VFD on CNC problem?

I posted about the torque issue on the CNC forum so there is more to the story however this is the latest issue and I am hoping the amazing electric guys here can help out.
The machine is intermittently slowing down in a medium cut and the first time it happened I had my hand on the feed knob so I slowed the feed and it recovered and completed the cut. So this happened a few times. I checked and tightened all wiring and had to relocate the VFD so I could open and check the wiring. (it was on the back against the wall) The wires seemed fine except one control wire jumped out of its spot when I first tried to straighten out the wires. It was either doubled with the red or on its own next door. I could not read the number that was poorly written on masking tape. The terminal for the red wire which was labeled 17 was a bit loose so I put them both in 17.
Then tested the motor and it started to pulse at 2200 rpm, revving or pulsing up and down maybe 40 to 80 rpm as seen on the VFD readout. I then tried the white wire in spot 16 and there was no change in the result. Hopefully I did not do any major damage with my fumbling as I was not expecting one of the wires to jump out so easily.
Earlier in the day the motor amp draw tested a little low on one leg and the same on the other two and the motor sounded fine at 2200 rpm, no pulsing. Also the voltage at the motor was perfect across all three legs. The control is able to change the speed fine and at 3500 rpm the pulse is less noticeable maybe 15-30 rpm.

So any ideas or more info and testing needed? Is the VFD on its last legs?

More info would increase the chances of an intelligent reply.

Machine? VFD? Pics if appropriate?

Do you know what the wire comes from?

I don;t want to be a pain, but when someone asks about where a wire goes on some unknown thing, how would anyone know?

Need diagrams and part numbers.

You'll want to get a readout of some kind of commanded speed, actual speed, loading, and overload. The question is whether the VFD is slowing down because it's been asked to slow down, because the VFD can't keep up, or because the motor can't keep up.

Machine is an HHRoberts Topwell mill, 5hp. The VFD is a Baldor and the machine is running with an Anilam 3000 control that was updated from the 1000 series in about 2010 by the previous owner. It looks like the VFD was installed in 2010.
The wire in question went in one of two places and I was just explaining what had happened. I did not expect anyone to tell me which spot the wire was supposed to go as that would be quite difficult. I tried it in both spots and got the same result from the spindle so I am still not sure if it is in the right terminal.
The rpm readout on the VFD does not match the selected control rpm input, in that the VFD readout is higher. I have never noticed this before because I could not see the VFD readout so don't know if it was always like that. It appears to be responding correctly to the input from the control except for the pulsing issue and the rpm percentage control also works as it should, the motor smoothly changes speed instantly.
I can take some pics of what is there and I do have one good diagram of the electrics inside the cabinet but not sure if I have the info for the VFD.

If it is a control wire, the chances of damage are reasonably low. You'd have to trace the wire to see what it logically should connect to.

"Pulsing" is a question. IF a vfd is overloaded, and IF it is set up to avoid overload by slowing until the overload is no longer present, than you can have a repeating cycle of that behavior. Overload leads to slowing, which releases the overload, so it may speed up, etc. Depends on what programming selections were made and/or what is available. You should see that in the display, because it is commanded behavior.

The fact that you could recover by lowering feed seems right for that.

But a few rpm of change seems less likely to be that issue, although the change of speed may indicate a change by the VFD. If the VFD is sensing actual rpm, as in vector mode, then it is possible that it may be set to read motor, and not commanded, rpm. That would be a VFD-dependent feature, look in the manual.

Actual speed changing with load is not unusual, induction motors slip more or less under load. So small changes in load due to feed variations, etc, etc, etc, lead to changes in motor speed. The difference is that you may be actually seeing them on the display.

JST,
Thanks for the reply. To be clear the pulsing on the spindle is a new issue that has never happened before and I have not made a cut with the machine since it was noticed. More testing today and I will trace the rogue wire to find out where it goes. There are 22 terminals for control wires and about 6 of those are jumpered together in the middle section on the vfd.

When the slowing spindle occurs with a small end mill it is brief and recovers on its own quickly. The above example when I slowed the feed I was running a 3/4” endmill about 1” down taking a 30 thou side cut. Everything fine until about 10” along in a 20” pass. I had been making repeated cuts for at least an hour before the problem.

Not sure if it is related but about 1/2 hour before the first slow down I had a crash and broke a 3/4” endmill. ( part came loose, operator error) No noticeable damage and the part damage was removed in the final passes.

I finally had some time at the end of the day today to do some testing and wire chasing. I have discovered that the pulsing spindle only happens when I select an rpm, press start and then I can start and stop the spindle on the control pad. When in mdi mode the spindle runs fine, no pulse. I did not try a stored program but suspect it will be fine like the mdi mode.

I also found the manual for the VFD but I am no closer to figuring out if the rogue white wire is in the right place. There are three wires in the cable and the black/ red go to 9/10 at the VFD. I traced the wires inside the machine control cabinet where our rogue white wire goes to the forward reverse contactors for the motor.
The black goes to an unknown contactor and the red goes to a terminal block on the main circuit board.

While tracing wires a control wire broke off but it was clear which terminal it went to so was an easy fix. It was red/black with a green ground on the main circuit board. It went to terminal 1 and 2 on the VFD but there was no ground wire connected. Seems odd to have only one end of a control wire grounded? Probably doesn’t make any difference?

I will update tomorrow if I get a chance to run some test cuts to check the torque problem.

Grounding only one end of the shield in control wires is standard practice. It prevents circulating currents flowing in the shield (a ground loop).

SSwhere,
Thanks for that I learned something today.
I was able to do some testing today and the results were a bit surprising. The spindle seems fine at higher rpms, it doesn’t pulse or vary in speed at all. I was checking with a digital tach. However at 2200 rpm it pulses and does not sound good. What was surprising is that the commanded rpm on the control did not match the actual rpm. At very low speed, 1000 rpm it was actually reading 2000.

I did some test cuts with a 3/8” end mill making a full cut through 1/4” mild steel flat bar, no torque problem with several attempts. Spindle rpm was 3700 and at higher rpms the actual is closer, maybe a few hundred over.
In regards to the VFD I checked the manual and as far as I can tell the white wire that had come loose is connected correctly. It is part of the forward/reverse control and those functions are working fine.

So is the VFD on the way out? What causes the variation in speed output from the commanded speed? Or is it possible the problem is with the machine control?