The Dreaded X axis stepper error using UC400ETH and UCCNC !!!

[Clive S]

Guess what.... all skips happening on parallel moves..... ! Strange

I you sure that this is not binding up or overtuned

[dachopper]

I am positive it is not mechanical, if I leave spindle turned off, it does not skip. If I turn spindle on and run job in the air, it jams under no load, and if I try to force it to jam, I can't, so it's 100% electronic interfere

[Clive S]

Guess what.... all skips happening on parallel moves..... ! Strange

what do you mean by this

[dachopper]

It;s the type of Cut the machine is doing when the skipping happens. Its a a sideways X movement with vertical Z at the same time, repeat x 1000

Doesn't seem to skip doing a normal constant Z level manoeuvre....

Now I must be 3.5 hours into this next cut, so far no issues.

[m_c]

Given the 'zapping' I would be checking you do have a good earth. The normal easy test is one of those plugin testers with LEDs on, but given you're down under, I'm not sure what would be available. It's also worth running the machine at the same time as having the tester plugged into the next nearest socket, just to make sure a good earth remains with things under load.

If you don't have a good earth, then the various filtering capacitors typically found in mains connected electronics (VFDs especially), can cause a voltage build up, which will earth through the nearest thing, which can be the person in contact.
You may need to install a new earth rod, or figure out why you've got a poor earth.
(does Australian power use L,N+E, or a centre tap L,L,N/E? to get 110V and 220V?)

[dachopper]

This is so frustrating, so I had another 4-5 skips / jams on the x and z axis doing this parallel job, so I reduced the speed, and the acceleration by 50% , still jamm, turned off the spindle, no jam. Put the spindle power plug and spindle / vfd cable earth on separate circuit ( not going through control box ) still jam.

Then I disconnected the spindle earth ( which runs inside the vfd cable so is probably a massive noise carrier, from the spindle and IE from the machine, and so far no jams.

This ground cable was effectively attaching the vfd noise onto the machine. Since the stepper motors are on the machine, they are then subject to the effectively AC pulse made by the VFD earth, and by the vfd cable shielding.
...


Update:
So disconnecting the VFD ground wire from the frame got me much further into the cut, infact the two medium sized parallel ops finished correctly without any issues, then I ran the really big one, it had a single skip 80% through. So now I'm thinking because the frame is now not grounded as I removed the spindle ground, I'm thinking of running a separately shielded spindle / Z block ground, to get rid of the static or whatever build up is occuring. Could Cat5 cable be useful in that scenario? I'm just trying to find a way to let the noise / static discharge without inducing any further issues...

Nick


So

[m_c]

I think you may need to add ground wires between the various parts of the machine, to ensure everything is grounded.
I've just noticed in your photo, that you're using aluminium extrusion, which will be anodised, and anodising is a brilliant insulator.

Did you try running the machine with the VFD cable screen connected at both ends (and check that with it connected, there is continuity between the spindle body and the VFD earth point)?

[dachopper]

Hi MC,

I measured the voltage that the shielding on the VFD cable produced when the Spindle was turned on, it's was straight to a 40 Volts AC signal, between the shielding, and the CnC frame. When I connected that shielding to the frame, the steppers went ape shi*T. Motors were stuttering and I could " hear" the motors struggling to just run normally. Now, the machine wasn't grounded at this stage, but that was the effect of attaching the noise directly onto the frame of the machine.

I then unattached that end of the shield, and tried to attached what i thought would be a suitable ground ( it was an unused wire from inside the VFD shielded cable ) Big mistake - now this wire WAS on the Earth ground at one end, running through the VFD cable, and attached to the Z axis / stepper motor mount at the other end. I think all I did, was make a big long antenna inside the VFD cable, which even though it was grounded, was helping to turn the machine into another noise generating machine.

Then I got desperate after about 5 jams in 10 minutes, so I undid the " VFD " ground wire from the spindle / Z block and this made a big difference, but now the machine is not grounded any more.

So I am going to run another two grounds, one from the machine frame to house ground ( which is the same as the start point, but from a different socket) , and another shielded cable up the cable tracks and onto the spindle/ z block, which will then go to the other house ground.

I keep reading about these star grounds, but the fact is the VFD cable is creating an AC type noise signal, and If I run that into my electronics enclosure base plate, what is stopping it going straight back up my Stepper controller shielding which is connected to the same base plate? hence I'm connecting these two at a different socket of house earth, in the hope it weakens before reaching the base plate again.

I don't think Im going to connect the VFD cable screen back up to the frame, it just seemed to make the situation far worse.

[m_c]

The entire point of the VFD cable screen is to absorb the EMI from the VFD cables, and drain it to earth. Due the VFD cables carrying a fast switching AC signal, an AC signal will be generated in the screen.
The fact that there is a voltage present, tells me that's it not correctly grounded. And also that connecting it to the machine frame makes things worse, tells me the machine frame isn't grounded correctly. If the screen was correctly grounded, then there should be negligible voltage in it, as it should be getting drained to earth without causing any problems.

Which brings me back to my earlier question, are you sure your power supply earth is correctly earthed?

[Clive S]

Which brings me back to my earlier question, are you sure your power supply earth is correctly earthed?

This is what I mentioned several posts back but you make a good point re the anodised stuff so as M_C has said ground each part ie Z plate to gantry to machine frame to Star point