#1 – bdring 于 2024-12-18

Just to clarify

– It it does not work with 3.9.0 and 3.9.3.
– It works with 3.9.1
– It works with versions earlier than 3.9.0

#2 – bdring 于 2024-12-18

What closed loop motors are you using?

#3 – jkingdcs 于 2024-12-18

> Just to clarify
>
> * It it does not work with 3.9.0 and 3.9.3.
> * It works with 3.9.1
> * It works with versions earlier than 3.9.0

this is correct

#4 – jkingdcs 于 2024-12-18

> What closed loop motors are you using?

nema 23 from stepper online 3nm model 23HS45-4204D-E1000 closed loop motors 4.2A per phase. motor controller is cl57T v4.1 closed loop drivers

#5 – bdring 于 2024-12-18

Is it possible to place an LED on the direction signal of the controller on the axis that is having trouble?

#6 – bdring 于 2024-12-18

!image

#7 – jkingdcs 于 2024-12-18

> Is it possible to place an LED on the direction signal of the controller on the axis that is having trouble?

yes, I can put an led, oscilloscope or multimeter

(I can jog for about 10-20mm before the gantry skews so much it jams)

#8 – bdring 于 2024-12-18

The direction signal is persistent. just jog a few millimeters. The state of the direction signal will stay until you reverse.

#9 – MitchBradley 于 2024-12-18

It is very strange that there is different behavior between 3.9.0 and 3.9.1. The changes between those versions are in an area that is very far removed from the step/direction generation code.

#10 – jkingdcs 于 2024-12-18

Y motor connected to motor socket B on rootcnc : (bad motor always turns same way)
multimeter negative on dir and positive on ref = -4.95V (motor connector B)

Y motor connected to motor socket Y on rootcnc : (good motor operates as expected)
multimeter negative on dir and positive on ref = -0V (motor connector Y)

#11 – MitchBradley 于 2024-12-18

I presume that the voltage on the good motor socket changes depending on the command direction, while the bad one does not change?

#12 – jkingdcs 于 2024-12-18

> I presume that the voltage on the good motor socket changes depending on the command direction, while the bad one does not change?

correct the motor on socket B has -5V dir to ref regardless of the direction of travel Y requested,.

I just put 3.9.1 back on, switched off and on the power and jogged in the y positive direction and BOTH motors work correctly and both motors have 0V between dir and ref when moving positive Y and -5V when moving negative Y.

#13 – MitchBradley 于 2024-12-18

In the previous test, where the B voltage did not change, was it 3.9.0 or 3.9.3?

#14 – jkingdcs 于 2024-12-18

> In the previous test, where the B voltage did not change, was it 3.9.0 or 3.9.3?
3.9.3
(I had exactly the same motor behaviour from 3.9.0 but stopped using it when it got pulled from the releases)

#15 – jkingdcs 于 2024-12-18

I should mention I am using firmware (3.9.0 3.9.1 and 3.9.3) I compiled from sources in github and not running the pre compiled firmware in https://github.com/bdring/FluidNC/releases.
The only change I made was to uncomment the 8MB flash option and comment out the 4MB flash option in platformio.ini. the compile environment and version of visual studio code was the same for all. (I checked the processor and it is a 8MB flash version)
I am happy to test 3.9.1 and 3.9.3 original firmware downloads from git hub and/or compile with the 4MB flash option if that would help

#16 – MitchBradley 于 2024-12-18

Yes, please do test the released versions, just to eliminate variables.

#17 – jkingdcs 于 2024-12-18

using the original downloads from git hub
3.9.1 works as expected no issues
3.9.3 ALSO works as expected…
im going to try recompiling 3.9.3 but with the 4BM flash option and see if it works or fails.

#18 – jkingdcs 于 2024-12-18

UGH this is crazy,
I switched rootcnc back on with the firmware 3.9.3 downloaded from git hub and the Y axis problem has struck again. I did nothing expect reboot the pc and leave the rootcnc off from sometime.

#19 – oliof 于 2024-12-18

sounds like s loose crimp on one of the motor wires to me.

#20 – jkingdcs 于 2024-12-18

3.9.3 4MB flash from git hub does not work after the second reboot.
if I have 3.9.1 flash installed, run install-wifi to flash 3.9.3 git hub firmware download and then quickly power cycle the root cnc
3.9.3 works BUT when I power cycle the second time 3.9.3 fails with skewed axis. All subsequent power cycles 3.9.3 precompiled from git hub does not work. I even tried reflashing 3.9.3 git hub again over the 3.9.3 git hub and it still does not work

#21 – jkingdcs 于 2024-12-18

> sounds like s loose crimp on one of the motor wires to me.

just put 3.9.1 from git works first time reboot still works reboot again still works….

ok if I swap the Y and B cables at the rootnc, then if its bad wiring the problem will NOT move from the left motor to the right motor. If the problem is the rootnc not outputting the direction correctly then the problem will move to the right motor
lets see…

#22 – jkingdcs 于 2024-12-18

I swapped B and Y motor connectors at the rootcnc,
with all 3.9.1 (compiled by me 4BM flash and from github) motors run correctly
with 3.9.3 from git hub installed the right hand motor now moves the wrong way, if it was bad left hand motor wiring on the left motor then the left motor would still be doing incorrect things regardless of if it is attached to channel B or Y
I believe the B motor channel direction is jammed at -5V Y with 3.9.3 after flashing if you switch off for more than a second or two.
I guess the next step is to see if channels A and C have the same direction problem.
I will also unplug the control lines from the rootnc to the motors and test the voltages with no motor wiring attached.

#23 – jkingdcs 于 2024-12-18

B channel direction is locked to -5V with 3.9.3 if hand compiled as 4MB or hand compiled as 8MB or using original git hub download, this is the case even if the motor control wires are disconnected from the rootcnc. (Apart from the single time I went from 3.9.1 to 3.9.3 and did a very quick power cycle) all other experiments are consistent. I can’t get 3.9.3 to work correctly on channel B.
channel B always works as expected in every 3.9.1, (hand compiled 4MB or hand compiled 8MB or download from git hub.)
the problem consistently moves from left motor to right motor and back when I swap the B and Y channel wiring and now the right hand motor is on channel B I can’t get the left motor to misbehave at all.
its rather late now and so I’m not going to experiment with channels A and C tonight and I will be away for two weeks visiting over Christmas so I will run some ore experiments in Jan.

Thank you Mitch, bdring and oliof for your support.

#24 – MitchBradley 于 2024-12-19

We are at somewhat of a disadvantage as I don’t have a Root Controller. Maybe Bart has one but I don’t remember him mentioning it. Nor have I found schematics online.

One thing that comes to mind as a possibility is signal integrity problems causing the I2S shift registers to misread the serial data. The 3.9 series runs the I2S at a faster clock rate allowing higher pulsing rates. Bart’s designs, which feature careful layout for good I2S signal integrity, have no problems at the faster clock, but who knows for the root controller? The fact that both Y motors are on the same shift register chip makes this somewhat of a long shot though.

#25 – jkingdcs 于 2024-12-19

> We are at somewhat of a disadvantage as I don’t have a Root Controller. Maybe Bart has one but I don’t remember him mentioning it. Nor have I found schematics online.
>
> One thing that comes to mind as a possibility is signal integrity problems causing the I2S shift registers to misread the serial data. The 3.9 series runs the I2S at a faster clock rate allowing higher pulsing rates. Bart’s designs, which feature careful layout for good I2S signal integrity, have no problems at the faster clock, but who knows for the root controller? The fact that both Y motors are on the same shift register chip makes this somewhat of a long shot though.

hmm, that’s an interesting thought which makes me wonder if its possibly a marginal I2S shift register…
I definitely want to move a y axis motor to channel A and then C to see if they are equally effected.

#26 – jkingdcs 于 2024-12-19

Some GOOD progress:
I just tested with one Y axis motor on Y channel (always reliable) and the other on (the previously unused) A channel and quick testing with 3.9.1 shows correct operation.
channel A also seems to work with 3.9.3. So far 3.9.3 has survived 4 power cycles and works correctly. I tried leaving a longer time between power cycles and still correct operation. This is by far the most reliable I have ever had 3.9.3…

I think there is now good evidence to suggest that either there is a signal integrity problem that 3.9.3 is triggering on channel B or the shift register that channel B is on has a marginal bit that 3.9.3 is triggering the stick. This might also explain the single correct operation I had with 3.9.3 earlier and then after rebooting it lapsed into incorrect behaviour…

I repeated the testing for channel C and the channel C dir pin is stuck at -5V for BOTH 3.9.3 AND 3.9.1 – C direction is on the third shift reg. So far I have not been able to get channel C direction to work correctly at all.

went back to channel A and it works fine with 3.9.1 AND 3.9.3.

So it looks like:
channels A,Y,Z,X work with 3.9.1 AND 3.9.3
channel C stuck -5V direction with BOTH 3.9.1. AND 3.9.3
channel B stuck -5V direction with 3.9.3 only.

I guess i should try a pre 3.8.0 to see if channel C will work at all….

P.S.
I found a schematic for an older version (v2.1) of rootcnc and it shows SN74HC595DR shift registers.

#27 – mattydboom 于 2024-12-19

This all sounds like my previously reported issue with v3.9.0 here #1364

That version was ultimately pulled, but I was experiencing exactly the same gantry skewing problem, and run a RootCNC ISO v3 with similar set up (except standard high torque and not closed loop steppers).

Bart and I did a bunch of tests over DM to try and get to the bottom but never did pinpoint the exact culprit….. and was fixed (and remains so for me) in all subsequent releases.

Just thought I would share in case I can be of assistance in trouble shooting with the Root controller, or if my previously shared config file could be used for comparison / testing.

#28 – jkingdcs 于 2024-12-19

> This all sounds like my previously reported issue with v3.9.0 here #1364
>

thanks for the info.
is 3.9.3 ok for you? which motor control channel is your second y axis motor on?

#29 – mattydboom 于 2024-12-19

Uploaded 3.9.3 firmware last night and behaves correctly just like every release (including the Pre releases) for me since 3.9.0.

Yes, my Y2 is wired to B port like yours. See attached pin mapping cheat sheet I created when compiling my config.yaml file. It may help you check for similarities / differences in your set up.

#30 – MitchBradley 于 2024-12-19