Z height unstable w/ Duet/FSR setup

[marcom]

After many months of happily printing, it appears my printer has developed an issue that no longer allows it to print properly - basically, my first layer is very uneven.

My setup:

• Rostock Max V2
• Stock hotend
• Duet 0.85
• Ultibots FSR kit w/ mhackney's FSR mounts
• Ball joint arms and molded carriages
• PEI bed
• RepRapFirmware for Duet v 1.16
• Web Interface 1.13

I was noticing that my first layers were starting to become pretty unpredictable. They'd be thin in some places, thick in others.

If I issue G1 X0 Y0 Z0.2, the center passes the paper test, but if I issue G1 X0 Y100 Z0.2, the nozzle is too high, G1 X0 Y-100 Z0.2 the nozzle is too low. Also, it seems the more I move between those two points, the worse it seems to get (though I can't tell for sure).

First, tightened all the belts - the issue was still there.

Then I realized I heard creaking in the joints so I thought it might be an arm issue - lubricated the arms and the creaking went away - the issue was still there.

Then I came across the "Unsolved mystery. Weird Z0 behavior" post and it seems like the fix there was a change in Repetier Firmware (Carriage Horizontal Offset). But I am not running Repetier.

I haven't yet tried replacing any of the hardware since I don't really know where to start.

My questions:
- Is this still an issue for other folks, particularly anyone running Duet w/ FSR's?
- Is there an equivalent parameter in the Duet RepRapFirmware that is perhaps in need of adjustment? If so, any idea what that would be?

Any help anyone can provide would be appreciated. Like I said, the printer worked well for ~9 months or so, and I have really enjoyed it until this problem crept up and now it's pretty much unusable since the first layer is so variable.

[geneb]

marcom, do you have the ball & socket arms, or the older u-joint type? If the older type, are the u-joints plastic or metal?

g.

[marcom]

Hi Gene!

Thanks - I have the newer ball joint arms - bought the printer in January of this year.

- Marco

[geneb]

Ok. On my Duet-equipped printers, I calibrate before the start of every print job. Maybe try that?

g.

[mhackney]

Thiis sounds like you are not getting reproducible probing at every point. Most likely to slight binding of your FSR mounting system. It doesn't take much. You can test by probing at each of the points multiple times (3) and checking the reproducibility of the results. Read this: http://sublimelayers.blogspot.com/2016/ ... -duet.html

[marcom]

For posterity's sake I am posting here the final resolution to the problem:

After tearing my hair out and not knowing what to do, coupled with a severe lack of time due to other responsibilities like twin boys, I finally figured it out! My Z-axis pulley had slipped ever so slightly and was catching the belt. This was subtle and easily missed when inspecting (except of course for the messed up printing). However, after continued running the pulley slipped so much that the Z axis would just outright catch, finally making the problem hard to miss. A tightening and resetting of the pulley fixed matters.

I also yanked the Astrosyn damper there, since putting the pulley in the right alignment with the damper was tricky and probably partly responsible for it slipping (without the damper you just flush it to the axle). Anyways, I'd rather have a noisy but properly-printing machine.

(FWIW, prior to this I did try to clean up my FSR mounts and even replaced the FSR sensors, but the problem persisted - luckily the continued running and testing of the FSR replacement was what ultimately led the pulley to finally catch!)

From the looks of my latest calibration piece, looks like it's BACK! Thanks for your help and sorry it took me so long to post a followup - I just figured it out tonight. Back to happy printing!!!

[mhackney]

Thanks @marcom Little things like your Z pulley make a big difference sometimes. Good catch! I am no longer using AstroSyns or any form of damper. You can actually see the AstroSyns wobble as the printer moves. The Duet WiFi drivers are sooooo quiet that they are no longer needed and they do adversely affect print quality.

[DeltaCon]

Nice of you to close this
Since I am on Duet 0.6 I have absolutely no need for stepper dampers anymore. Higher version seem to do even better reducing the stepper noise. Yes slipping pulleys are easily overlooked indeed, and it is a pain to fasten them (if you don't have a long enough wrench!)

Most likely to slight binding of your FSR mounting system. It doesn't take much. You can test by probing at each of the points multiple times (3) and checking the reproducibility of the results.

Yes especially calibrating at different bed-temp can lead to inconsistent variations in the drag in the mounts. I make it a habit to tap a finger lightly in the bed at each mount position, to test it's functionality. Also I mounted the SL board behind the acrylic side panel so that the lights are visible. When I do a calibration, I look at these lights and se only one light up at the positions of the mounts, and 2 at all positions between the mounts. Probing the center of the bed gives me three simultaneous lights. Then I know that the probing is consistent

[Jimustanguitar]

Then I came across the "Unsolved mystery. Weird Z0 behavior" post and it seems like the fix there was a change in Repetier Firmware (Carriage Horizontal Offset). But I am not running Repetier.

Also for posterity... The carriage offset value didn't really effect anything that changed print quality or that people reported fixing their problem. The number is used once in the code to do a subtraction equation for the horizontal radius, and that's it. You could add a million to both values and get the same subtracted difference, so while it's good record keeping to have the proper value entered, there's not a behavior difference at all.

[mhackney]

I have now eliminated any possible binding/drag with my new mounting system. I also put the JohnSL behind the acrylic panel - great minds think alike! A nice subtle indication during probing.

[marcom]

New mounts?! Do tell! I am using your previous system and it's served me well but always looking for more upgrades, especially now that I'm up and running again.

[DeltaCon]

I have now eliminated any possible binding/drag with my new mounting system.

Can't wait to see!

I also put the JohnSL behind the acrylic panel

Yes, the pins really should extend to the other side of the board to make that easier. I contacted ultimaker about that, but ofcourse they never heard that before I have too much lack of solder skills to do that myself. Connectors and wires I became pretty routined in lately but electronics I don't dare touch...

great minds think alike!

Haha, Thanks, I obviously have a little less self esteem

[mhackney]

@marcom @DeltaCon and all, new mounts: http://forum.seemecnc.com/viewtopic.php ... 27#p102727