[rbleth51]

Hi all. This is my first post, so here it goes.

Trying to get an older (1983-84) Hitachi Seiki CNC lathe with Fanuc 10T control up and running. It has sat for a while and been moved on a flat bad over the road to get to my shop.

All was going well chasing down the usual gremlins, loose cards, broken/losse cable, etc. Things are moving as they should and looks like it will run nicely, except for the spindle. It will not respond to a run command, M03/M04. The spindle drive is a Yaskawa Varispeed 626 MTII. A quick look at the drive after it failed to start the spindle revealed the "T DET" error LED was lit, meaning too high a tourqe on the motor was sensed before start up.

So I began a real basic inspection of the drive, cable conections, loose components, broken wires, etc. When I opened to swinging door on the drive to reveal its inner workings, I discovered an obvious problem. A component inside has a lead broken off from one end of it. Bellow is a picture of the component.




It looks to be a capacitor of some sort, but the markings are pretty vague, at least to my non electronics trained eye. Obviously the seperated lead pictured on the left was attached to the unit at the center of the side it is pictured on.

I don't know if this is the cause of the problem that is keeping the spindle from running, but it should be fixed before I go too much farther I think. I called Yaskawa and they were not a ton of help. The drive is so old that they don't really support it anymore. But the engineer I spoke to thought it was probably a filter capacitor of some sort. But that was all he dared say.

I have nor wiring docs for this unit and the ones I found on the Yaskawa site do not seem to contain this compnent. I tried to solder the lead back on but to no avail, the remaing part was not long enough even after scraping away most of the plastic coating around it.

If anyone could help me ID this part and figure out its specs so I can find a suitable modern replacement I would be forever gratefull. I have scoured the net and can find nothing based on the numbers on the unit. If any more info may be required to help someone who may know figure it out, let me know and I will get it for you ASAP. Below I will insert a pics of where the part was located in the drive. Thanks so much in advance for any help/info you can provide me with. Sorry for the length of this post.





Best Wishes,
Rob

[Al_The_Man]

That appears to be part or a RC snubber for High frequency suppresion, I doubt very much if it is affecting the immediate operation of the drive, The value appears to be .02µfd, I would say this jives with the physical size.
Did you try disconnecting the motor and see if you still get the same error?
There should be no torque output from the motor before the command is issued?
That probabally is a DC motor with wound field and tach feedback for that vintage.
Are you sure the LED does not come on after the spindle attempts to turn?
Al.

[rbleth51]

That appears to be part or a RC snubber for High frequency suppresion, I doubt very much if it is affecting the immediate operation of the drive, The value appears to be .02µfd, I would say this jives with the physical size.

Yes, I was thinking for the size 0.0022µF. I read an artical somewhere saying that if no unit was given for a Farad rating, assume pico Farads. And the suppresion suggestion coincides with what the Yaskawa engineer I spoke with was thinking.

Did you try disconnecting the motor and see if you still get the same error?

I did not, I wasn't sure if that would cause a problem or not, don't want to let out any "magic smoke" from the drive by accident. Is it safe to try that?

There should be no torque output from the motor before the command is issued?

I agree, that is part of what has me puzzled. The spindle will move a little when the gear change comand is given (M40/M41), but does not even twitch when an S word for RPM and a run command (M03/M04) is given.

That probabally is a DC motor with wound field and tach feedback for that vintage.

I had thought it was an AC spindle drive, but I am not sure. My knowledge is very little in this area. But I try to learn all I can every day.
Are you sure the LED does not come on after the spindle attempts to turn?
Al.

I beleive that that is exactly when the LED comes on. I have not specifically checked the state of the LED's before the run command to see if it was on before or not. But I don not recall it being on any of the times I checked the LED's before I tried to start it. I looked them over before I attempted running and checked the meaning of all of the LED's that were lit against what the manual said they meant to be a little more sure it was safe to try and start the spindle/drive.

Would it be safe, in your opinion, to repower the drive, without the cap in place, for the purpose of checking that out? I would assume yes as the cap was not connected on the broken off end anyway. If you think yes, I will power it up in a few and check. But like I said, I am pretty sure it does not come on until after it gets the run command.

Below is a photo of the drive nameplate data. Maybe that will help to determine whether or not the motor is AC or DC. I feel kind of ashamed that I don't know for sure.



It says inverter drive on it so I had assumed AC for the motor. But like I said, not sure.

Even if the bad snubber cap is not my problem, and help on figuring out a replacement cap for it? I would like to replace before I run the spindle too much, if I can get the spindle to run at all. What about polarity? None is labled on the originall. All the caps I have looked at at Mouser are of differing voltages. Heck, I am not even sure if the lable of "600V" on the original in the photo really is 600 volts. And if so, is it VAC or VDC? My electronics ignorance is really showing through now I am sure. And would mees things up if I replaced the cap with one from Mouser rated for 1000V? The one I found at Mouser that, in my un-educated judgement, best matches the orignal was rated for 1000V and 0.0022µF.

Thank you so much for the help so far! Hopefully I can get this thing going with the help here.

Best Wishes,
Rob

[Mike Nash]

The capacitor is a 2.2uF 600V non-polarized film capacitor. It's a snubber. Newark has this discontinued (non-Rohs compliant - no big deal in the US and solders more reliably) part that should work for $5.94. http://www.newark.com/jsp/Passives/F...sp?sku=46F1232

Be warned that not all film capacitors make good snubbers. It has to do with the way the end of the "tape" is connected to the leads.

Since you are not getting any actual trip condition, it's probably not your immediate problem. You could have other problems due to it however. I can't advise running for long without it. The TDET light is just telling you the torque level has reached the point set by the TLIM pot. IF this light is on but the motor is not turning you may have a mechanical stall condition, i.e. the motor or load is requiring more torque than the drive is set to deliver. Turning TLIM up might start it turning or it might burn something up or break something. I'm wondering if there is a spindle brake that is not disengaging or some other reason such as rust or gunk that the spindle cannot turn freely.

Your drive is an inverter, hence AC motor. The manual I'm looking at is here http://www.yaskawa.com/site/Support....ucts%20Defined

Good luck!

[rbleth51]

Mr Nash,

Thank you so much!! That is exactly the kind of info. I was looking for. You even found and provided a link to a suitable replacement for me. I had spent hours and hours trying to find one. Chasing my tail mostly as I was not certian what values the orginal had that I was replacing. SO you think it is 2.2uF? The fact that it was labled 2.2K threw me, I was thinking K might be referencing or = 1,000. I will place an order for a few of the new ones you linkd to for me when I am done this post. For that kind of money, even if it is the wrong thing, it is worth trying.


I was pretty certain it was an AC drive and motor. But I have spent so much time chasing my tail on this one, running into dead ends, etc. that I had an overwhelming feeling of not being certain of anything any more. LOL


The spindle nose seems to turn fairly freely by hand. Doesn't seem to be bound by any pins, gears, shifting mechanisms, etc. It may have a brake, but I don't think so. I have seen nor reference to one in any of the docs I have or on the machine. If it does have one, and it is "on" it is pretty weak as I can turn the spindle by hand without too much trouble. And it does move without any visible trouble when the gear change is comanded, and seems to change the gear succesfully.

I did find those docs/manuals on the Yaskawa site, but was unable to determine any answers from them on my own. I will be reading them through pretty closely over the weekend though.

I will be powering up and checking the LED states for certain after I order the new/old caps. I will post what I find.

Any other information, opinions, help are still much appreciated and wanted as the spindle is not turning yet. Thanks!

Best Wishes,
Rob

P.S. Sorry about my poor spelling, my education is in Mechanical Engineering, so numbers more than words are my thing. And I have not figured out if a spell checker is available here.

[Al_The_Man]

The fact that it was labled 2.2K threw me, I was thinking K might be referencing or = 1,000. I will place an order for a few of the new ones you linkd to for me when I am done this post. For that kind of money, even if it is the wrong thing, it is worth trying.

.

You need to be sure, if the value is in fact much lower than 2.2 µfd, a 2.2µfd can fry the resistor if it is across any kind of AC supply.
I would have considered the µfd value high for a HF snubber.
If everything is free but as soon as the motor moves you get a Torque alarm, it could possibly be a connection on the Resolver feedback.
You could remove the motor leads (3ph) and try to run, you may get another alarm, but if you still get the TL alarm, I would suspect the switching components.
If it sat for a while.I assume It wasn't mothballed because of a shorted spindle problem?
Al.

[Mike Nash]

You need to be sure, if the value is in fact much lower than 2.2 µfd, a 2.2µfd can fry the resistor if it is across any kind of AC supply.
I would have considered the µfd value high for a HF snubber.
If everything is free but as soon as the motor moves you get a Torque alarm, it could possibly be a connection on the Resolver feedback.
You could remove the motor leads (3ph) and try to run, you may get another alarm, but if you still get the TL alarm, I would suspect the switching components.
If it sat for a while.I assume It wasn't mothballed because of a shorted spindle problem?
Al.

It's definitely a 2.2uF. I looked around to see if we had one in the bone pile but didn't see any. We don't see many (if any) of the spindle drives in for repair, but get a lot of the larger Yaskawa VFDs. We've been selling/repairing them for years. Unfortunately, we have a limited series of prints and the spindle drives are not one of them. I reckon I've personally only seen one and that was a few decades ago.

I'm going to correct my earlier statement after rereading the (poorly) written manual. The TDET pot sets the torque level at which the TDET LED turns on. It is actually just there to make available a contact output to signal some user adjustable torque level has been reached for some user determined reason. I.E. it doesn't mean a whole lot unless it is being used for some external circuit that may be shutting you down if the torque exceeds some level. Try turning up both TLIM and TDET. But also record where thay are now before you do.

Are you getting any other LEDs?

[rbleth51]

If everything is free but as soon as the motor moves you get a Torque alarm, it could possibly be a connection on the Resolver feedback.
You could remove the motor leads (3ph) and try to run, you may get another alarm, but if you still get the TL alarm, I would suspect the switching components.

That is something I had not dared to try yet. I wasn't sure if something would fry if I tried running the inverter with no load connected.

If it sat for a while.I assume It wasn't mothballed because of a shorted spindle problem?
Al.

No, it was making chips the morning we picked it up from the shop that sold it. They had taken delivery that day of a brand new Daewoo Puma I think it was. Much bigger machine and brand new. They needed the increased capacity of that machine.

Thanks for the advice/ideas.

Best Wishes,
Rob

[Al_The_Man]

It should not make a difference with the motor 3ph disconnected.
Did you try what Mike suggested?
Al.

[rbleth51]

Mr. Nash,

Sorry it took so long for me to reply to the last message you posted. I wound up with strep throat and a couple other nasty things and have been down for a while.

Unfortunately, we have a limited series of prints and the spindle drives are not one of them. I reckon I've personally only seen one and that was a few decades ago.

Unfortunately prints for this drive seem to be near impossible to locate, at least for me anyway.

I'm going to correct my earlier statement after rereading the (poorly) written manual. The TDET pot sets the torque level at which the TDET LED turns on. It is actually just there to make available a contact output to signal some user adjustable torque level has been reached for some user determined reason. I.E. it doesn't mean a whole lot unless it is being used for some external circuit that may be shutting you down if the torque exceeds some level. Try turning up both TLIM and TDET. But also record where thay are now before you do.

I did not get a chance to turn up the TLIM pot yet. But I did change the TDET pot a little. 2 changes resulted. First the spindle actually rotated a little, maybe 1 rev, and then stopped. Second, the TDET LED did not light at any time during my atemps to run the spindle. Hopefully I will feel well enough Tuesday to try advancing the TLIM pot a little.

Are you getting any other LEDs?

Sort of. I did get my wife to come down to the shop today so I could watch the drive while she ran the control to try to start the spindle. But it is late and I can't find the paper I noted it on.

I ordered a couple of the capacitors that you pointed me to with the link at Newark. They are supposed to arrive Tuesday. I don't want to play with the drive to much until I get that put back in place. Thanks so much again for the info on that. It would have taken me forever to ID that part, if I ever figured it out at all.

I did notice another anomally on another electronic component in the drive. But I will post that in a seperate message.

Thanks for all your time.

Best Wishes,
Rob

[rbleth51]

While I was adjusting the TDET pot I noticed something. The part pictured here:



It caught my eye as it appeared that something may have "oozed" out of it at some point. I am reffering to the white material that is on either edge of the part. Unfortunately I do no recall if the white goo had been there before or not. It was not warm at all to the touch. And it was not sticky or in any way runny/liquidy (please excuse the poor teminology, it is 1:15 A.M. here). That is to say it was fairly solid. Not hard, but like a spray foam that had long since set.

Like I said, I had no idea if it was like that before I started trying to get the spindle running or not. The only reason it caught my eye was that it resembled the look of some capacitors in my PC's power supply that had failed and similar material has flowed out from them.

There are many similar parts to this one on the same card, and none show any evidence of the white stuff on them. And not one of them have the same numbers as this one printed on them nor do they appear to be the same size. Some seem to be a little larger and most appear to be a little smaller.

As with the capacitor from before, I have no idea what this part is. So I have questions.

1. Anyone have idea what this part is? (in case the white stuff is bad and it needs to be replaced, if it can be replaced)

2. Knowing what the part is, could the white stuff have come from it?

3. If the white stuff could have come from it, and did come from it, is it bad? ( I think the obvious answer to this one would be yes, but I really have no idea) In other words, is it an indication that the part has failed?

Here are a couple more pics to help in the ID/diagnosis.






Thanks as always for your time and help. It is greatly apprciated.

Best Wishes,
Rob

[Al_The_Man]

The 'White stuff' appears to be silicone rubber, it is used to cushion components like those that are prone to break off due to vibration, it is applied at board assembly time.

Was anything disconnected during the move? such as spindle motor etc?
Al.

[rbleth51]

The 'White stuff' appears to be silicone rubber, it is used to cushion components like those that are prone to break off due to vibration, it is applied at board assembly time.

Phew! That is what it looked like and felt like to me. But all I could picture was the many failed caps I have seen and seen pictures of that had oozed stuff out. It has exactly the consistancy and feel of silicone rubber.

But I am still very currious now as to exactly what kind of electronic component that is. The artwork on the PCB almost appears to lable it "Ic". Integrated circuit? Any ideas? Anyone?

Was anything disconnected during the move? such as spindle motor etc?
Al.

Nope, nothing that I know was disconected. But that doesn't insure that nothing was.

As far as the resolver goes for feedback to the drive, where would it be? In most cases I think I could pick it out pretty easily, but not here. The motor itself is burried deep in the castings of the lathe, and inside a couple more, fairly large (and I am sure really heavy) sub castings. So seeing it is quite a trick. But when the flexible conduit going to the motor is examines it yeilds a few things. 4 fairly large wires, I would assume the 3 legs of the 3 phase power to the motor and a ground. 3 more smaller wires (maybe 4 but I don't recall now) that I am assuming are the 3 phase supply lines to the motor cooling fan. And lastly a cable with several, much smaller conductors in it. I thought at first that that cable must be the resolver line.

But then I noticed another unit. It has a small pulley on its face and is driver by a belt directly from the spindle tube, right beside the belt that drives the spindle from the spindle motor. It has a faily large conduit running to it with a big Mil. connector hooking it to the conduit. I have not snaked around under the lathe enough yet to find where that conduit goes. I will take some photos of the unit latter and post them if that would help.

I am not very familliar with resolvers. Mostly I have just removed them from servos to replace with encoders. Not idea how many wires should be going to them, etc.

Any thoughts?

Thanks for your time.

Best Wishes,
Rob

[Al_The_Man]

Normally 6 small gauge wires to the resolver, it would be right on the rear motor shaft usually.
You may have an encoder also mounted on the final spindle shaft for threading and CSF.
The component looks like a potted Opto Isolator P&B make some that look just like it.
Al,

[Mike Nash]

The part with the silicone on it is called a hybrid. It's a ceramic substrate with some components (resistors typically) deposited on it and others soldered on (ICs etc.) then coated (or not). The silicone is definitely factory. Look for loose wires since it was running before you moved it.

[rbleth51]

Well, I feel pretty foolish now. Why will become quite apparent I think in just a few moments. I am no electrician, nor very experienced in electronics. But I know enough to get by. And I can't beleive this slipped by me.

So first, the replacement capacitor arrived about an hour ago. I carefully installed it in place of the broken unit and powered things up. All seemed the same during power up. I was carefully to observe the drive LED stated and wrote them down so I would not forget them during the long 50' walk from the shop to the house. Here they are:

Main machine power on/control power off - not started yet
Z SP - lit
NDET - lit
all others off (assume that unless I list it as lit, it is out/off)

Main machine power on/control power on - control started
SOURCE - lit
Z SP - lit
NDET - lit

Same power state as previous but with spindle commanded to run
SOURCE - lit
RUN - lit breifly, then out
AGREE - possible flash on for an instant, but may have been light from RUN LED reflecting as they are beside each other
Z SP - lit
NDET - lit
TDET - lit

The above attempt at comanding the spindle to run was done with the TDET pot reset to it's original position. I was planning or turning it up some, but decided to check a couple of things first.

I have been reading the manuals for the drive from the Yaskawa website. Last night, in one of them, I do not recall which, I ran across a table listing places to check voltage and the allowable values for them. The first listed was for the mains into the drive.

I decided that this was another one of those very obvious so get it out of the way first things to check. So I got out the old Craftsmen multimeter and set to work checking the voltages on the drives 3 phase mains input.

Phase to Phase readings all seemed resonable, between 243VAC-248VAC. I am really close to the substation here, so mains voltage is a little higher than in most places.

But then I decided to check Phase to Ground just to be thorough. I got the expected 120VAC-129VAC on two of them. The third is genrated by a pahse converter. It is an ARCO Roto Phase CNC 15 (rated for CNC duty, condtioned power and all that jazz). I was shocked to see the voltage from the generated Phase to Ground was 220+-VAC!

When the converter was installed the guy that did only checked Phase to Phase. Aparently he never checked Phase to Ground. Or if he did something has drastically changed since it was put in.

Not sure if this is my whole problem or not. But I am betting it is at least part of it. And at the very least, a problem I want/need to fix before going any further.

The worst part is that I have no idea what might have cooked in the drive as a result of the way high voltage on that generated pahse. For all I know it might be a non issue, and I am sure you will correct me/set me staright. But I am betting it is an issue and probably a big one.

But now I have no idea how to go about getting that voltage down on that phase. Any ideas/suggestions?

For all I know, the hight voltage on the phase generated by the converter is a sign that something in the converter circuitry has failed. On the caps in it blew or something.

Any advice/suggestions/etc. is, as always, greatly appreciated. This one has been pretty frustrating. But I kind of like tracking down these kinds of problems in older machines. It gives me an opportunity to learn a great deal. Or at least learn just how much I don't know.

Thanks again for the time and thought. And in case I do not get a chance to speak to anyone before tomorrow, have a happy Fourth of July!!


Best Wishes,
Rob

[Al_The_Man]

I suspect it is not causing any problem as the 3ph goes into a fullwave SCR bridge, which does not appear to be referenced to ground, I think that excessive to-ground voltage is due to tuning caps etc.
Also I would expect something drastic to have happened when you powered up, if that was an issue.
It may be more likely that you have the controller etc and single phase loads on the artificial leg, energising the large spindle causes a 'bump' on the artificial leg, which although not always affecting the spindle, will knock out the control if on this 'artificial phase'.
Al.

[Mike Nash]

I also don't think the 220V to ground is an issue. You have 120V to ground on the two legs because it is a center tapped transformer producing them and the tap is grounded. You can see this in some plants with 240V 3 phase power.

I think the TDET relay output may be signaling something else to shut you down, hence the loss of the RUN LED. Try watching the lights again with the TDET turned up a little. You could probably turn it all the way up since you have no load. It's not doing anything internal to the drive. Just turn it back down later after you get the spindle running. I still think this is for external shutdown when too much torque to the load is detected, not a drive fault.

BTW, that is one amazingly clean drive!

[Al_The_Man]

Something that crossed my mind is the logic supply for the drive, although it is shown as a 3 ph transformer, maybe some models may have had a internal 1 ph transformer for the control, may be Mike can verify.
If it is a 1ph then it is important to make sure the two particular phases that feed it are on the main 240 1ph that feeds your convertor, (not the artificial ph).
If it is a three phase transformer for the control board, then it is possible it also uses phase loss detection, and if you get a momentary drop on one phase, which is common on high loads on RPC's etc, then it could be dropping out for this reason.
Al.

[rbleth51]

Good afternoon. You were both correct about the high voltage not being the problem.

The spindle is now running. And you were right about the cause as well. It was on the control end. But not due to any signaling coming back from the drive. I feel kind of stupid again about what actually turned out to be the problem.

On the 4th I had some time after the big familly BBQ to relax before more familly activities got underway. So I sat down with yet another beer for the afternoon and the operations manual for the lathe. I figured I would thumb through it yet again and see if anything jumped out at me as an explanation/solution to my problem. And low-and-behold it did. I randomly landed on a page that mentioned what I needed.

It talked about canceling a "spindle rotation interlocking mechanism". To do this one needs to, with the machine/control running, hold down the center retract button on the control panel for 5 seconds or more. As soon as I returned home from the familly get together I tried this procedure and the slindle ran up to speed as soon as I commanded it too.

You have to love old, poorly translated manuals for just this sort of thing. The note was burried away in a location that I never would have intuitively looked for it.

Sorry if I have wasted alot of your time trying to help me through this part of the problem. But I am very greatfull for the oportunity to learn from both of you. Like I said before, I try to learn as much as can at every oportunity that I can.

I also want to say thanks again for the help in identifying the capacitor that had the broken lead and with finding a source for a replacement. That is one that I truly did need help with and really wanted to have replaced before I ran the drive anyway.

Now, if I can only figure out why none of the panel LED's on the control will light. They seem to have no voltage going to them at all, and no change in voltage on the lines feeding them when the LED state should change. But that is one for a new thread.

Thanks again for all the time and help. I really apreciate it and am thrilled that the spindle and drive are now running and cooperating with each other happily.

Best Wishes,
Rob