had a printer head jam (PTFE tube broke free and caused a massive mess, soaking in acetone now) with two parts left to print so I have the winder rigger together right now. Wouldn't do Ian's design justice to post images or tolerances as is. Bug even rigged up it makes a nice difference. Should have replacement  parts in today or tomorrow so I can get back printing and then I'll post results. :-)

146

Re: Filament Winder

I just noticed http://www.filabot.com/products/filamat … ing-system . What do you guys think of it?

147

Re: Filament Winder

  • No instructions for assembly

  • No instructions for use

  • No means of keeping the filament tight for a nice wrap

  • No empirical data on filament diameter (Filabot claims +/-0.3mm)

  • Pair of photosensors, that means if the filament drops below the bottom photosensor or above the top photosensor then control is lost.

  • No means of walking the filament back and forth across the spool to ensure a even wrap

  • Seems to be on/off (bang-bang) motor control, not PID based control

  • No video of it actually working

  • Seems to run on 120v directly with a bare PCB!

  • Motor runs on 120v

The last two are biggies for me, and highlight something I've been noticing - the other extruders (ExtrusionBot/Filabot/Filafab/Strudittle) all appear to use 110/220vAC to power at least part of their apparatus directly, then leave the wiring up to the buyer! This is incredibly unsafe. It also appears that some of the designers of those kits don't recognize that powering a 110v heaterband with 220v QUADRUPLES the heat produced, and can easily produce hotspots that exceed the rated temperature of the heater band, resulting is short circuit or fire.

Furthermore, a lot of those systems run on 300w+ heaterbands. What happens when a 300w heaterband fails on, due to a failure in the cheap PID controllers we all use? Answer: The polymer thermally decomposes into some pretty nasty gasses and can catch fire.

The Fiiastruder (and Ian's spooler) run entirely on 12v DC, and the Filastruder's heater is 40w - if left full on, it will only reach 210C or so, still within the safe zone for ABS and PLA by a good margin.

148

Re: Filament Winder

There isn't much to work with there.  There isn't any mechanism for putting tension on the filament to make it lay flat, and nothing to guide it onto spool letting it mostly bunch up in the middle.  Also those photo gates make a pretty narrow target for the filament to drop into.  The filament has a way of moving side to side, and the bottom of the loop can shift back and forth.  There needs to be room for the filament to overshoot the sensors up or down without getting hung up on the table or the top gate.  Also if you have to guide it into too tight of a space, it is easy to overconstrain it and have it start to kink rather than go where you want.

149

Re: Filament Winder

elmoret wrote:
  • No instructions for assembly

  • No instructions for use

  • No means of keeping the filament tight for a nice wrap

  • No empirical data on filament diameter (Filabot claims +/-0.3mm)

  • Pair of photosensors, that means if the filament drops below the bottom photosensor or above the top photosensor then control is lost.

  • No means of walking the filament back and forth across the spool to ensure a even wrap

  • Seems to run on 120v directly with a bare PCB!

  • Seems to be on/off (bang-bang) motor control, not PID based control

Thanks for the analysis.

Why on earth would you need a 120v motor for winding filament? If nothing else, I'd think you would never want to be able to pull too forcefully, because it could stretch or break the filament near the extruder, where it's hot/soft. I'd think a wimpy little motor would be just fine for this application, since all it has to do is turn a filament spool extremely slowly (i.e. geared way down), which I think a 5v motor might even be able to do.

150

Re: Filament Winder

laird wrote:

Why on earth would you need a 120v motor for winding filament? If nothing else, I'd think you would never want to be able to pull too forcefully, because it could stretch or break the filament near the extruder, where it's hot/soft. I'd think a wimpy little motor would be just fine for this application, since all it has to do is turn a filament spool extremely slowly (i.e. geared way down), which I think a 5v motor might even be able to do.

If it were up to me, it is not what I would use, but I imaging because they're cheap and (if synchronous), they're constant speed. I think this is the one he's using:

http://www.ebay.com/itm/12-RPM-Synchron … 1252476308

Also motor voltage doesn't dictate torque output. You can get 220vac motors with 1N-m of torque, and 12VDC motors with 1000 N-m of torque.