RE: Stepper Motor Upgrades to Eliminate VFA's (Vertical Fine Artifacts)
...and now for some 3dBenchies:
Posted without commentary, because I don't want to influence anyone's opinions. I'm curious what other people think besides myself!
I prefer the TMC ones.
RE: Stepper Motor Upgrades to Eliminate VFA's (Vertical Fine Artifacts)
I tried the newest firmware you created, and also added the info, above, into my gcode to be run after the firmware update. All my numbers matched what they should be without needing a factory reset! Great job!
Also, here is a copy of that cube you sent me a while back. The size is good and I think it looks pretty good...but my Benchy (downloaded from thingiverse instead of using the Prusa provided one - but it is from the same people as Prusa got it from) has lines on it. Any idea what I should change to make that better?
Also, how do I use the new LA15?
RE: Stepper Motor Upgrades to Eliminate VFA's (Vertical Fine Artifacts)
That's about the limit of VFA reduction we can get out of the Moons' 0.9 steppers. The OMC stepper can do a tiny bit better when linear correction is set to 1.130
Which reminds me, I have to go back and set linear correction for my OMC which is on my Y-axis. Forgot to do that after my last factory reset.
I don't know how to use the LA15 firmware. My branch LA15 and 0.9 motor branch has the code from Prusa's LA15, but I have not seen documentation on how to use LA15 features.
RE: Stepper Motor Upgrades to Eliminate VFA's (Vertical Fine Artifacts)
BTW, that nasty horizontal line you see in the Benchy is the "solid surface" defect to which Evan was referring. It's unclear what causes that happen. It occurs in coincidence with tops of solid horizontal surfaces. Shows up in all manner of models and is super annoying. Slicer has been implicated. Also seems to worsen if print cooling is marginal at that outer surface.
RE: Stepper Motor Upgrades to Eliminate VFA's (Vertical Fine Artifacts)
Thanks! What is a good Linear Correction to use as a starting point for the moons? I do not have mine turned on at all...have to read up on it yet.
I will print it using S3D 4.1.2 and see if that line is there for it...but in a few hours, doing another test print at 0.15 layer height - the one I posted was a 0.05 layer height.
RE: Stepper Motor Upgrades to Eliminate VFA's (Vertical Fine Artifacts)
What is a good Linear Correction to use as a starting point for the moons?
OFF - the Moon's 0.9's have never tested to be better with linearity correction on in my experience.
RE: Stepper Motor Upgrades to Eliminate VFA's (Vertical Fine Artifacts)
Perfect, I like easy.
RE: Stepper Motor Upgrades to Eliminate VFA's (Vertical Fine Artifacts)
The values for LA-15 are much lower than version 1.
http://marlinfw.org/docs/features/lin_advance.html
PLA seems to want a M900 K value of 0.04
PETG seems to want K value of 0.11
I found my CF-PETG prefered 0.09
RE: Stepper Motor Upgrades to Eliminate VFA's (Vertical Fine Artifacts)
My print of the Benchy sliced in S3D does NOT have that solid layer line. You are correct, it is caused by Slic3r. Excuse the color change, the black ran out mid print. Was a good test of the MK3S filament run out sensor...it works and still has enough filament left for me to grab and pull.
RE: Stepper Motor Upgrades to Eliminate VFA's (Vertical Fine Artifacts)
My print of the Benchy sliced in S3D does NOT have that solid layer line. You are correct, it is caused by Slic3r. Excuse the color change, the black ran out mid print. Was a good test of the MK3S filament run out sensor...it works and still has enough filament left for me to grab and pull.
RE: Stepper Motor Upgrades to Eliminate VFA's (Vertical Fine Artifacts)
I have been running fine with the stock Bondtech motor (1.8) and your firmware with 830 e-steps. Is there any benefit to be had going to the version with 415?
also, could the new lower e-steps mean that maybe that .9 LDO pancake that had some problems at 830 might work well now with the Bondtech and your firmware? As I seem to remember, the problem was max speed for retractions being out of spec but you would gain a 2x benefit of speed here if I am correct? Or because these are microsteps does this not change anything.
I guess I am wondering what their rationale was going to 1/2 the steps on a 1.8 motor...don’t you loose some extrusion accuracy?
RE: Stepper Motor Upgrades to Eliminate VFA's (Vertical Fine Artifacts)
...don’t you loose some extrusion accuracy?
I wonder about this as well.
RE: Stepper Motor Upgrades to Eliminate VFA's (Vertical Fine Artifacts)
Let's keep the two issues separate in our minds.
The lowered microstepping is to avoid exceeding EINSY CPU max rate capability during fast extruder moves like retractions and MMU2S filament load / unloads. That could occur on the 1.8 motors at the higher microstepping rate. Yes, it theoretical drops mechanical resolution. However it isby an amount less than our mechanicals can achieve about 1 micron). For practical purposes, it doesn't reduce real world extrusion accuracy during extrusion. For higher speed maneuvers it increases accuracy because EINSY is no longer forced to skip over positions to keep up. Another minor benefit is extruder motor noise is reduced.
The 0.9 pancake issue was one of torque. My firmware drops microstepping for 0.9 so the effective step rate was the same as with 1.8 motors. It isn't a CPU being overrun issue. So, halving the microsteps doesn't fix that. What does appear to be making operation of 0.9 pancakes possible it the new LA15. Dobewan has tested a 0.9 pancake under LA15 and gotten good results. I have not had a chance to test on my machines, but I have added LA15 to a new branch of my 0.9 degree support firmware.
https://github.com/guykuo/Prusa-Firmware/tree/LA15-and-0.9-Degree-Stepper-Support
That new branch implements both the halved step rate and LA15 changes.
Be sure to either do a full factory reset and erase if going to a current version of either branch. Failure to do so will let old settings in EEPROM override what my firmware attempts to set.
At the very least, do the M350 and M503 sequence to force microstepping and verify the settings. You have to use whater the e-steps and microsteps are corrrect for your particular extruder gearing ratio and type of motor.
For example, a BNBSX with 54 tooth gear and 0.9 motor would need (omit everything in parenthesis)
M92 E473 (set e-steps to half what it used to be for 16:64 gearing)
M350 E8 (set microstepping to half former rate)
M500 (store settings in EEPROM)
power off/on
M503 (check the settings)
RE: Stepper Motor Upgrades to Eliminate VFA's (Vertical Fine Artifacts)
For clarity, I changed the firmware instructions to read as...
Geared extruders now set for lower microstepping to avoid overruning EINSY during fast retracts or MMU2S filament moves.
Factory reset and delete all data after installing this firmware. Otherwise EEPROM settings override settings in this firmware.
After installing this firmware, send M350 and M92 commands to force correct micro-stepping and e-step rates.
e-steps values for M92 depend on your extruder gearing.
xxx = 280 for non-geared extruder
xxx = 415 for BMG extruder (special Bondtech compensated value)
xxx = 420 for 3:1 extruder
xxx = 473 for BNBSX with 54:16 gearing
xxx = 490 for BNBSX, Short Ears, Skelestruder with 56:16 gearing
non-geared extruder, 1.8 degree motor
M92 E280
M350 E32
M500
non-geared extruder, 0.9 degree motor
M92 E280
M350 E16
M500
geared extruder, 1.8 degree motor
M92 Exxx
M350 E16
M500
geared extruder, 0.9 degree motor
M92 Exxx
M350 E8
M500
Follow with power off/on and M503 to verify settings are correct.
RE: Stepper Motor Upgrades to Eliminate VFA's (Vertical Fine Artifacts)
The 0.9 pancake issue was one of torque. My firmware drops microstepping for 0.9 so the effective step rate was the same as with 1.8 motors. It isn't a CPU being overrun issue. So, halving the microsteps doesn't fix that. What does appear to be making operation of 0.9 pancakes possible it the new LA15. Dobewan has tested a 0.9 pancake under LA15 and gotten good results. I have not had a chance to test on my machines, but I have added LA15 to a new branch of my 0.9 degree support firmware.
https://github.com/guykuo/Prusa-Firmware/tree/LA15-and-0.9-Degree-Stepper-Support
That new branch implements both the halved step rate and LA15 changes.
Be sure to either do a full factory reset and erase if going to a current version of either branch. Failure to do so will let old settings in EEPROM override what my firmware attempts to set.
At the very least, do the M350 and M503 sequence to force microstepping and verify the settings. You have to use whater the e-steps and microsteps are corrrect for your particular extruder gearing ratio and type of motor.
For example, a BNBSX with 54 tooth gear and 0.9 motor would need (omit everything in parenthesis)
M92 E473 (set e-steps to half what it used to be for 16:64 gearing)
M350 E8 (set microstepping to half former rate)
M500 (store settings in EEPROM)power off/on
M503 (check the settings)
Yes, this worked very well and I am still adjusting, but this alleviated all of my issues with the BMG/LDO 0.9 stepper I was running into. To be 100% honest I am unsure of the actual quality improvement vs the moons 1.8 pancake seeing that I have been using this stepper before the LA 1.5 with very low retraction set. Now, with this firmware, it has 0 issues with retraction, but now I am trying hunt down occasional overextrustion issues which is not a big deal.
That being said....
I will be possibly converting to the mosquito soon, so narrowing my issues down is temporary until I mock up a BMG/Mosquito setup with all the bells and whistles.
Getting rid of VFA's has now turned into a obsession 🙂
RE: Stepper Motor Upgrades to Eliminate VFA's (Vertical Fine Artifacts)
The Zen zero detent motors are out of the running. Lin Engineering won't sell less that 5 units. Each motor is $150. That's an entire Prusa MK3S to do the test. Nope.
RE: Stepper Motor Upgrades to Eliminate VFA's (Vertical Fine Artifacts)
What's the PN on the Lin's? I have a box of LIN steppers here somewhere...
RE: Stepper Motor Upgrades to Eliminate VFA's (Vertical Fine Artifacts)
It's one of the Zen series . The Zen series isn't available as a "web motor" so they only want to sell in quantity and with an estimated yearly purchase rate.
ZN417-11-06-RO is the model I think would be a best match for our printers.
RE: Stepper Motor Upgrades to Eliminate VFA's (Vertical Fine Artifacts)
Ah, nvm then. My box is 211s and 4118s. They're 1.8 degree but they don't feel like they have a lot of detent. I'd have tried the 4118s for kicks if they weren't 2A motors which is more than the TMCs can handle.
RE: Stepper Motor Upgrades to Eliminate VFA's (Vertical Fine Artifacts)
(the 211s are NEMA11 and while ~0.6A I doubt they have the holding power/torque to stop an extruder or bed without skipping.)