Clearpath Servos with 1XD Expansion

jballard86

@JoergS5
Its a core xy setting steps/mm is irrelevant, you use the Mcode to scale the x and y

Testing at later change could give some interesting results but I'm using hard stops, so not viable if the shifts are in the wrong direction. And as I can't log both servos count position at the same time and have to swap cables it could take an afternoon or longer to run and rerun tests.

Do you think the servos are loosing position or the duet? Im 100% confidant that it's the duet as I've already shown that.

jballard86

@JoergS5 Post crossing is happening often

jballard86

@JoergS5 I'm adding a g4 s.5 on layer change, I predict it may stop the shift, but is also not an acceptable solution. Results in a few min.

JoergS5

@jballard86 M400 before layer change would be a test also. Maybe XY throws a movement away when Z move starts.

jballard86

@JoergS5

g4 reduced a lot of the shifts, trying:

m400
g4 s.5

results in a few

jballard86

@JoergS5
still shifting with m400 and g4 s.5

JoergS5

@jballard86 at the moment I don't have additional ideas. I will think about and tell you when I have a new one.

jballard86

@JoergS5

This is interesting, it had shifts mid layer on the skirt

I'm out of ideas as well, maybe @dc42 can give some more input.

JoergS5

@jballard86 slip through of the belt to the pulley is a possibiltiy also, but you would hear it. But I don't know how to verify it.

jballard86

@JoergS5 an I am unfortunately aware of that noise from when I was setting up hard stop homing

tae wimju

Hi. This may be long-shot, but...have you check that all ClearPath motor has the exact same RAS setting, and that the motor are not set to "Auto RAS"? Keep the RAS setting fairly low, like no more then 25mS (mine only let me set maximum 16mS which work OK fine for me. But make sure the RAS values match exactly the same. Good luck with print machine!

NitroFreak

I had the same exact problems on my JMC servo,
I guess you are not using the duet expansion breakout board? That fixes it.
You need the 5V logic for the steps to arrive properly, the 3.3v logic from the mainboard is enough to trigger some steps to run normally but not all of them.

So yes you need the expansion breakout board. here is my thread on it
https://forum.duet3d.com/topic/8948/3-3v-logic-for-5v-servomotor-optocouplers

T3P3Tony

@NitroFreak the 1XD produces 5V level signals so i don't think its that.

@jballard86 Can you do a simpler test to try and reproduce the missed steps (if it is that that is causing the shifts). Do the servos have an interface where you can see the ticks received, or work those out from movement received? try just a movement of a set distance back and forth between two points (without printing).

jballard86

@tae-wimju Both have the same RAS setting, I actually turned it to none on both servos to rule out RAS early on.

jballard86

@T3P3Tony The problem is not reproducible while other steppers are not operating. I made a simple gcode file with about 50 moves, many of them with 1mm oscillations and then return to home. After running the code dozens of times no change in origin for the duet or servos was noticed.

Im pretty sure that the issue has to do with communication between the duet 3 and the 1xd boards.

Ive thrown some steppers in place of the servos temporarily.

Any other good tests I could do? to help narrow the issue down?

Teknic_Servo

Hi @jballard86,

I'm a Teknic engineer and I came across your post. While it doesn't seem like the motors are causing the issue you are seeing, the ClearPath MSP software has a lot of useful diagnostic tools that can help us figure out where the problem may lie (mechanics, controller, wiring, settings, etc.).

Based on the information in this thread, I'm not quite sure what the issue is, but I'd be happy to set up an online session and troubleshoot with you. Please feel free to give Teknic a call if you are interested.

EDIT You may have seen some of the support videos that Teknic produced to help customers learn the MSP software and diagnostic tools. In the event that you haven't (or in the likely situation that someone else comes across this post and needs help troubleshooting some aspect of their machine), here are some links to those videos:

-"Using MSP (Motor Setup Program)": https://www.teknic.com/products/clearpath-brushless-dc-servo-motors/model-support-sd-dc/#model-support-running

-"How to Use ClearPath's Diagnostic Tools" (we plan to eventually move this video to the link above as well, but right now it's located on our "Videos" page under "Demos and Tutorials"): https://www.teknic.com/watch-video/

Best,
Bridgette G. - Teknic Servo Systems Engineer

jballard86

@Teknic_Servo

I may take you up on that. Im also pretty confidant that this issue has nothing to do with the servos. But the 1XD are a beta device and im sure the problem revolves around that.

I will be putting the Servos back in the printer when I get some more ideas from Duet, then I may call yall up to help diagnose.

jballard86

@T3P3Tony

What is the max step rate of the Duet 3? That may be the problem, the less micro stepping I do the faster the printer will actually print.

Im debating putting the Servos back in and try them at maybe a 2x equivalent microstepping to see if the error continues.

Any more ideas on your side?

dc42

The max step rate reduces as you increase the step pulse width and interval. So set the M569 T parameters to the minimum values given on the datasheet for the drivers.

8000 pulses per rev is equivalent to 8000/200 = x40 microstepping. So reducing pulses per rev may be advisable.

Use M122 to report the number of hiccups on the expansion board. High values indicate that the 1XD cannot maintain the requested step rate.

jballard86

@dc42 towards the end of my testing with the servos I had dropped down to 3600/rev and hiccups were in the single digits usually 1 or 2

pulses were setup at 2.7:2.7:2.7:2.7

still had shifting