[Leblondmakino]

Hi all,

I have a dc brush servo that refuses to work properly.
When I switch the machine on it rotates clockwise (looking from the motor shaft end) then trips with 401 and 410 alarm .

Sometimes it just sits there and then trips with 401 alarm.
It never does the same speed on start up more than once!!

I have a Fanuc 6 control with Siemens 6RB 20 transistor chopper drives.

The machine had a violent overtravell/shunt and thats when my problems started.

Is it possible that the case magnets have become damaged in some way.

I have done the usual checks on the tacho, encoder and windings and they all seem fine.

I've had the stack card and drive card checked and repaired by a specialist.

I swapped the motor to another axis and the fault followed the motor.

I had the motor checked out by a specialist and they say its fine??

If I reverse the tach polarity it runs away like crazy! so the motor will run!

Thanks for ant help in advance
John

[Al_The_Man]

Its seems it is fairly conclusive it is the motor if the problem follows when you switch them.
A 6 control can be fitted with resolvers or encoders on the motor, it may be this is the problem, especially if it starts to run away at start up.
Al.

[Leblondmakino]

Thanks Al I wasn't sure if this type of motor could become demagnetised..

I think I will get it checked again and may be try and replace it as a quick way of solving the problem..!!

John.

[Al_The_Man]

Thanks Al I wasn't sure if this type of motor could become demagnetised..

John.

Yes it can, rarely, but this does not cause it to run away.
There is a test I do for the feed back device that may work.
What I do is remove the armature leads so that the motor cannot turn, switch on and go to Hand Wheel mode, low resolution for that axis, and slowly turn the handle slowly, obviously the axis is not going to move, but you should see the display count go up or down, after a certain amount, you will get a Following error alarm.
Record the amount the display shows up to the alarm, e.g. .325"
Now for the tricky part, reset the machine, and this time crank the handwheel slowly but as you crank it, move the motor shaft or ball screw approximately the same amount the screen displays.
IOW move HW .2 , move BS .2 (it has to be in the right direction), keep inching the HW up at the same time move the BS in tandem.
The idea is, is to move the BS to keep up with the HW and the control thinks the motor is turning, and never reach the error point.
If you can go past the point of previous following error by doing this, this virtually eliminates or proves the feedback defective, depending on the result.
It may be worth a try.
Al.

[Leblondmakino]

Thanks Al...so in effect you are checking that the control is receiving the correct feedback signals from the encoder and from the tacho because the motor has no power and will not alarm the machine out.

Thats something I will try tommorrow.

One strange thing I am finding is that I get a different tach voltage every time I switch on and it can vary between 2volts and 0.002volt without me having adjusted anything on the machine! (tacho is 20v/1000rpm)
I have even had a -0.003 voltage a few times when the motor decides to turn the other way!

I have checked all the segments on the commutator and am getting 0.3 ohm between pairs all the way round.

Perhaps my volt meter isn't accurate enough to detect winding damage..?

Btw I swapped the tacho from the faulty x axis motor with the working Y axis motor tacho and the fault stayed with the X axis motor.

Time to do more tests!!

John.

[Al_The_Man]

The controller does not normally see the Tach, only the drive sees it.
The tach should only output when the motor is turning and is proportional to rpm.
The test I suggested does not concern the tach for this test.
Al.

[Leblondmakino]

I see Al will check this today and see what happens.
John.

[GrantM]

Assuming it's a +/- 10V Velocity command to the drive, Disconnect the Vel command to the drive, if it's now stable then it's very likely a problem with feedback (ie. encoders/resolvers), maybe damaged encoder optics? If it's near stable but slowly drifting, then adjust your drive offset.

[philserveng]

You've had some good advise here, but if the fault follows the motor....change it. Don't mess about with having it checked, they cannot check for all circumstances. The reason for different voltage on the tacho could be you are checking an pulse encoder or you have an ac brushless tacho. When the motor is running you will get a mean (median) voltage on your meter, this is known as RMS and is quite normal. If you have brushes on the tacho and motor remove them all and clean the commutators of both with a pencil eraser, this will remove the graphite from the comms. Make sure the comms are not grooved or the segments are burred over and touching, again this will cause motor instability.

[Al_The_Man]

One problem you may have with changing the motor is getting one.
Both Mitsubishi and Fanuc originally used Getty's DC motors, or at least a model that was based exactly on Gettys.
Getty's do not make DC motors anymore and both Fanuc and Mitsubishi sometimes quote delivery time in Months.
BTW I have interchanged all three types of these motors in a pinch, Gettys/Fanuc/Mitsubishi and I suspect Siemens have one that is close.
Al.

[Leblondmakino]

I think I am agreeing with philserveng as I have managed to source a brand new identical motor at Fanuc spares UK.

I have tried disconnecting the +/_10v velocity command as suggested by Grant but then the drive doesn't power up at all and I get a not ready signal, maybe i'm not doing this right..!

I also tried removing the armature leads as suggested by Al and the encoder is working fine ie. it will turn a full rev in tandem with the handwheel without tripping out (took a bit of doing though!)

Time to purchase the new motor and see how I go..

Thanks guys for all your help!

John.

[gridley51]

I have had the same problem with an HU motor and I cured it by doing exactly what Philserveng is suggesting.
There are loads of these motors about in the UK so you should be able to pick a used one up.I have a load of them on machines and find them very reliable.I also have seven or eight as spares but want to keep them.If you do need to buy one,email me and I will give you the names of some places to ask.
Before condemning it remove the drive connections from the cnc and run it using a battery box.
I have also found with these motors that taking them off the machine and running them off a 24 or 36 volt supply is a good pointer to condition.If they run smoothly at these low volts they tend to be ok at all other voltages.I use an electric forklift as a dc supply.
What size is the motor and what machine is it on?
On edit:I just remembered that many years ago I bought a big cnc lathe with these motors on and it had a similar fault in the Z axis.I found it was the front bearing on the motor that had worn,Fitting a new one cured it.

Mark.

[GrantM]

Are you sure you are removing just the Velocity command and not some other control signal like Servo Enable? With just the Vel command disconnected, there should be no fault unless the control is trying to move the axis for some reason and then finding it isn't moving. The other possiblility is that there is drift (offset) which may trigger and alarm.

Be sure that the tacho brushes haven't bottomed out and that the connections to the tacho are good and the tacho is clean (the darm film on the comm is fine, just not heaps of carbon between the segments or badly worn comm). If you have 'ify' brush connection similar to what Phil suggested, that could cause your problem.

We also service DC servos (amongst other things) and very often have people bring them to us with the old "Acme Motor Repairs checked it and said it's fine" comment, yet when we put it on one of our servo system test-rigs, it's sick as a dog! Many standard motor shops just don't understand all the ins and outs of servo motors / servo systems and very few seem to have a proper servo system to set it up on, but rather just feed it with DC.

Anyway, if you can easily get a new one at a reasonable price, then that seems like a decent way to go.

[Leblondmakino]

Thanks Mark I would be interested in any contacts you have for finding one of these motors...do you have an idea on the price I should be paying for new and used..?
Penine Automation have 5 new ones in stock but I'm always keen to look around..

Servo Spec: 1HU 3074-OAC Stall Torque 7Nm

continuous rated current 9.6 amps

2000rpm 200volt

Machine: Beaver VC35 with Fanuc 6ML

Yes Grant my mistake I removed the servo enable wires by mistake!
I removed the velocity command wires on the drive card today and the motor still moves as before and alarms out.
It just doesn't want to respond to anty adjustments I make on the drive card..
It's really clean inside ...hardly any brush wear and the Tach and comm are very clean too.

At least if I get a new/used motor on I'll be back running and the faulty one will do for a spare if I manage to fix it!

John.

[gridley51]

John,
I still have my doubts that it is the motor.I`ve got a Beaver V5 that I`m working at as time permits,same drive and motors but Heidenhain control.Where are you based about?Can you email me your phone number?I`ll give you a ring Sunday.
Mark.

[GrantM]

If it's still running away with the Vel cmd off then it's not the feedback (enc/res), it's almost certainly either a dicky drive or dicky tacho (or connections). You indicated the drive had been tested, so make absolutely sure your tacho connections have continuity right the way to the drive. Check the tacho voltage at the drive when it runs off and see if it approximately matches revs. You said earlier it varies speed from ones start-up to the next, does it also do that with the Vel. Cmd disconnected (and preferably shorted)?