Power supply Gremlin in Fanuc O-M vmc

[carl laniak]

So,
this 1996-ish Leadwell AP-760 VMC with FANUC OM-F control has been a golden child since I acquired it 10 years ago...
Until recently.

About a year ago the control started not wanting to come up when the power button was pushed. I figured out the OFF, then ON button key strokes to get it to turn on....and have since found forum posts that indicate that issue is a dying cnc power supply. Cool. I've got one of those (used/good) arriving today. It got to the point that I might have to try for hours pushing those buttons to get it to turn on (but once on it would run flawlessly for days).

Some may recall that after changing my x-axis thrust bearings a few weeks back I turned the machine on and found the servo power supply module was fried (one of the internal fuses blew, and when I replaced it it blew again immediately upon turning on the control).... so I replaced that with an Ebay (used/tested) unit....which worked perfectly....for 2 weeks. (in hindsight it seems my thrust bearing replacement had nothing to do with that failure but I mention it because it did fail at the first startup after the bearing swap).

It's now got the same errors as when that PS module blew the first time.
409 serial error
414 x axis detect
424 y axis detect
434 z axis detect

So, first question is:
Is the failing CNC power supply possibly related to (causing?) the servo power supply module failure?
If so, then I'll replace them both at the same time and hopefully that issue will clear up.
If not, then I need to figure out next steps in trouble shooting why the servo PSM is getting fried.

Input from the learned people is appreciated.
Carl

 

[dandrummerman21]

Have you put in the new control power supply yet? The way it reads, you haven't yet replaced that (coming today?)

I'd put that in first before anything. Then replace the fuse that you had tried before.



414 424 434 alarm should generate alarm bits in diagnostic parameters 720-727 (I think 720 = x, 721 = y, 722 = z). What bits are on for those diagnostic parameters?

Are there alarms on the drives themselves? Or are they not powered again like last time?.

My boss has taken the power supply units apart and replaced the transistor packs (for lack of the correct term). I believe there's a way to check them with a multimeter. If you do have two dead power supplies, you might have enough good transistors between them to make a good one, at this point.

When we're troubleshooting this type of issue, we usually swap the drives/psu's between machines (most of our machines are mid 90s to very early 00's with MOSTLY the same type and size drives). But since I assume you don't have another mill or lathe with the same hardware, you're stuck buying used ones, or refurbs, or new ones, or having fanuc come out. Throwing money at the problem sucks!


To be honest, I'd still be wary that you moving the servo motor and cabling was the cause from the getgo. Could be an errant chip and some bad insulation in the cable. The "track" that the wires go are on that side of the table on that machine, in that square tube "way cover". I can't recall, have you meggered the cables?

 

[Garwood]

Bad capacitors.

Realize any power supply from 25 years ago is a decade past it's expiration date. If it works when you swap power supplies, crack open the old one and replace every electrolytic cap in there with quality new. Probably cost you $20 plus the equipment to do it (you own an old CNC, you should own a good soldering station and desoldering gun).

Then swap the re-capped one back in and you will be good for a decade or two.

If you do the same thing to the drives in the machines it will run like brand new and all the weird pixie gremlins will go away.

 

[dandrummerman21]

An update to my suggestion.

A note that I have says that if the power supply for the drives is OFF with no status LED:

"Control power has not been supplied. The control power circuit is defective"

You'd check for 24v and 5v, should be +/- 5%

So again, I think it is control power supply related. And as Garwood suggested above, I've always read that it's the caps that go bad in them, even if they don't look swollen.

 

[carl laniak]

Thanks to Dan and Garwood.
Dan, I'll look in to your suggestions in the next couple days, thank you.
You are correct that the CNC PS arrives today, so I haven't put that in yet.
And no, I did not check my motor cables resistance earlier like you suggested...can I get away with a normal ohm meter, or do I need a special one (megger?) for this purpose? I'm happy to buy the right tool if I need it.
And...I'll check those diagnostic bits tomorrow.

Garwood, your words are encouraging. I really do enjoy fixing the old machines, and I have replaced a capacitor here and there over the years....when trying to decide between selling off the machine parts and buying a new(er) one and just digging in and fixing things, it's nice to hear that this old beast can be made to keep running - with some effort on my part.

also, for anyone interested in Dan's comment about combining good 'transistor packs' between drives to repair things, member Ewsley has a video that I believe fits my situation:

Fanuc Power Supply Repair and Other Electrical Adventures - YouTube

I'll follow up in the coming days...
Carl

 

[carl laniak]

My klein multimeter has a range of up to 40Mohms (although I haven't found a spec for what voltage it uses)...it seems like from what I'm reading that will be enough to show whether I have a likely issue with my wires/windings....can anyone confirm that for me?

I see the Megger instruments used specifically for this purpose go way higher, but much higher than 40Mohms and it seems like all is good....so if I get reading of infinite resistance with my meter I'm likely just fine, but if it shows something less than 40Mohms then I am in the zone where that is likely my issue....right?

Thanks
carl

 

[ifixcnc]

The difference between Megger reading and a Multimeter, is the amount of current (I) used to make the reading.
If you read low on the Multimeter you have an problem.

 

[carl laniak]

............
To be honest, I'd still be wary that you moving the servo motor and cabling was the cause from the getgo. Could be an errant chip and some bad insulation in the cable. The "track" that the wires go are on that side of the table on that machine, in that square tube "way cover". I can't recall, have you meggered the cables?

So,
I got my brand new mega-ohm meter in, and sure enough....
I disconnected the cable from the drive and the motor, and I get 4000+Mohms on two of the cable leads....but the other reads 28.5M-ohms.
So apparently it's in that zone where the machine can run but if it moves just right then there's a short, right?

The cable goes through a pretty heavy duty jacket, through some "way covers".... Is replacing the cable (mostly asking Dan here since he has similar machines) as easy as unscrewing the connector at the motor end, pulling the cable out, and then pushing the new cable back through to the cabinet?
Or will I have to remove some things to be able to fish the new cable back through the cable guarding?

Sounds like at the very least I'm replacing this servo power cable, the cnc power supply that I've already got...and hopefully I'll be making chips again soon-ish.

thanks,
Carl

 

[ifixcnc]

There is not really a ZONE, it sounds like one cable has an break down of its insulation.
I use the old wires to pull in the new.
The power supply is likely OK, the servo amp may have been damaged by the imbalanced load.

 

[dandrummerman21]

So,
I got my brand new mega-ohm meter in, and sure enough....
I disconnected the cable from the drive and the motor, and I get 4000+Mohms on two of the cable leads....but the other reads 28.5M-ohms.
So apparently it's in that zone where the machine can run but if it moves just right then there's a short, right?

The cable goes through a pretty heavy duty jacket, through some "way covers".... Is replacing the cable (mostly asking Dan here since he has similar machines) as easy as unscrewing the connector at the motor end, pulling the cable out, and then pushing the new cable back through to the cabinet?
Or will I have to remove some things to be able to fish the new cable back through the cable guarding?

Sounds like at the very least I'm replacing this servo power cable, the cnc power supply that I've already got...and hopefully I'll be making chips again soon-ish.

thanks,
Carl

I've never changed a cable in a 760 or 1000 but it is fairly simple. Yes, you'll want to use the old wire to pull the new wire through, as suggested by ifixcnc.

 

[carl laniak]

Alrighty then,

I've been pulled in a few different directions lately, but I was accumulating the necessary parts and tools to get this thing put back together (hopefully) correctly.

There were a few extra feet on the length of the existing servo power cable, so first I tried shortening that, hoping that the issue was close to the motor (more movement) - but alas, after shortening it as much as I could there was still one conductor that was showing only 28Mohms...so I ordered a new length of cable.
I ended up frustrating myself for a while trying to solder 10ga conductors to the rather large terminals in the motor connector....but I'm happy to report that I bought a nicer (higher wattage, more adjustable) soldering iron, so hopefully I'm set on that front going forward.
Then there was the matter of the waterproof flexible conduit that went through the machine guarding...trying to remove the old piece, it was clear that the rubber (pvc?) sleeve was WAY past it's service life...it was cracking everywhere when I pulled on it....so I acquainted myself with the conduits available, and matched it as best I could from McMaster. I got a couple new fittings for that, but ended up re-using the old ones because I liked them better than what I bought.
Of course not having done it before, I had to snake cable through the conduit, and snake conduit through the guarding more than once ("dang it, I needed that nut on BEFORE I put the cable through in to the cabinet!").
But I got there....

I got it together, but the ON button still was resistant to starting the control, so I switched out the CNC power supply per dan's suggestion....and voila, the ON button worked on the first push!

and.....
for 20 minutes anyway, everything appears to be functional.

Will report back if anything else crops up.
Thanks so much,
Carl