[xyz-cnc]

Hey guys

First a little background. I've designed and built a few CNC router tables over the last couple of years, so I'm not completely clueless. I've got a good combination of PC, driver, inverter, spindle, etc etc that work well together and are very reliable - I use my table daily.

Just before X-mas, I built a table for a friend of mine. About a month later, he tells me he has been snapping endmills because the spindle stalls while cutting. Surprised, I go to have a look. At the time, he is cutting MDF, 1/4" single edge bit, doing 1/8" deep passes at about 80IPM at 12,000RPM (IIRC). In other words, the spindle is under little load. After 4 to 5 minutes cutting, sure enough, the spindle sound changes, like it's missing or cutting out. A second or 2 later, it happens again, then about 5 seconds later, it rapidly slows down & stalls.

It's a 2.2Kw water cooled spindle with a huanyang (sp?) inverter. I programed the inverter and tested everything before he took it. I know the settings are correct, as they are the same as my table that has been running for almost 2 years with zero problems...

So we re-zeroed and ran the program again, but this time it was following the previous cut lines, so it was under no load at all, and it still stalled a few minutes into the program, but it happened at a completely different part of the program. The spindle was not even warm, and it was not seized. Pressing the F5 key turned it on again and it continued to run for a bit, then missed and stalled again. I was watching the inverter, and as the spindle slowed and stalled, the inverter display showed the RPM dropping rapidly.

So I thought lets swap the inverter. I pulled one out of my machine and put it in his, ran the same program but nothing changed, spindle stalled again. I then replaced the spindle with another good spindle, but the problem persisted. While I was at it, I replaced the driver with a known good one, but no luck. Finally, I connected up my PC to his table and ran the program again but, you guessed it, spindle still misses and stalls after a few minutes of use. We ran a different program, but it made no difference, spindle stalls after cutting for a few minutes....

All the wires are shielded, all connections properly terminated, all plugs tightened, everything is exactly the same as every other table I have built, and every table has performed flawlessly except this one.....I'm out of ideas, what could be causing this to happen?

[JerryBurks]

Since it looks like the spindle VFD is controlled by the PC, did you check the control line for intermittent or the BOB or controller for a defect? That seems to be the only thing you did not swap.

[xyz-cnc]

I replaced the driver, so that rules out the BOB. However I did not replace the printer cable so that is worth a try. While I'm at it, I'll also remove all the other cables and re-route them through the cable chain. He did an OK job installing them, but not as neat as I do it. I doubt it will change anything, but it's the only thing that is left to do.

Thanks for the suggestion

[ger21]

What are you calling the driver, and how does that rule out the breakout board?

How are you controlling the rpm and spindle start/stop?

[xyz-cnc]

Sorry, I should have mentioned it has a gecko G540.

Obviously you know it has a built in BOB which has a 0 to 10v analog output. That, along with VFD out and VFD gnd are wired to the inverter. Spindle on/off and speed are controlled through Mach3. He is using Aspire to generate the Gcode.

I spoke to him again this morning, he tells me he straightened out all the cables inside the cable chain but it made no difference. He ran it some more, just cutting circles in chipboard to test it and it continues to do the same thing.

One thing to note is it seems to run OK for about 4 mins, then begins to miss/bog down and stall in the next min. He noted that if he restarted the spindle by hitting the F5 key and going back about 20 or so lines in the Gcode, it would stall in almost the same spot. If however he re-zeroes and rewinds the Gcode back to the start, it will take about 4 mins for it to happen again...

[Christian Knüll]

Hi,

I'd connect a Voltmeter to the 10V spindle speed input and see if it drops together with spindle rpm. Then exchange everything until the culprit is found.
(And check the grounding wires - maybe the input is more or less floating?)

[fixtureman1]

Have you tried running his file on your machine to see if there is anything in the code that could cause this.

[JerryBurks]

Hi,

I'd connect a Voltmeter to the 10V spindle speed input and see if it drops together with spindle rpm. Then exchange everything until the culprit is found.
(And check the grounding wires - maybe the input is more or less floating?)

Or maybe even disconnect the control line and operate the vfd manually for test purposes.

[xyz-cnc]

Have you tried running his file on your machine to see if there is anything in the code that could cause this.

No, haven't done this. But we have tried running many files on his machine, and they all do it if they are longer than 5 mins.

[xyz-cnc]

Or maybe even disconnect the control line and operate the vfd manually for test purposes.

Yes, tried this already. Changed the settings in the VFD to manual control, set the spindle speed to the default 24000RPM and ran the program. Still happened. We did it twice to make sure, it happened both times....

I think today I will swap the whole controller enclosure over to a new one. Not sure what else to try....

[fixtureman1]

have you tried just running the spindle without even turning the controller on for 10 minutes to see if it is the vfd or spindle

[JerryBurks]

Yes, tried this already. Changed the settings in the VFD to manual control, set the spindle speed to the default 24000RPM and ran the program. Still happened. We did it twice to make sure, it happened both times....

I think today I will swap the whole controller enclosure over to a new one. Not sure what else to try....

Wow....that is weird. Maybe you let the Magic Blue Smoke out and need a refill kit now

[harryn]

When problems are really wierd like this, it often points to RF or grounding related. I wonder if somehow his system is picking up something from the building ground, or if maybe the grounding is a solid copper wire vs the needed stranded wire to carry RF to ground ?

[ger21]

Any chance the G540 is overheating? I've seen a few other posts with non spindle issues in the last month that were caused by overheating G540's.

[neilw20]

Is there a small voltage regulator, somewhere in the system, that overheats?

[bushwakka]

Hey guys

The spindle was not even warm, and it was not seized. Pressing the F5 key turned it on again and it continued to run for a bit, then missed and stalled again. I was watching the inverter, and as the spindle slowed and stalled, the inverter display showed the RPM dropping rapidly.

If this is the case, then something is telling the inverter frequency to slow down. The haunyang inverters are open loop, and have no feedback to tell them what speed the motor is running at (just assuming a slight slip frequency below synchronous speed).So the lowering frequency is because the frequency setpoint is being lowered.

So there is probably something going on in your 0-10/0-5V spindle speed control loop. I'd suggest the volts on this circuit are dropping back to zero, most likely to something heating up or overheating.

However, I don't understand why it is still stalling in full manual control.

Paul.

[fixtureman1]

When I posted to just run the spindle I mean do not run the program do not even turn the control pc on and see if it slows down. I think it may be the VFD or spindle that is at fault.

[ger21]

I think it may be the VFD or spindle that is at fault.

He said he's already replaced both.

[xyz-cnc]

have you tried just running the spindle without even turning the controller on for 10 minutes to see if it is the vfd or spindle

Just running the spindle while it's stationary doesn't make it stall, only while it's moving

[xyz-cnc]

When problems are really wierd like this, it often points to RF or grounding related. I wonder if somehow his system is picking up something from the building ground, or if maybe the grounding is a solid copper wire vs the needed stranded wire to carry RF to ground ?

Good point, the machine is in the shed which was just recently rewired. However today he ran an extension lead to the machine from a completely different circuit from the house, but there was no improvement.