TMC2660 problem

dc42

@jerolee please can you provide a link to the pancake stepper motor that you purchased. I will then purchase on so that I can test it with Duets.

@jerolee said in TMC2660 problem:

This means that my Z-axis will always run out of sync go skewed in height while the motors are not exactly the same?

Only if your Z steps/mm is not the same for all your Z motors, or they can't all be run at the same current.

GeneRisi

@dc42 I have exactly the same motor. It works. @jerolee , here is my suggestion. Connect the motor wiring from the motor all the way to the duet board. The duet board connector has the wiring between coils on adjacent pins, so one coil will be pins 1 and 2. The other coil will be on pins 3 and 4. Measure the resistance on the connector you are plugging into the duet board. If you do not get approx. 2 ohms for both pairs, a wire is crossed. Typical Nema 17 steppers have the pairs interlaced. The LDO pancake stepper does not.

dc42

@generisi, thanks for your observations.

@jerolee in your original post you said:

I cant get mine to work, which ever combination of wiring, tried 8 combo's
result a little hum.

Unfortunately, connecting stepper motors incorrectly to a stepper driver is the best way of blowing it. We took steps when designing the Duet to improve the resilience of the drivers when a motor is disconnected under load. However, the current regulation in the stepper drive relies on the motor being connected correctly, and we can't do anything about that. This is why at https://duet3d.dozuki.com/Wiki/Choosing_and_connecting_stepper_motors#Section_Connecting_stepper_motors we stress the need to connect the motors correctly.

The TMC2660s are high-current drivers and the short-circuit protection built into them is unfortunately not fast enough to protect the driver in all situations (even though RRF sets it to maximum sensitivity), especially when connecting low-inductance motors incorrectly. Therefore I suggest you check whether the motor output that you are testing the pancake motor with is still capable of driving one of your Nema 17 motors correctly.

jerolee

@dc42
All of the drivers still operate normally with all NEMA17 on it. So none of the drivers is defective or burned out. my 3D printer functions, it homes all axis, it does the tramming. it just wont let the pancake motors rotate. i connected them to all of the drivers and they just dont work.
Please stop replying on the wiring because i connect the coils correctly. 1 coil on 1-2 pin other coil to 3-4 pin. i looked at the wiring schematic of the Duet board.
So nema17 just works fine, i know the wiring can be different for motors, thats exactly why i measured which pairs are the coils, to not make mistakes here.

Im completely in the dark about what can cause this. I posted the config file im using. Can someone comment on the settings for the Extruder?
I cant think of anything wrong there tbh because it has all the settings the other drivers get.
Also tried different current settings, nothing changes. tried without interpolation, etc...

I connect the pancakes to my ender3 board which runs on a SKR turbo with Marlin and there the round pancakes just work fine, rotating both directions
No worries.

If you guys dont have a clue either, then just say it. dont bring up the wiring again
Thanks anyways.

i just changed the Y and Z steps to just 1 value for each letter. Does the same goes for the current settings and others like acceleration, speed?

After getting used to reprap and DWC i would hate to change back, a lot of work again, time and money invested lost. im not happy, the printer as good as finished but without the extruder running its useless sitting there.

jerolee

pancake (1).JPG

@dc42 Same measurement, same wiring like other picture but just more clear in view.

Hurray the stepper is rotating!!! The only thing i changed today was that i removed the (not needed) steps per mm in the config file from the Y and Z axis. This has nothing to do with E axis right?!
BUT appearantly , how in the hell this is possible i dunno, it fucked up the pancake steppers. Im completely stunned here and i would like to ask DC42 to try and replicate this. the more values registered under Y and Z axis must have crazed the system in some kinda really odd way.
Wow unbelievable, but im so happy it works now, jeeeez im speechless.

GeneRisi

@jerolee , I am only trying to help you out. My final comment is to look at the section of my config file below. The first extruder is a 20mm LDO pancake, the second extruder is a 17mm LDO pancake. They are both driving a version of the Orbiter extruder.

; *** Motion Settings ***
;
M566 X400 Y400 Z8 A8 C2 E300:300:300:300					; Set maximum instantaneous speed changes (mm/min)
M203 X35000 Y35000 Z1200 A1200 C5000 E3600:7200:3600:5000	; Set maximum speeds (mm/min)
M201 X6000 Y6000 Z400 A400 C400 E600:800:600:5000			; Set accelerations (mm/s^2)
;M593 F48													; cancel ringing at 50Hz (https://forum.e3d-online.com/threads/accelerometer-and-resonance-measurements-of-the-motion-system.3445/)	
;
; *** Drive Currents ***
;
;M906 X2000 Y2000 Z1330 C400 E1000:1680:1680:1380 I30 	; motor currents (mA) maxes
M906 X1400 Y1400 I30
M906 E1200:650:500:1380 I10
M906 A700 C400 I10
M906 Z1250 I70
;
M84 S120 												; Set idle timeout```

For the 20mm LDO pancake, I have the current set at 1200 mA. I also have a lower "maximum instantaneous speed change.

I hope this helps.

jerolee

@generisi
Hi, i appreciate your help ofcourse.

Good news mate, after i removed the values from Y and Z for the other motors on that axis in the config file, so only 1 value for Y and 1 value for Z, after upload and reset the pancake extruder WORKED!!!
It has nothing to do with the wiring but appearantly the values from Y and Z in the config file did some really odd magic preventing the 36mm round pancake to rotate. It did not bother a NEMA17

I ask DC42 to try and replicate this oddly behaviour as i gues its a very interesting conflict somewhere??

So my problem is solved and im a happy camper.

M584 X0 Y1:4 Z2:5:6:7 E3 ; set drive mapping
M671 X-60:-60:660:660 Y-48:378:378:-48 S7 ; set the coordinates of the 4 Z-axis drivers
M350 X16 Y16 Z16 E16 I1 ; configure microstepping with interpolation
M92 X80.03 Y79.84 Z1619.80 E704.00 ; set steps per mm
M566 X900.00 Y900.00 Z60.00 E300.00 ; set maximum instantaneous speed changes (mm/min)
M203 X10000.00 Y10000.00 Z500.00 E1000.00 ; set maximum speeds (mm/min)
M201 X500.00 Y500.00 Z20.00 E500.00 ; set accelerations (mm/s^2)
M906 X800 Y800 Z800 E800 I30 ; set motor currents (mA) and motor idle factor in per cent
M84 S30 ; Set idle timeout

GeneRisi

@jerolee I am glad you figured it out. Check my settings for current, etc. they are working well for me so far.
Gene

droftarts

@jerolee What firmware version are you using? Send M115 and post response.

Multiple values on gcodes that should only have one (like M350, M92, M566, M203, M201 and M906) used to cause problems, but I thought had been fixed in more recent versions of the firmware. If you're using the latest and it's still causing issues, I'll raise it with @dc42 for him to check. Also, @PCR 's suggested changes included setting single values for each axis, so if you'd tried all of what he suggested, it should have worked. My suggestion (if I'd read this thread earlier) would have been to create a blank config.g with one motor defined, and test. You can also send M98 P"config.g" to check your config.g for errors, though I'm not sure it spots multiple values being specified when they shouldn't be.

Next time you post a problem, it's best to post what hardware you are using, firmware version, and your config.g in your first post; it will make helping you a lot more straightforward, and quicker. See https://forum.duet3d.com/topic/5909/

Ian

jerolee

@droftarts
RepRapFirmware for Duet 2 WiFi/Ethernet FIRMWARE_VERSION: 3.3 ELECTRONICS: Duet WiFi 1.02 or later + DueX5 FIRMWARE_DATE: 2021-06-15 21:45:03

I did the changes PCR suggested otherwise i would never have found the solution
@PCR
Thanks mate, by removing values from the config as you suggested, the odd behaviour of the software was gone and round pancake worked.