VFD in a lathe application 5

[itsmoked]

Hi, does anyone have the rundown on using a VFD'd motor in a lathe?

We've put a 5hp VFD specific motor into a Monach lathe. The original motor was a DC motor run by tubes! This was later swapped out for SCRs in a DC controller in the 80's. That's crapped-out just like the company that made it and as a last resort we put in this VFD induction motor that even came with an encoder.

I'm running it in sensorless Vector mode since we're not trying to do positioning or holding.
Tuned it successfully with nothing attached to it. Hooked it up to a gear box whose output belt drives the spindle.

The lather came with 1hp, 3hp, and 5hp options.

Everything operates fine except at lower speeds you can grab the spindle and stop it with your hand. While I don't expect massive low end torque that's seeming... under-the-bottom.

On a second note I see large currents being sent into the motor even if it's stopped. !4A FLA and I'll see maybe 10A, .8A, 4A on the phases when the motor is stopped. Is that the drive trying to keep the rotor somewhere known? Can that be avoided?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[TugboatEng]

If it has an encoder why are you not using it? I wouldn't think that your application would need one but this will provide the best performance for all applications. The only reason not to run an encoder is cost.

 

[itsmoked]

That's not the only reason. You need to hunt down an additional expensive board and install it into the VFD.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[LionelHutz]

Motor is actually taken down to zero speed and not stopped when you still see these currents, right? There is a difference. Stopped is VFD off.

 

[Gr8blu]

itsmoked: There are two modes involved with a VFD, usually. One is a region where the drive can effectively control power vs speed, and the other one - which is typically below about 5 Hz - is where it has difficulty. In the higher speed range, you'll pretty much see the torque you're expecting. At the bottom end, though - all bets are off, except that you're likely to see less torque than expected. One way to get around this is to acquire a MUCH better (and therefore more expensive by at least an order of magnitude) drive. Another is to just not operate "down in the mud" so to speak.

As to seeing current at nominal zero speed - that's precisely it. The drive is not "off", it's simply holding the shaft from spinning - kind of like riding the clutch on a hill. Remember, the control isn't all that accurate down there either, so the drive is using relatively coarse current adjustments as it "hunts" for the best position.

Converting energy to motion for more than half a century

 

[itsmoked]

Lionel; That's what I expected but nope! The external input set to STOP it still keeps the motor excited.. Still puzzled. The manual made mention of a timer one could set to stop, what I thought was this, when I wasn't even looking for it, but when I went back doing an extensive search I could never find it in an hour of hunting.


Hi Gr8blu; Thanks for that. I'm aware of the vulgarities of sensorless down below 5 hz or so. I set the minimum to 10Hz to avoid this region of mud. Didn't expect the gutless result. There is a torque-boost setting that ranges from 100% to 350% which I believe is being tried. I don't know the results yet.

I just wish the excitation would go away since one expects a lathe to spin freely when "STOPPED" and I'd like a stopped period to go towards cooling the motor where still cranking 10 amps thru it doesn't probably allow any net cooling.

I'm starting to wonder if I should go back to V/Hz mode and run in a PID mode, though, I'd think a PID mode would be for a fixed application like running a fan or a conveyor not an all-over-the-place application like a lathe.



Keith Cress
kcress -

http://www.flaminsystems.com

 

[LionelHutz]

PID for what, speed?

What make of drive?

 

[TugboatEng]

Have you considered that you may have an IGBT that has failed closed?

 

[itsmoked]

Lionel: CFW500C24P0T2DB20, VFD, 7.5HP, 230V 3PH IN, 230V 3PH OUT, 24FLA Running a 5hp motor.


Tug; Brand new so I'm pretty sure it's not "stuck on" though I hadn't considered it.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[TugboatEng]

Brand new? That puts you on the left side of the bathtub curve.

I'm thinking the stuck on gate only affects one phase. There are multiple gates in the inveter. Some time back I had a triac on a control board fail in short circuit. It controlled the contactor for a small air conditioning compressor. Twice we burned out compressors when they didn't shut down on high pressure cutout before we found the shorted triac.

 

[itsmoked]

Hi Tug; I see your points, they're pretty good. One point though is that those measurements I mentioned above were taken between several jogs and they changed around to different phases. For instance the 10 or so amp phase moved and sometimes the max was more like 6A and other phases got higher. They did seem to kind of algebraically always add up to some consistent value.

I can shift the drive to V/Hz and that would kill any and all static currents. If that happens then a 'stuck' pass-element wouldn't be.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[LionelHutz]

I'm not familiar enough with that drive to know what it might be doing, but stopped should be off. There can be an odd mode occur where you set the drive to force the motor to zero speed before it stops and the motor regenerating can't be taken to zero speed causing the drive to get stuck trying to stop but never getting to the stopped condition where it would shut down. It might be that you've got this happening even though the load is not regenerating to cause that.

I do agree that something is definitely wrong if you can just grab the chuck and stop the machine with the motor running at 10Hz or more.

There must be someone at WEG with enough of a clue about that VFD to suggest some programming changes to make it work correctly.

A VFD has 2 IGBT in series across the DC bus for each output phase. If one IGBT is failed on, the drive will immediately blow up as soon as the other IGBT turned on. Since the IGBTs switch on and off at the carrier frequency it'd blow within milliseconds of being given a start command. Well, it might not blow up but rather trip on a hardware short, it's kind of a crap-shoot on which would happen. If it did fault and you reset and tried it again you'd greatly increase your odds of getting a nice explosion. You've got to know how a VFD works before suggesting how it might have failed.

 

[EdStainless]

We had always set a minimum speed.
There comes a point at lower freq where the waveform of the VFD and the winding style of the motor just don't like each other.
I would never have thought of trying to run down to 10Hz, or anywhere near that.


= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[TugboatEng]

Our winches use cam operated limit switches on the speed potentiometer to force the stop command when the lever is placed in the center position. However, I believe in itsmoked's case the drive is indicating that it is in fact stopped but still has output?

Can you hear the switching or carrier frequency in the motor when it is supposed to be stopped?

 

[itsmoked]

I did not hear the carrier though the motor is a NonVentedTotallyEnclosed model that does wonders for bottling up VFD excitation noise.

I did try hand turning the spindle while "stopped" and I could turn it but you could tell it was excited in some form as it would take effort to turn then advanced herky-jerky.

As Hutz mentions though, if one channel was stuck ON it would be crow-bared almost instantly.

I'm going to give another try searching for that "Note" in the manual I subsequently couldn't find again. I saw it while speed reading the manual the first time just looking for things that would matter for this job not actually realizing that note WOULD matter to me.

LionelHutz; Thanks for confirming the hand-stopping. I haven't messed with a motor this size at low speed in vector to have much judgement about hand stopping it. (Especially since it's being geared down some thru a gearbox!)

Keith Cress
kcress -

http://www.flaminsystems.com

 

[LionelHutz]

A motor might never stop if you set the VFD to current limit and the motor is being driven hard enough to overcome the decelerating torque the motor is producing with the VFD current limit. The VFD timer may be to protect against this where it will time out and eventually trip off if told to decelerate but it can't get the motor to decelerate. It'd be odd for the VFD to be thinking a stopped motor is still rotating, but maybe it is.

 

[TugboatEng]

I had one application where the motor would spin at 25 rpm when clutched out but once clutched in the gearbox load was enough to stall it. Nobody ever noticed anything but I could hear the carrier frequency faintly through the motor casing. I hooked up Connected Components and saw the motor speed. I never did get to troubleshoot that one as we transferred the vessel out of our fleet.

 

[cswilson]

Some vector drives will keep the "magnetization current" active even when there is no request for orthogonal "torque current". This way, when there is a request for torque current, there is not the delay in getting the rotor to magnetize.

One of the trickiest things about "sensorless" (actually, shaft-sensorless) vector control is calculating the rotor angle at very low speeds. The back EMF on the phases is so low that you have a horrible signal-to-noise ratio. This can mean that what is intended to be torque current is not oriented correctly.

I don't know if these factors are what you are seeing, but they should be considered.

Curt Wilson
Omron Delta Tau

 

[RedSnake]

You might try this for the STOP mode.
I am assuming that Coast is what we call free rolling.

P0229 Stop Mode Selection
0 = Ramp to Stop
1 = Coast to Stop
2 = Quick Stop

“Logic will get you from A to Z; imagination will get you everywhere.“
Albert Einstein

 

[TugboatEng]

If it's using DC injection braking then it may cause the situation described but if that were the case the drive status should be "running" and not stopped.