I got Fanuc Power Supply / Amp Issues...

[cncmek]

Hey Ox,
After reading through this a couple of times the only reference to a temp alarm is you mentioning you had on a couple of years ago. You have swapped amps and it seems to not follow the amp. You keep getting a current fault.

Do you happen to have a amp meter with min/max function? If not, pull up a chair and watch. You could monitor the the motor leads at the drive and see if you are actually getting an over current or mabye loseing a leg/single phasing and thats whats's causing the curent fault.

Could be any number of things a winding opening up, a loose connection or a bad contactor.

You could meg the wires and motor, may some bad wires/isulation.

Just throwing these out there as food for thought.

Dave

 

[Ox]

After reading through this a couple of times the only reference to a temp alarm is you mentioning you had on a couple of years ago.

Agreed

You have swapped amps and it seems to not follow the amp.

Agreed

You keep getting a current fault.

The only "curent" fault that I git are on power up after powered down overnight - before eny servo engagement. The colder it is - the longer that I need to leave it "up". Then cycle power at case level. Alarm gone. This powered up time before cycling power has gotten longer and longer over the last 5 months, but it is still only at maybe 2 minutes. One axis will clear before the other'n if I don't wait long ernough. I will not have an issue with that enymore the rest of the day. We are assuming that it must be a failing capaciter or similar. If it was enything else - I don't see how cycling the power would change the alarm or not be an issue at varying times during the day.

The only reason that I started with that issue is that it IS a funky quirk that the amps are dooing currently, but I also suspect these are two totally diff issues. Unless possibly a weak cap could somehow be blamed for the less than 100% duty cycle?

Also - I am assuming it would be the cap in the power supply as that quirk has not migrated yet. ... Maybe I should play musical power supplies too? The one having issues is a 130 and the other one in there is a 115. I wonder if I could at least swap them for a bit to see if it migrates? Hmmm??? Guessing the 115 wouldn't power the 25hp main spindle motor?



----------

Think Snow Eh!
Ox

 

[cncmek]

Ok, so two different issues.

As fas as the power supply the 115 probably wont power the spindle, but you are getting this alarm just turning on the machine in the morning not running it right?

A swap may tell you a few things. If you get a fault on the other rack the it could be a faulty power supply.

If the fault stays on the same rack the something is dragging it down.

I didn't look up the numbers but assuming Alpha series drives you diconnect the buss bars and reconnect on at a time ( after letting the buss voltage dissapate ) to see if one the drives is dragging it down. This also gives you the chance to check the buss conections. I've seen a loose screw on a buss cause current faults before.

As far as the turret alarm you may well be hitting the duty cycle. It would amaze you how limited they can be. We just went through this on some Duplomatic turrets. Been having a rash of turret failures for some time on a particular model. Had to go back to Duplomatic to get the duty cycle and come to find out that the damn things almost had sit idle as much as they had to run. Customers were wearing out the turrets.

If you are able can you shuffle the the live toy work from the one turret to the the other to lighten the load on the one giving you problems. If it a duty cycle issue it should show up on the other turret after a while.

I hope this made sense, it's early and I am still on my first cup of coffee.

Dave

 

[Ox]

As fas as the power supply the 115 probably wont power the spindle, but you are getting this alarm just turning on the machine in the morning not running it right?

A swap may tell you a few things. If you get a fault on the other rack the it could be a faulty power supply.

If the fault stays on the same rack the something is dragging it down.

Yeah - I was thinking it should be at least good for setting overnight and power-up in the morning. That should seal the deal I would like to think...

If you are able can you shuffle the the live toy work from the one turret to the the other to lighten the load on the one giving you problems. If it a duty cycle issue it should show up on the other turret after a while.

This last job was all C werk and I could put the holders enywhere, but the next one is mostly Y werk - and turret 1 will be loaded with lives. (10) Turret 2 will have only 2 or 3 lives, so I may have the same issue aggin - only on the other turret. We'll see...

I hope this made sense, it's early and I am still on my first cup of coffee.

Too bloody early for me! LOL!


-------

Think Snow Eh!
Ox

 

[cnctoolcat]

Hey Ox,

Try 'n get the caps replaced in the power supply. This should eliminate the caps as an issue.

I recently had a spindle drive power supply that did the same, not-wantin'-to-power-up-at-power-up thing. Took a few minutes to get the caps warmed up, with just the main machine breaker on, before turning on the control. I replaced the 30 or so caps, and all was well.

Well, except for my dumbass installing the power supply "slightly" wrong...that's another story for a new thread when the saga is complete.

 

[Ox]

Well, except for my dumbass installing the power supply "slightly" wrong...that's another story for a new thread when the saga is complete.

Your ***Saga*** or mine?



---------


All-righty! I have more to add to this here saga.

Now y'all aint gunna believe this here...

CNC Toolcat - here hold my Dew while I tell this eh? (I may need both hands. )


Now what we have here aint exactly a failure to communcate (Ruled that out with swapping encoders) but it is more a really strange issue. Kind of one of them ... uh .... them really strange things that make no sence no matter how you look at it? .... Oh yeah - I got it - it's named after that really whacked out chick - you know - Anna Molley? (Not to be cornfused with her sister ***Hatchet Molley***. )


Well enyhow - here's the current state of affairs here in the corn fields....

The next job has been running for a month. There is a LOT of milling on it, but most all of it is done on the other turret which uses a completely diff power supply and motor than I was dealing with before. (Keep note that I had already swapped the encoder as well as the drive unit. They are still swapped.)

I have been running 24/7 since with hardly eny down-time.

It ran right through a week ago last Thursday when it was 96* and the humidity was sympethetic to ***New Orleans***.

However I have had the lower turret get "lost" a few times here lately when the temps have been lower. (Had to cover up the 'maters this week!) Temps are cool, door is open, and yet I seem to have this issue that seemed to be heat related before. ??? Worster - I had it down for unrelated reasons Friday night and needed some supplies from the hdwr on Sat morning.

By 1:00 pm Sat I had it farred all back up and making chips. (Sat all night and had been cool eh?) Took off for the afternoon for dinner at in-laws. (Dinner was late as my FIL took his Goat and his chumm took his Cobra Repli through the parade in the morning. Needed the heater in the Cobra. LOL!)

Came home 4-1/2 - 5 hrs later to find that the lower turret had been spinning "lost" since 10 minutes into the cycle. (No time to build heat) So I reset it and walked away. Came back a few minutes later to find it "lost" once aggin.

So now I decide to put my thinkin' tuk on and try to finger out what the bloody werld has changed in the last few days?

In previous weeks I had noted that the machine seemed to respond well to having the operator fan on. And with the cooler weather I had it shut off. (No need to catch P-Newmonia just to chack parts eh? I've heard that you can "catch'yur death" that way )

So I reset the machine onced aggin and turned on the bloody fan. That was 24 hrs ago now. Machine has not had an epeleptic eppisode since.

Fan is pointed to where the operator would be, provided it was a manned job. The control cabinet is behind the machine. (Not even mounted to the machine) Now the face of the control would "see" some of this wind. ???

The real kicker is that the last cpl times especially - I can be remember specifically - and that the machine did NOT have time to warm up enough to have had a heat issue. ALSO - on the same line - the fan had NOT been running long enough to cool enything off that may have been warmed up from previous eppisodes.

This is what I would call an Anna Molley!


So that's where this story sits currently. Now - can I have my Dew back?


This Dew feels lighter than it did a few minutes ago. HEY!


--------------------

Think Snow Eh!
Ox

 

[Tonytn36]

Ox, just a thought....

On our Fanuc'd controls, there are 2 (iirc about 40 or 60 mm) fans over the display on the control. These cool the control itself. We've had these go bad before and the control did all kinds of freaky things. This is on 15/18/21 series.

 

[Ox]

Have narrowed one issue down:

Looked back in this thread a bit to see that the main issue noted here had been the "lost" turret. I don't know that I have had that issue since then. ??? I mean - it's definately been quite a while since I have seen that happen now!

What has changed since then? Well - the only thing that comes to mind is that (Now you REALLY aint gunna buy-in to this!) I am running my library on empty. I have had trouble on 2 diff Fanucs with files getting messed up and this particular one actually tied up so tight that I had'ta send it out to git "unlocked"! It was a feasco to say the least! (3 yrs ago)

Even since that excorsism I have had files git messed up. I have learned that if I keep my library emptied out that I don't have this issue. Could it be that [for whatever reason] this also played a roll in my 'lost' turret? It is too crazy to believe, but....


So - enyways - on to new business.....

I had thought that all my issues were relative to a common power supply. Nothing on the other supply was messing up. I assumed it was a weak capacitor as I was having troubles getting it to light up after the case had been powered down long enough to cool down. (caps to drain)

Well the issue has gotten to the point that I can't get the machine up at all now. We kinda got it narrowed down to a 24V issue. So now we are playing musical daughter boards, as well as switching to a diff 24V supply - but the meter read a good value, so ...

Have it nailed down to the daughter board on the one 2 axis Alpha amp that is creating the issue, with the same issue on a nother amp right beside it getting worse, but still able to clear the alarms yet if allowed to warm up a while. There is yet another amp almost the same with no issues at all.

My teckie found a small cap on the 24V circuit that looked like it could be an issue, and was easilly servicable. He gave it a 15-25% chance of fixing it he said.

It didn't...

There are three other little doo-hickies on there that one of them is likely the cuplrit - but they are not easilly serviceable. This board is very densely populated and he is not comfortable just throwing parts at it.


So - I called TIE and they did not have any reman boards to exchange. They priced me new, but didn't have that either...

I called Fanuc. This board has been superceeded to another number. But wait - it's not that simple! The new daughter board won't neccessarily work with the current mother board! It is highly suggested to buy a whole new(sed) amp with the updated set-up in it.

Now I can't fuss aboot the price of the new(sed) unit from Fanuc, but I decided to call TIE to see aboot a complete unit as well. ???

Yep! They have the whole drive! BUT! There's always a but eh?

BUT! (via TIE) There is a service bulletin from Fanuc (and they sent me a copy of it) you may need to update your SRAM to accomidate the newer drive unit. The SRAM has a $500 deposite on it. (for a flash card? It would seem that they could jist e-mail it to me eh?) I send it back - I git my $500 back. No biggy there.

But FANUC made no mention of this a'tall!

The next issue tho is that I have THREE of these on my machine! If I need a software reflash to change one of these - I hafta wonder if I will hafta replace ALL THREE?

So this is where 5:00pm Friday leaves my sorry assets.

Hopefully someone that has these troubles will be able to learn from my pain.



-----------------------

***Calgon - Take me Away!***
Ox

 

[Bruce Griffing]

Ox-
Given what you face, I would go back and look at the 24V board. Replacing all of the components sounds good to me... of course I am not there looking at it. Can you post a photo of the 24 volt board that has difficult to service components??

 

[Ox]

Yeah - I guess so. I'll try to fetch the camera tomorrow.

There is a LOT more going on than 24 tho. All encoders and such cable in here as well.


-----------------

***Happy 75th for the US Male***
Ox

 

[Bruce Griffing]

If you are going to post photos of the board, also post a photo of the schematic page (if you have it) that shows what goes in and comes out of the board.

 

[Ox]

I only have Prog and 1 basic Maint manuals for this machine.

My guy didn't actually replace, but rather dbld up the little purple cap in the BRH corner of the 3rd pic. (from the back side)

His thought is that it is likely one of the three little alum w/black striped cans on the LH/middle of the board. One small and two smaller yet.

My alarms are 6, 8, or 9 and message on screen says X and/or Z Abnormal Current. (Depending on the alarm) A little "soak" time to allow to charge up and then a quick "power cycle" will usually get one of the axis to clear, but the one drive won't clear both anymore.

Once fired up - if power to the case is left "up" - it will run 'till the cows come home. But once powered down - all bets are off aggin. Of course I could always git it to clear untill now.

Board number A20B-2100-025 3
Current replacement is a -054 I think they said - but won't work with current mother board. So whole drive needs swapped out. But the new drive number is the same as the old. ??? (A06B-6096-H208)

Little cans spec:
33, 25V, 7ZD (Bigger one)
2.2, 50V, 7ZO (Smaller 2)

He says that those are fastened in a challanging way. Also - he is jist guessing at the cause.

-----------------

Think Snow Eh!
Ox

 

[Bruce Griffing]

Ox-
I don't see anything obvious on the board. Surface mount parts can be hard to replace, so I understand the reluctance to do so. On the question of voltages - a voltmeter is a great thing, but does not compare to the information an oscilloscope gives. If your electronics guy has not already done so, I would look at all of the power supply voltages on a scope. Of particular interest would be the high voltage bus. I say this because the fault could be low current as well as high current - caused by a weak supply. This test would require either a differential input for the scope or a scope that will do channel A minus channel B. I would check all other voltages as well. I think this is very worthwhile, given the alternative of replacing most or all of the control. If you lived nearby, I would come over and do it. OTOH it is unrealistic to "think snow" in Texas so you probably won't be moving here soon.

 

[adama]

Im guessing here but i personaly would replace every capacitor on that board. Its going to cost peanuts compared to the other option.

Gotta say though i have major douts about 2 - 3 parts dieing at once. Have you given all the controls boards - other drives - anything in that machine a good blast of fresh air. Only takes the smallest most random spec of even slightly conductive dust to create some realy wonderfull faults.

 

[Ox]

Not 100% following your path Bruce, but the 24 is supplied elsewhere. Not generated on this board.

High/Low V differential? What doo you call high? AFAIK the 24 is likely the HV on here. Now I see the 50v rating on those two little shiney [not exactly]caps. So maybe there is?

As per "buss" are you referring to the 24v or the buss to run the motors? - I don't even need to hook up the buss lines or even the motors at all to test. The amps are not "engaged" yet when I am getting this fault.

I originally thought that it was that big cap in the power supply feeding the buss, but that is ruled out as we git this fault before we ever engage the amps. This comes up as soon as we put power to the case level. (24v on)


Adama:

Not saying that there is no truth to your statement - but if it was a pc of shit shorting something out - why would it go away once it had time to soak? And also behappening exactly the same on two diff drives at the same time. One still able to clear, the other finally not able to clear. And taking a yr to progress to this point.... (3rd one is fine and dandy to date.)


Bruce - if you think that this is something that you want to tackle (quickly) I will NDA you a good board as well as a bad one.

I don't know about getting schematics tho. I searched this part number and all Google came up with was two hits, both were in Japan I think. ??? Does Fanuc usually enclose such info with purchase? I would think they are all about keeping it proprietary eh?

You've had as cold or colder temps than we have so far this week I think. (supposed to change tonight I guess?)


------------

Think Snow Eh!
Ox

 

[Bruce Griffing]

Ox-
My reference to schematics was to what comes with the machine. I did not expect to have a schematic of the board, only a schematic that shows the board and its connections to other stuff. In my Kitamura manual (Kitamura VMC), I have a schematic section before the ladder diagrams. It shows all of the boards, power supplies, servo amps, etc. The utility of such a diagram is that understanding the signals coming and going from the board can help in the diagnosis of a problem.

My comment about the buss (or bus) is the high voltage bus that is used by the servo amps. I catch the drift that you have come to the conclusion that the HV supply is probably not your problem, but in the spirit of checking things that could contribute to "abnormal current", I would check all of the power supplies with a scope anyway. This could uncover a problem that a voltmeter will not see. It may not, but it is worth a try.

The next thing I would try is to connect one channel of a digital scope up to the fault line and trigger on that. I would have the other channel(s) connected to things that might cause the fault - like power supplies, motor leads, etc. Once a fault occured, you could look back in the scope memory at the preceeding signals to see what caused the fault. That would probably take several tries to zero in on the problem, but you could probably establish the root cause.

On the boards, I think you are better off trying to debug the whole system rather than a board by itself. It is conceivable that the problem is not even in that board - though your evidence is against that. If I were you, I would figure the cost of the replacement deal and compare that to the cost having a crack tech come out and try to debug what you have. I know it is a roll of the dice - but that is the best suggestion I have.

 

[Ox]

It's on this board. No doubt in my mind.

I have swapped this, that, and the other thing all around. The error chart consistantly follows this board.

It makes sence too as Fanuc has superceeded this board - and likely for a good reason. ???

Now that I think about it - I think I doo have a wiring diagram!


-----------

Think Snow Eh!
Ox

 

[Bruce Griffing]

Well..... we are waiting.... where is the schematic??

 

[Ox]

Machine is UP !!!

I had one wildcard up my sleeve to be played yet as it seems...

Along the lines of "warm-up" or "soak" time on the cards before clearing the faults, I tried a heat lamp. Didn't see any results after a while. So I added a nother. .... Nothing ....

So then I closed the cabinet doors and left the bulbs burn overnight...

Next day I threw the switch and it came up all --'s! (that's a good thing) Seems to be the opposite problem of a cracked ignition coil?

Finally got time tonight to hook up all HV wires and re-zero all the axis.

I left the power on to the case now.

Hopefully this will last a while.


One thing that IS still up in the air is if I have the amps in the right place or not? One is a dbl 18.7A output, the other is 18.7 on one channel and less on the other. I eyeballed the amps and find no obvious difference anywhere. Would hafta doo some real explorations to get deep enough to maybe see some differences.

If I have the daughter boards in the wrong place I git a motor/amp mismatch alarm. But if I have the right D board and switch the amps around I don't get that alarm. It only seems to follow the board. We are wondering if there IS really any diff in the drive itself - or if the only diff would be in this board?

I'm gunna run it and if I find that I am having torque issues on an axis - then I will know why and switch at that time I guess. ???




-----------------

Think Snow Eh!
Ox

 

[cnctoolcat]

We are wondering if there IS really any diff in the drive itself - or if the only diff would be in this board?

Ox, chances are the drives are different. The base transistors (igbt's) are probably sized for a specific amperage output. You would have to see down deep in the guts of the drives to see the base transistors and other high-power components.

Now, if all the base components are similar, then the output difference would be in the circuit boards...probably just a setting or two..