Hello,

I really hope you can help me here. I have a problem where my stepper motors (well more the motion of the steppers) randomly stutters under load. Sometimes it just stutters a little and then carries on, other times it stutters then stops and then there is the times where it stutters skips some lines and continues and ruining everything.

The funny thing is that when it moves in a completely different direction (after the stutter and the stop) Mach3 actually tracks it motion correctly.

Here is a little video of the problem:
https://youtu.be/URxRGks_US0
Running in air there is no problems, but under load the problem appears.

Running windows 7 x32 - Have a parallel port but not in use - using UC100
Steppers are Nema23 WT57STH115-4204A 425oz 4.2A
The system feels like it can pull through walls, so I am quite sure it's not power / torque problems.

I have done the following so far:
Added RFI filter on the VFD
Used version 066 (final) and 040 of Mach3
Gone over the CV settings a milion times - Still havent found a good answer.
Shielded cables are: Spindle - Motor - Limit switches - USB cable

I am really at a dead end here, so I hope you guys can give me some input to find and correct the problem.


Here is some images of the setup:

First problem is you are using mach 3, V0.066.... it has known problems and should be not used. Last stable version is 0.062.  Second problem is I don't see any common grounding (star) between components.   

Yes I have also tried the v0.040 version (as suggested by another guy 2015) but will try the v0.062 version.
As per grounding, everything is grounded to the common ground, also the shields from the spindle - motor and limit cables. The mounting plate is used as a ground, so some of the cables are grounded to the plate instead of the terminal.

Please don’t take this the wrong way, I am not being picky just making a few suggestions regarding your wiring based on what I would try…

The inputs wiring (Estop, switches etc.) needs to be shielded right up to the BoB then the shield taken to the single point Earth. The switch ends of these cables need to be shielded close up to the switch but the shield left unconnected.
Earthing to various points on the metal backing plate is not a good idea – Earth to the single point only.
It looks like you are running a.c. wiring together with your TTL level signal wires – not a good idea.
Move the UC100 away from your a.c. wiring and perhaps connect it direct to the BoB without that extension cable.

Tweakie.

Yeah... I was kinda coming to the same conclusion. Although I think I will rebuild the large cabinet into two smaller with the BoB in one and the motor drivers in another to separate it even more. I was just hoping that there where someone who had a simpler solution... with a full time job, my own small business on the side and a kid around 1 year I have to rationalize my time quite heavily.
Didn't know that using the mounting plate to ground to wasn't a "good" option. There are no ground loops of course.
And I also didn't know that running a regular AC wire next to signal wires could do that much... the cabinet was also much cleaner wired until I remembered the 3-4 components that I forgot

I also agree with Tweakie, separate your low volts cables from the mains volts cables

but you mention stuttering! I regard stuttering on my machine (homemade cnc router machine) a question of speed I get stuttering if I push the axis motors too hard, also on my old setup I got stuttering because the brushes on my  router were sparking, I dont suppose there are brushes on your spindle?

PS nice panel did you do it yourself?
 I didnt like the wires in the speed controller hanging down like that possibly putting strain on the cable connectors, but more than likely thats not the problem its just me!!!!! nitpickin lol

Ye, will separate the ac and dc, and add some more rfi / emi filters to some of the other electronics.

Well I call it stuttering because I don't have a better word for it. It is more like a short (very short) pause of motion, the pauses then comes more frequently until it suddenly stops.

The spindle have bearings, and there can occur "bad" currents in the motor, but if that happens it is not the stuttering I should worry about ☺

Yes, I build the panel my self since I have access to everything where I work. We build high end UV systems, and the smallest control cabinet we have is the one I build my cnc controller in. We will see if I shot myself in the foot with that decision.

Yes the vfd cables doesn't look that great, but there are not any strain on the cable and it is only temporary. Moved the vfd out of the cabinet to see if that helped with the noise issue, and it didn't ☺ but now it is not getting back in since I have become paranoid with noise.

Success!!! Pulled out most of the wires and separate ac and dc completely, plus added an rfi/emi in front of the stepper drivers. I also drilled a hole on the side of the table and pulled the vfd cable down under the cnc to get it further away from the control cabinet. But it was most likely the separation of the ac and dc.
Now the only problem is that I have some leak current to ground which some times takes the hpfi relay on turn on or shut down. But the cnc is running flawlessly so far, have run 2x 30 min jobs without an stutters or stops.
So thank you for the support, it was really great to get some feedback.
New Christmas star's in oak