1 Topic by tonycstech (edited by tonycstech 2014-08-02 22:39:49)

  • tonycstech
  • tonycstech
  • Member
  • Offline
  • Registered: 2014-08-02
  • Posts: 315

Topic: Upgrade request. Perfect "String Theory".

Filament puller.
Stepper motor at the end of the nozzle (far enough to get a grip on cooled down filament).
Speed adjustment knob with display value or some sort.


Its to Speed up/Slow down of filament exiting the nozzle to fine tune diameter due to swelling.
The need for filastruder to stay up certain distance from the floor will be eliminated.
Curling up of the filament will go away. It will keep it more if not perfectly straight because pulling force is constant VS gravity and some accidental resistance caused by curling up on the floor.

It is a theory but i've seen it in action. I dont know if the motor was a stepper motor or not, but it did have speed adjustment to control how fast it pulls on the filament. Its purpose was to fine tune the diameter.

https://www.youtube.com/watch?v=cRJEcWHZEmM

my thingiverse http://www.thingiverse.com/tonycstech/designs
my 3d models http://www.turbosquid.com/Search/Artists/tonycstech

2 Reply by elmoret

  • elmoret
  • Filastruder
  • Offline
  • Registered: 2012-10-11
  • Posts: 6,048

Re: Upgrade request. Perfect "String Theory".

http://www.filastruder.com/products/filawinder

Filastruder - the robust, inexpensive filament extruder
E3Dv6 All Metal Hotend - the premier 3D printing hotend

3 Reply by tonycstech (edited by tonycstech 2014-08-02 22:43:38)

  • tonycstech
  • tonycstech
  • Member
  • Offline
  • Registered: 2014-08-02
  • Posts: 315

Re: Upgrade request. Perfect "String Theory".

Filawinder winds extruded filament on to a spool. It does not effect extruded diameter by pulling faster or slower nor does it make it any straight.

my thingiverse http://www.thingiverse.com/tonycstech/designs
my 3d models http://www.turbosquid.com/Search/Artists/tonycstech

4 Reply by elmoret

  • elmoret
  • Filastruder
  • Offline
  • Registered: 2012-10-11
  • Posts: 6,048

Re: Upgrade request. Perfect "String Theory".

It keeps it straight by holding a constant height loop, which keeps the filament straight and more consistent. Ian can explain better, or I can once I'm at a computer. A puller in open loop won't work like you think - I'll explain more soon.

Filastruder - the robust, inexpensive filament extruder
E3Dv6 All Metal Hotend - the premier 3D printing hotend

5 Reply by IanJohnson

  • IanJohnson
  • Moderator
  • Offline
  • Registered: 2012-08-07
  • Posts: 2,815

Re: Upgrade request. Perfect "String Theory".

I'm working on that now. 

https://www.dropbox.com/sc/yxie46m1l3vm … BVYB3G-yxa

It's harder to keep tolerances this way, because the filament is fully constrained.  Changes in extrusion pressure cause changes in filament diameter, because it can't speed up or slow down.  With the dropped loop, gravity pulls the filament from the nozzle, but doesn't exert force in any other direction so the filament can move as it needs to maintain diameter.

The puller needs to be absolutely consistent in speed so it doesn't introduce any variation on its own.  I'm trying to figure out a roller that can grip the filament without slipping, while still being rigid enough that the filament won't compress it.  The latex roller in the video is problematic because the filament pressing into it changes the effective diameter of the roller, and the  amount of compression changes with filament diameter.  I'm going to set it up with an extruder gear and see if I can get by with a light enough grip that it won't mark the filament.  Pre-made urethane rollers are $15 each, so I'd like to avoid that.

The key to using a puller is measuring the filament and using that to control the speed of the motor.  If the measurement involves handling the filament, it has to happen after the filament is rigid.  That is typically 2-3 inches from the nozzle, which means the PID has to work from measurements taken every 2.5 inches or more.  For it to be effective in controlling the diameter, the system needs to be stable enough that it won't change much between readings.

6 Reply by tonycstech

  • tonycstech
  • tonycstech
  • Member
  • Offline
  • Registered: 2014-08-02
  • Posts: 315

Re: Upgrade request. Perfect "String Theory".

Who does change extrusion pressure ? I dont even mess with the motor unless it does it by its self.
Gravity does work to, but it sometimes make the filament to go sideways. I guess i need some kind of a guide ?

Yes, puller does need to be consistent, otherwise why would stepper motors be used ?

Check out the video i linked on my 1st post, guy uses pretty good roller. I think any plastic will do the trick with spring tension applied.

There is no need for any electronic measurement and self adjustment, it will only complicate the device and skyrocket its price.
Manual spring tension on rollers and manual speed adjustment for stepper motor with 1 stepper controller board. Shouldn't be too hard for someone who know what's he doing.

my thingiverse http://www.thingiverse.com/tonycstech/designs
my 3d models http://www.turbosquid.com/Search/Artists/tonycstech

7 Reply by elmoret

  • elmoret
  • Filastruder
  • Offline
  • Registered: 2012-10-11
  • Posts: 6,048

Re: Upgrade request. Perfect "String Theory".

tonycstech wrote:

There is no need for any electronic measurement and self adjustment, it will only complicate the device and skyrocket its price.
Manual spring tension on rollers and manual speed adjustment for stepper motor with 1 stepper controller board. Shouldn't be too hard for someone who know what's he doing.

This is 100% wrong, as someone that has designed and used a filament extruder for 2 years now. The pellets are not uniform size and the way they lay in the auger is obviously random so the rate material is delivered is variable. Therefore, any puller/spooler must have a closed loop sensor.

Filastruder - the robust, inexpensive filament extruder
E3Dv6 All Metal Hotend - the premier 3D printing hotend

8 Reply by tonycstech

  • tonycstech
  • tonycstech
  • Member
  • Offline
  • Registered: 2014-08-02
  • Posts: 315

Re: Upgrade request. Perfect "String Theory".

I start to see your point.
It pushes the actual mass, not the volume.

my thingiverse http://www.thingiverse.com/tonycstech/designs
my 3d models http://www.turbosquid.com/Search/Artists/tonycstech

9 Reply by IanJohnson

  • IanJohnson
  • Moderator
  • Offline
  • Registered: 2012-08-07
  • Posts: 2,815

Re: Upgrade request. Perfect "String Theory".

Pressure still changes in the extruder due to slight differences in the size and number of pellets going into each flight of the auger, especially with the small size of the auger in relation to the pellets.  There are probably other factors at work in the melt zone, maybe slight changes in temperature or uneven distribution of heat throughout the melted plastic.  Even the industrial extruders use laser micrometers to measure the filament and control the pull rate.  Stratasys actually has a patent on the idea, but everyone else does it anyway.

I worked on a puller when first developing the winder, and could never get better than .15mm tolerances even when I used an extruder gear.  I decided closed loop control would be necessary, and if I went that route it might be another 6 months before I had something ready to launch.  The dropped loop takes up more space, but is far simpler and produces better quality filament.  I'm revisiting the concept using a magnet and hall sensor.  The filament pushes up a bearing attached to a lever with a magnet on it, and a hall sensor detects changes in distance caused by changes in diameter.  It is a lot simpler and cheaper than optical measurement, but I can only get as close to the nozzle as I can cool the filament.  Lyman's v5 extruder is using optical measurement to drive a puller, but still doesn't get better than .1 variation, so I'm not sure I will do any better.

Russ's rollers are like the ones I was talking about, and probably more expensive.  I have to keep the total cost of the BOM from getting out of hand, and can't devote $30 of a kit's material cost just to the rollers when I'm already adding a stepper, stepper driver and LCD.