Reprap Visit Sunday 10th Details
Contents |
Details of Debugging on Sunday 10th
Notes taken by Mike.
When Lisa and I arrived, Drew and Skippy were already working. They gave us a run-down on the current state. I helped debug from then until I left. Jeff Crews dropped by for a while and contributed. Drew attended the Mapping as Activism class while Skippy and I continued to debug. When I left, Drew and Skippy were still at it, and Jeff was on his way back.
Main problem: extruder not consistently extruding plastic
Had tried PLA the previous day
Trying ABS today (since the PLA had gone home with Jeff)
phone call with Neil Underwood
call placed around @ 1:20pm
using PLA:
- water gets in and it swells
- end gets the most swelled
- both ends of the PLA are fatter (even when new from manufacturing)
- can tolerate a little bit of difference but not as much as 3.5mm (in a 3mm hole)
using ABS:
- temp is the default (230)
- problem is extruding, then slip, then extrude, then slip, ...
- seems to be working better with more tension in the extruder assembly
feed in, feed out -> should be knarly nibbles, big gouges
Neil didn't know we were feeding in large (3.5) PLA
might have had PLA metlting before getting through the nozzle
brass nozzle has 3mm hole
thermal barrier could be full of PLA
might have to take extruder off and melt out stuck PLA:
- unscrew 3 bolts that hold stepper
- cut or extract plastic
- take extruder off top (2 crews)
- hot end comes off
- Neil put plugs on extruder to it just unplugs
- take thermal end off
- take silicone jacket off hot end
- take thermisotr off -> be gnetle
- cut tape
- take thermistor off
- take brass nozzle
use blow torch (propane, buttane, lighter, etc)
melt plastic out of nozzle
probably have a plug of plastic in there
another way to test: warm nozzle up, try to turn extruder wheel by hand, see if there's a lot of resistance
if we hand feed and there isn't tension -> there's something going on with stepper motor
if we hand feed and there is tension -> can try swapping E and Y stepper controllers (since Y doesn't take much force) [theory being that E controller is providing enough juice for the extruder stepper)
thinking the problem is that we were feeding 3.5mm PLA through a 3mm
Neil has printed a bunch with this machine so it should work (is/was known good as some point in the past)
possible problem with some of Neil's rewiring
Debugging
PLA being used yesterday was old
- diameter should have been < 3mm
- Skippy measured the PLA at 3.2mm and 3.3mm (but his measurements might have been off)
we took extruder apart and disassembled nozzle
used butane soldering iron as a torch to heat it up manually
there was red ABS (what we're using today) in there, but no PLA gunked up
problem:
- ABS is getting stripped / gouged at the knurled bolt and not feeding (@ rate of 300mm/min)
variables:
- feed rate
- change tension on the feed bolt
- temperature
- current coming through the stepper controller (set by potentiometer)
- heat at the controller (too much heat will shut down controller and make it "stutter")
- cleanness of gear teeth (on the knurled bolt)
- aged of plastic
- tooth flossing issue (too much ABS in the feed bolt teeth)
reduce rate to 100mm/min
- result 1: better - fed smoothly, but for only 81mm (19mm short)
- result 2: no good - feed for a few mm then stopped with ABS gouged out
I noticed skate wheel bearing was not mounted on an axle and was not turning at all when feedstock is feeding
Jeff said that in all the extruders of this type that he has seen or heard about, the skate wheel is mounted on an axle so it can turn, and the tension is very high on the tension screws
cut a piece off the end of one of the X-axis rods using a Dremel to use as the axle for the skate bearing
mounted skate wheel on the axle and put the extruder back together
another problem Jeff noticed is the teeth on the knurled bolt were full of both PLA and ABS from when it had been gouging the feedstock
Jeff uses dental tools to floss the teeth
after flossing and getting skate wheel on an axle:
extrude at 100mm/min -> short by 5.3mm
extrude at 200mm/min -> stopped, gouged the ABS, gummed up the teeth again
[stopped and flossed teeth again]
extrude at 100mm/min -> short by 5.2mm
extrude at 100mm/min -> short by 5.7mm
OK, we now assume we have a consistent feed
steps to go through before printing:
- check that print platform is level
- check all of the limit switches (X, Y, Z)
Skippy and I tried a test print of the RepRap 10mm cube gcode file that was on the desktop of the PC
extruder dwell - "G1 E1250.0" in the gcode - is a problem
causes extruder motor to sit and whine
Drew said a 5% error between what control s/w thinks and what extruder is actually doing is way too much and the math needs to be changed/updated to match in the s/w
Notes
hex bolts holding extruder to X-axis gantry is 7/64" allen wrench
3mm wrench needed for extruder tenstion bolts
the three hex bolts holding the X-axis stepper motor take 2 different size hex wrenches (SRSLY!?!)
wires to extruder assembly
- green and red go to green (power for heating)
- blank and white goes to white (thermistor)