My "2 Wifi" is un-baby-stepping

Phaedrux

So I had a chance to take a look at your config on a proper screen now, and I still think there are some things that could definitely be causing you issues.

M350 X64 Y64 E256:256 I0 ; high enough microstepping  to cause hiccups
M350 Z4 ; Low microstepping on z axis
M566 Z750 ; high jerk for z axis
M203 Z1000 ; high max speed for z axis
M906 Z2600 ; max motor current into a tiny pancake motor

Combined with all of your quoted commanded Z moves would be requesting the full max z speed.

I think you might have best results with x16 microsteps across the board with interpolation to 256. Possibly higher for the extruder if it results in less than 400 steps per mm. I wouldn't be surprised if your microstepping settings are high enough to trigger hiccups and possibly missed steps.

Reduce the max Z speed to 300mm/min.
Reduce Z jerk to 50mm/min

It's not clear to me if you are using the Za and Zb motor connectors, or if you're using independent drivers for your Z axis. either way you may want to revisit your current settings to see what the motors are actually receiving, as your M906 Z2600 would seem very high at first glance.

gnydick

@phaedrux I am using the a&b setup on the same driver. The reason for the settings like that are because the pancakes are not strong. So the larger the microstep the more torque it can handle.

Also, the current is high because there are 2 motors. As far as I understand, if you double the motors on a circuit, you double the current to compensate. Is that correct.?

Phaedrux

Sure, I get what you're trying to do with the low microsteps but the high feed rate and jerk settings are kind of working against you. Lowering those might help reduce the initial torque required to get moving. You've already got a nice low acceleration value.

Things can also behave strangely at low microstep values like that. Things like motor direction flipping from what would be expected.

Maybe try x8 microstepping. The amount of torque lost won't be as much as you'd think.

If you have a free driver, why not give each motor its own? Keep in mind that the Za and Zb connectors are internally wired in series, not parallel, so they should be getting the full current each, not half. At least that's my understanding. Maybe I have it backwards, but I think that's why the Duet has them wired that way internally in the first place.

A stepper motor may be wired in either series or parallel, depending on the needs of the application. A series-wired motor will deliver more stall torque, but torque drops quickly as velocity increases. A parallel-wired motor typically maintains its (lower-than-series) torque to a higher velocity.

Do you have the specs on the motor? Someone more versed in stepper motors than I could probably tell you whether they would be better used in series or parallel.

gnydick

@phaedrux

Yeah, the jerk can change, the feed rate being that high isn't ever used.

The problem I'm experience isn't an effect though. It's definitely being executed. I've converted the Z to a 3-to-1 gearing, so it's much more visible because the motor has to spin further to get the same distance. I've tried pretty much every different microstepping setting.

I don't think I have a free driver?? X, Y, Zab, E0, E1

https://e3d-online.dozuki.com/Document/124/.pdf

Phaedrux

Yeah if you're using both extruders then no free driver.

I still think 2600ma is overboard considering they are rated at 1700ma Have you tried 1200ma? Or even 1700ma? Are the motors getting hot?
Especially if you're using 3:1 gearing now.

Other than that I'm out of ideas. Definitely odd behaviour.

gnydick

@phaedrux Yeah, they're getting hot. They're undergeared, 20 teeth on the motor, 60 teeth on the lead screw.

aidar

If you are using Za and Zb for motors, than they are wired in series, that means you must NOT double current. Double current is needed when motors are wired in parallel.

deckingman

@gnydick said in My "2 Wifi" is un-baby-stepping:

............... I've tried pretty much every different microstepping setting...............

That must have been a pain because the way you have things configured, each time you change the micro stepping, you have to re-calculate and enter the the steps per mm for that micro stepping. So here is a little tip. Set the steps per mm for 16X in M92 but put that line before the M350 micro stepping line. Then you can change the micro-stepping in M350 and the firmware will automatically re-calculate the steps per mm.

As others have said, if you set micro-stepping too high, especially with the gearing you have, then you'll run the risk of hitting the step pulse frequency limit which is about 200KHz. So I'd suggest you start with 16X with interpolation which should be fine.

wilriker

@gnydick said in My "2 Wifi" is un-baby-stepping:

@phaedrux Yeah, they're getting hot.

Depending on how long you have already run them at 171% of their rated current (1.4A according to the linked datasheet and 2.4A maximum current that the drivers on the Duet will put out - even if you command higher values) they might already have gotten to a temperature where their internal magnets have suffered and thus even further reducing their torque. So I would not count on them still being able to provide 13N.cm of torque anyway.

gnydick

@deckingman

OOOOOOHhhhhhh. That's a great tip for ordering the commands. Yes, it was a pain, because even though it was powers of 2, it never worked out exactly and I had to measure every time. Luckily, I have a spreadsheet where I just plug in the settings and the measurement and it tells me the new setting to put in to fix it.

But, there seems to be a chicken and egg problem. Are you saying that the default is 16x and that's what I'm measuring if I were to not even specify microstepping?

gnydick

@phaedrux I switched my extruder to x16 + 256 interp, and now the surface of my prints are pretty poor. They're rippled along the entirety of the path, not just beginning or end.

deckingman

@gnydick said in My "2 Wifi" is un-baby-stepping:

@deckingman

OOOOOOHhhhhhh. That's a great tip for ordering the commands. Yes, it was a pain, because even though it was powers of 2, it never worked out exactly and I had to measure every time. Luckily, I have a spreadsheet where I just plug in the settings and the measurement and it tells me the new setting to put in to fix it.

But, there seems to be a chicken and egg problem. Are you saying that the default is 16x and that's what I'm measuring if I were to not even specify microstepping?

To save me typing - check this out - https://duet3d.dozuki.com/Wiki/GCode#Section_M350_Set_microstepping_mode

Phaedrux

@gnydick said in My "2 Wifi" is un-baby-stepping:

@phaedrux I switched my extruder to x16 + 256 interp, and now the surface of my prints are pretty poor. They're rippled along the entirety of the path, not just beginning or end.

With your other axis at x16 you have open budget to increase the extruder micro steps. It's recommended to get your extruder steps per mm value in the 400 to 1000 range. Try x64 microsteps.

You can use this tool to determine what your max retraction speed with your microstep settings can be before getting skipped steps.

https://wilriker.github.io/microstep-calculator/

gnydick

@phaedrux gotcha. Thanks!

gnydick

So, it's still happening. It looks as though rapid baby-stepping is undone, but slow, infrequent baby-stepping isn't

dc42

Can you provide a GCode file and a sequence of steps to reproduce this?

gnydick

@dc42 sure, I'll get it to you today.

gnydick

0_1550902843214_Ghosting Test Makers Muse (1).gcode

To reproduce, just rapidly enter baby-steps. Say, 4 clicks, and what the z motors.

dc42

I'm sorry, I can't reproduce this. The only negative Z movements I see are the one at the start where you wipe the nozzle, and the ones that undo Z-hop in the GCode around travel moves.

Tested using the 2.03beta2+1 release.

gnydick

@dc42 can you print with the latest stable release instead of the latest dev version? Do you want me to upload my entire sys folder so you can try it?