Fanuc 0M Servo Motor Hunting For Position

Advertise with us
shop deals on amazon
Shop deals on ebay
TaoTech_ElectricianForum2
Talk Electrician Forum

Help Support Talk Electrician Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Sidewinder

Sidewinder

The Snake™
Supporting Member
Joined
May 6, 2010
Messages
15,375
Reaction score
401
Location
UK
Well...

I've got a problem with a sliding head CNC lathe, Fanuc 0T control, M series d.c. servo drive & motor.

Poor surface finish on the part & a "grumbling" axis.

Believed it to be the x-axis motor, so got it off & away to a motor shop that had done work for me in the past.

They pulled it down & the bearings weren't good, & the brush gear was heavily contaminated.

Full wash & brush up, with new bearings, ran fine, refitted seemed OK.

Drive taken back also & ran OK on test bench.

Couple of weeks later noise back again.

Discussed with client Engineer rebuild of 2nd motor, as it was same age & same use, they agreed.

Rebuilt 2nd motor.

Fault reduced.

A few days later overload alarm (shared overload).

Went to site, traced overload alarm to TM O/L on drive.

Checked overload changed setting through full range and reset to factory no more trips.

However, rough surface finish remains.

Spindle was growling, spindle motor OK, checked out spindle layshaft, rough NDE brg. rebuilt layshaft, spindle itself seems OK.

Ran machine rough surface finish, spindle quieter.

Now jogging X-Axis, this is vibrating quite badly.

Left machine powered up and not running for a few hours, motor very HOT!

Remove motor return to test bench runs fine, strip & check NDE brg. a little rough, replace & rebuild.

Align magnets with brushes as a double check even though the parts are marked, end up in same place!

Return to machine still rough in jog.

Swap X & Z axis motors, X now smooth, no vibration on Z, but, Z is now resonating when holding position.

Gut feeling is resonance is at same frequency as noise through X axis.

No drive changes to affect tuning.

Also as the motor has been moved from axis to axis, the feedback interpretation, control channel, cables, power stage etc. have all remained with the original axis & the fault has moved, thus it points to the motor, if it were a drive tuning issue I would have expected the fault to remain with the drive/axis rather than with the moved motor.

The only thing I can find is rough encoder bearings, however, the encoder pulses are all fine on a scope.

The encoder feeds direct to the CNC which outputs a speed tacho value to the drive for the drive to control the speed loop, whilst the CNC controls the position loop.

Any ideas, anyone, because I can't think of anything else!

 
Looks to me like you've confirmed it's the motor, but not something you or the motor place can trace. Is a new motor a realistic option? (Not familiar with the kit, sorry if this is a stupid suggestion)

 
Motor is basically obsolete mate, it "could" be the feedback package, I'm investigating a replacement now.

Short of testing each individual rotor winding, or getting it on to a "growler" then I can't see a solution.

I'm thinking about getting the other axis motor off & running a few comparison tests.

Being on my own on site sometimes it is hard to move away from your own train of thought.

Yes Noz, the fault goes with the motor, it is quite small, only perhaps a 63 frame size maybe 70.

 
Last edited by a moderator:
Ha,, I was going to ask if you'd tested the motor windings,, but I thought to myself "don't be daft" and deleted the question

I think full continuity, IR and growler testing of the good and bad motors to give you some comparisons would be good..... it'll only take a very small breakdown in a windings IR to show a fault like this

 
Trouble is Noz, with it being d.c. it has a commutator, and thus is bother to check!

The motor shop have experience of these & have made suggestions, we have & do work together quite a lot, so we know what each other do as it were.

I may just go & do an IR on the assembled motor whilst rotating the com now & see what I get!

Then I'll come back.

 
Yes admittedly, but it was basically cold yesterday when it was hunting, it only heated up after a few hours of being sat there & powered up, I suspect that the heating is due to the hunting.

 
Sorry, but 63 and 70 frame mean nothing to me,,, remember I'm coming at this from my time fixing avionics,, we had bombing computers that used servo and synchro motors,, but if we had a problem like this we'd just bin the part

We also had faults that would only be evident when the item under test had warmed up a bit

 
Standard commercial motors have "frame" sizes, this is the shaft height from the base as it were for a standard foot mounted motor of the type.

e.g a 100 frame size would be 100mm radial to the C/L of the shaft from the motor base.

I've not actually checked this one but it is around a 70 frame "ish".

I appreciate the thermal thing, but it was hunting immediately it was connected, so basically cold, i.e. room temp.

 
I think as much, when I have issues I tend not to give out my ideas as it can colour opinions too much!

The encoder is a 5V TTL type with 2 channels & inverses @ 2500ppr plus marker & inverse, so pretty basic.

Tacho signal for velocity loop is generated in the CNC from the encoder signal.

Tacho signal was rough when scoped, but at that time I did not have access to the machine wiring diagrams so guessed that the motor had a TG which it does not!

Would like to swap the FB packages from the 2 axis motors, but not "quite" that simple, but it may come to that!

 
Com is fine does not need undercutting or skimming.

Even I can tell that! ;)

As far as growler etc. goes the guys are just setting up, they have left another company, and gone alone so we are helping each other out, & they don't yet have a growler!

Can't find anything definite to prove it is the encoder, but, at this stage it seems the best bet, waiting for a price on a replacement.

 
NozSpark said:
Sorry, but 63 and 70 frame mean nothing to me,,, remember I'm coming at this from my time fixing avionics,, we had bombing computers that used servo and synchro motors,, but if we had a problem like this we'd just bin the part

We also had faults that would only be evident when the item under test had warmed up a bit
Click to expand...
going on the possible consequences, that was probably wise.

now, me or Deke, coulda sorted it with some cable ties, some duct tape, and a couple of bits of 0.75 flex,  :)

 
Have you tried re tuning the drive, oscillating or hunting is almost always too much gain.

As it's old dc kit, you can probably adjust that with a trimpot somewhere.

I'm talking here about reducing the gain of the velocity loop, which will be closed within the dc drive using the simulated tacho output.

Have you actually checked with a scope that the simulated tacho output is actually working? if not then it could be running without velocity feedback so no wonder it's getting in a huff.

 
Thing is PD, drive tuning does not move "overnight" really.

There is no backlash to speak of in the mechanics & the position & velocity loops are from the same feedback.

The machine holds position so the position loop must be functioning, the velocity loop is fed from the NC, and without this the drive shuts down, so it is working.

Yes there are tuning pots etc., been through all the pre-tuning checks with a scope, and according to the drive/motor manuals the tuning is correct!

Going to change encoders this weekend to see what happens.

It "could" be an intermittent/poor connection somewhere, but, there are a lot to choose from on these damn little CNC lathes!

 

Similar threads

Sidewinder
Not Looking Forward To Tomorrow
Replies
10
Views
902
Evans Electric
Evans Electric

Latest posts

Top