Stepper motors trip power supply only when run through linux cnc

I am having an issue that I just can't wrap my head around.

For some reason, my z axis motor will trip the power supplies internal protection. This happens reliably when I run my motors at 75 in/min and occasionally at about 50 in/min. (I have a 20 tpi lead screw so its something like 1500 rpm right?)

This only happens when I am using Linux cnc to control the motor. My stepper drivers have software so I can control it manually. If I drive them directly from that at 2000rpm and 500 r/s/s for acceleration, it doesn't trip. If I switch the connection so that I control my z axis motor as my x axis, it doesn't trip. If I try my x axis motor as the z axis, it will trip.
I might add that I have completely remove my motors from the machine to make sure it wasn't an issue with the lathe itself.

On occasion, the rpm of the motors will drop briefly but only when running linuxcnc. This happens to both motors on both axes.

According to the driver software I am using to control the motor, it pulls a max of about 2.5 amps. I couldn't get data for the moment the supply trips due to it taking about a seconds to update. I also tried hooking a cheap multi meter directly inline between the stepper and the psu and it only read .87A. I tried a clam meter between the psu and wall. It read 0.3A.

I am using CL57Ts driver and a 5.8A stepper motor. I have a 36v, 5.9A power supply.

Attached should be my entire config file for linuxcnc.

I previously posted about my machine "missing gcode". This was the actual issue.
I am completely baffled as to what is happening. My gut reaction is that its a psu issue but it seems to be tied to how Linuxcnc is controlling my z axis. Any insight is appreciated.

Not always. It will happen when I hold the arrow key to tell it to go in one direction.

As a side note, Is there gcode I can use for this instead to make sure the keyboard isn't losing connection. It's a cheap keyboard that stops working when I get too close to the stepper motors.

Nevermind, I remember it happened halfway returning to home.

I tried using a multimeter to see if the voltage fluctuates while the motor is running and it doesnt appear to.

If was only when it decelerated, would it be a setting in linuxcnc?

Running higher acceleration and deceleration value when controlling it with the driver software didnt cause the power supply to cut out.

This issue make no sense to me. I have successfully run the motors at extreme acceleration values and high velocity when controlling them via the driver's software.

It has to be something wrong specifically with my z axis software set up since it effects the motor is wired to the z output on my mesa board. Swapping in a different motor keeps my problem.

The only thing I can think might cause this is the brief drops in rpm I've seen on both axes. That might be enough deceleration to cause the motors to trip the over voltage. It doesnt really make sense to me though as to why the x axis wouldn't have the power supply tripping issue.

The only difference in my set up of the two in pncconf was that I turned backlash compensation on for my z axis.



If I can't find a solution to my issue, I might as well just buy some new power supplies.
Any suggestions?
I kind of want to go for an unregulated design but I have no experience with them and am having trouble finding a decent supplier.

For spikes on powersupply you might experiment with capacitors in parallel. Can also make it easier to measure if you come into overvoltage protection when getting "back supply" from breaking.
You write that you have emv problems with your keyboard.. Are you sure that this is not part of the problem. Do see the same thing if you running the test via a small gcode?
Mike

I've had issues where It cuts out halfway through a rapid so it's not necessarily the keyboard.

I like the capacitor idea. Any idea how I size one for the task?

As high elco you find. I love to have in the x0'000 myF. Plus some small ceramic in parallel. I also have cheepo V plus A meters between psu and capacitor bank. Drivers directly connected to capacitors.
If i read that your stepper make emv i gues that you are not using shielded cables..
So at least try to stick some ferrites into motor cables.
I know you can do without capacitors, ferrites, shielding, proper earthing... But i prefere to do first time right and having never problems..

P. S.
With one of my versions running my machine guarantied -> no ukw radio running. - > low WAF (woman acceptance factor) ;-( nogo = go back to the bench and redo.

Mike