What's new
What's new

I got Fanuc Power Supply / Amp Issues...

Ox

Diamond
Joined
Aug 27, 2002
Location
West Unity, Ohio
Starting back to maybe Dec when she got cold out, on the cold mornings that would dip below 60* in the shop I started to git "X axis abnormal current" alarm upon boot-up.

???

I powered back down at the case and rebooted. Then it was fine. ???

Within a few days it had progressed into a "X AND Z axis abnormal current" alarms.

I had asked here if enyone had eny ideas and - nodda.

It has gotten to where it is not just the cold days enymore and will doo it every morning even at 70* inside overnight temps. There may have been one morning this last few days that it did not throw that larm?

This is on a "4 axis" lathe and it only does this on the X2 Z2 amp. I believe the amp ratt next door to it is the X1 Z1 amp and has never thrown such alarm - and is on the same power supply. This is actually an 8 axis / twin spindle lathe - so there is a whole bank of amps going on here and I am labeling them once I am dead sure what they feed. But I am 99% sure of the next door amp being the X1 Z1 amp.

I have been trying to run a job recently that is 98% live werk. My live tooling and turret index are the same motor as seems to be pretty common these days. A cpl weeks ago when when I first set this up, thigs were still cool and I don't recall having issues the first day. But the shop may have been below 70* inside yet at that time. But also as it was early on in the new program - I may not have had it running 100% yet and still been tweaking it? And thus not seeing near the duty cycle yet? By the next day things were slightly warmer as I remember and from that day on the machine would only run for two hrs in the morning before faulting out. It will continue to fault out the rest fo the day unless you power it down for hrs on end. So what I was doing was to run as long as possible in the morning untill it would fault out and then power down untill maybe 8 O'clock at night and giv'r another go. This is semi acceptable for this short job, but I have a much bigger 85-90% milling job to put on next. I would be werking my way through proving it out by now had my machine been running properly this last cpl weeks.

Generally the alarm is "Turret Bevel Gear" which would assume that the live toy bevel gears had not meshed properly upon turret indexing/seating and so the turret had not fully seated. I have had this alarm many times in the past - but is always just outta the blue and not overly problematic. However - it aint really a bevel gear meshing issue at all. Most of the time the turret is "lost" and still rotating. Appearing to be more of an encodeing issue? Sometimes the turret is seated and the live toys are meshed just fine with the same alarm.

Also - near as I can tell - and I am having a hard time believeing this - but it appears that the live toy/index motors for both turrets are running off the same amp. ??? However - I am not 100% sure on this... but if it is, and if my troubles are actually an overheat (duty cycle) issue - then it would not seem possible to split up the work between turrets ta make eny diff. ???

I installed an 8K MacKlean A/C on this last week. Got'r ratt chilly in there too! To my amazement not only did it not seem to make all the diff in the werld - but it didn't seem to make eny diff at all! :eek: :confused:

Started checking fans next. I found a cpl aux (non-Fanuc) outboard fans were simply stuck and once freed - ran fine. :crazy: The small inside fan on the power supply was dead. However - I had not been getting errors on the power supply - but rather the drive amp. Fri afternoon I replaced not only the bad inside fan but also went ahead and replaced the bigger outboard Fanuc fan as well. Not long after start-up it faulted out aggin. Thigs weren't looking promissing again. And this was NOT 2 hrs in!

I reset and tried to push it some more. To date all issues had been on the lower turret dooing mostly C werk on the main spindle. Fri evening it was throwing "Live tooling trq" faults a number of times. It is hard to take such a fault seriously when it has a 5/16 C/D and/or a .078 drill in the werk at the time eh? :rolleyes5: I have not had eny alarms on the upper turret since then I don't think - and everything seems to be back to my new "normall".

Also - I don't think that I had mentioned this either - but up untill this time all the faults were durring an index and never a basic "overheat" that would occur at random times during a cut. I had still not gotten an over temp alarm - but faulting out during the cut was more inline with this. I HAVE had live toy overtemp alarms on a job a few yrs ago and on that job I would simply park the machine for an "X" minute dwell to have a bit of time to cool off before continueing. This is NOT the same situation that I have been having.

last night I left the power on to the case - in order to keep the A/C running as it is quite warm here now. (+80F in the shop) However when I came in around noon today - the temp in the case was 19C (Was running between 13 and 8 - depending on ambient temp previously) So that didn't look promissing. Not sure if the exchanger froze up maybe? Especially with the humidity all of a sudden higher? I did not pursue this one yet. Machine only ran for maybe an hr-1/2 at best this afternoon. More likely 1 hr.

I had left the O/H door open last week during the cool weather before I got the A/C on to git it cooled off below my comfort level for the benefit of the machine - and it DID make a notable difference. It ran for 3 hrs before faulting out this time! (That really was a big improvement!) But then I got the A/C on and it started warming up. (Less than 2 months to summer salstace - and then the snowy season will be headed back our way! :D )

There is obviously a heat issue here - but I can't seem to figger out just where? If it was actually in the case it would (should?) have been fixed - or at least allowing a much longer duty cycle at this point I would like to think. ???

All the C werk is on the main spindle which looks to be on the same power supply as the live toys/turrets. But again - I have not had a power supply alarm - and besides - running these little live toys is nothing compared to starting and stopping the main spindle - which it hardly is dooing at all in this part. I am dooing a face and turn, drill/ream, a deburr-ream and cutoff is all - in an 8 minute cycle! I am not cranking the live toys enywhere near their speed limits either.

The only alarm number that I have found on the amps is #7 on thne live toy spindle amp = Overspeed per an internet search.

Machine is 2000 model yr with 18i. 25hp main and 15hp sub. Twin turrets and Y. Live toys every position. Units of interest are A06B-6087-H130 power supply and A06B-6078-H206#500 live toy spindle amp.

I am beginning to think that my trobles are actually outside the case. I am leaning towards the live toy motor heating up causing issues. As the turret is still going around in a "lost" scenario - the motor must be OK. But I am wondering if the encoder on the live toy/turret motor could be messing up as it heats up?

The colder actual shop temps would have impacted the motor encoder - with the A/C not having eny impact.

If you have had enough interest to read this far :typing: - doo you have eny insight?


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

Think Snow Eh!
Ox
 

Ox

Diamond
Joined
Aug 27, 2002
Location
West Unity, Ohio
After setting for a few hrs I went and farred it back up. It ran for maybe close to an hr? When faulted - the turret was rolling around lost - aggin with a "bevel gear" alarm. Motor did not even feel warm.

I had this apart a cpl yrs ago and as I recall - the encoder is only on the turret drive train. Not the motor. I had to replace the belt driving the encoder was the reason that I was in there. Whudda job! Won't be nearly as bad the next time tho as I "fixed" a design flaw. ;) Had to break my arm in three places to git apart the first time. :crazy:

Eny edgycated guesses on this?

I think I'm gunna go see aboot trying to git a # off that encoder...

If someone has eny better ideas - I'm all ears...

I am guessing that my cold morning startups are a diff issue all-together. However I just wanted to include them as they may be connected somehow?


---------

Think Snow Eh!
Ox
 

Tonytn36

Diamond
Joined
Dec 23, 2007
Location
Southeastern US
Ox,
With my experience with Fanuc, I'd start looking at cables very closely. They have a bad tendency for the outer jacket to get hard, crack and split. This allows coolant to infiltrate, and it will creep toward the connectors by capillary action (even uphill).
I'd pull the connectors at the motor, inspect and clean with contact cleaner. CAUTION: When you do this, you will lose the absolute position and it will need to be reset.
We've had to change a lot of Fanuc cables on our 2000-2002 year model machines. If it gets bad enough, it'll arc over in the connectors and usually takes out the motor AND the drive.
The overcurrent alarm is what is pointing me in this direction.
 

Ox

Diamond
Joined
Aug 27, 2002
Location
West Unity, Ohio
I am NOT getting an "OVERcurrent" alarm.

It reads "Abnormal Current" and this is only upon initial power on if the power has been shut off to the case overnight. And also it migrated from just X to X and Z in a matter of just a cpl days or so. Both axis are on the same amp - and other than that - I am having no issues with that amp. We were guessing a cap getting bad in there. ???


In only a few minutes I had the cover off of the axis and to my excitement - the encoder was looking ratt at me on top. It is a dbl banger tho? Must have it's own tach as well. ???


----------

Think Snow Eh!
Ox
 

Ox

Diamond
Joined
Aug 27, 2002
Location
West Unity, Ohio
Let me have a go at that abnormal current issue one more time.

I git this alarm when the control comes up. This is before the servos are enabled.

After allowing a bit of time (presumably for warmup and/or recharge) I cycle power at the case level and the alarm goes and stays away the rest of the day.


And the turrets aren't abs. They doo need homed. (Thank GOD!)


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

Think Snow Eh!
Ox
 

Tonytn36

Diamond
Joined
Dec 23, 2007
Location
Southeastern US
Ok,
So abnormal current...
This still means something isn't right in the control circuitry. Abnormal could mean high, or could mean low. It's just not within a specified range.
This still leads me back to a wiring problem. The control does all these checks during the boot sequence, Yes the power sections of the drives aren't on, but that doesn't mean that some parts of the control sections are not alive.
If the turrets are analog, a spinning turret would lead me to check tach feedback for that motor. I'd still be checking cables....
 

Ox

Diamond
Joined
Aug 27, 2002
Location
West Unity, Ohio
I priced a tach this AM. However - they don't just sell the tach alone. There is a sub assembly with a casting, the pulley, a gear reducer, and then the tach. $2600 out the door!

Well that sat me back down a bit. I decided that I needed to make dern sure that was the problem.

Then it came to me that I was being a Gomer. :o

I thought "SHAZZZZZAAAAAMMMMMM!" :dunce:

I have a twin turret machine eh?

Both look ... sorta alike.....?

I switched the encoder assys around this morning. Went to make a pickup/delivery. Just got back to the same issue as normal. (Glad I didn't order those parts!)

I have a nother amp that's the same as well - I think I'm gunna see how they play - musically. :scratchchin:


----------

Think Snow Eh!
Ox
 

Ox

Diamond
Joined
Aug 27, 2002
Location
West Unity, Ohio
I have a nother amp that's the same as well - I think I'm gunna see how they play - musically. :scratchchin:

That was code for "Musical Amps".


I swapped the drives and were up to an hr+ so far. Should see results soon.

What doo you mean control side too?

FWIW - I physically removed and switched the amps. They are a foot+ away from each other. Cables won't stretch.

Doo you mean something beyond what I have done?


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

Think Snow Eh!
Ox
 

Tonytn36

Diamond
Joined
Dec 23, 2007
Location
Southeastern US
I meant make sure to swap the control side when you swap the amps around so that the correct axis works. (I've seen rookie electricians forget to do that before......)
 

Ox

Diamond
Joined
Aug 27, 2002
Location
West Unity, Ohio
I meant make sure to swap the control side when you swap the amps around so that the correct axis works. (I've seen rookie electricians forget to do that before......)
I know that I said that I bleed green - but I aint ^^^ that ^^^ green. :toetap:



Still running Ox?.....


Well - ????

It ran fine for an hr or so and then it threw the same alarm. Turret was locked down in place - not going around lost. When the turret is lost I hafta power down at case level to clear. So far - it has not dome that yet. But I have had to "reset" many many times. I just shut it down an hr ago for a while.

Even tho it is not "lost" enymore, I am still getting a #7 (Overspeed) alarm on the amp. Turret is seated and gears are meshed. A simple "reset" makes all right with the werld. Reset - scroll back to proper spot in prog and turn it loose aggin.

Now when I git there - the control is on the last line of the live toy macro. The macro that actually runs the live toys. Not the part program associated with the toy. However - if it was code in the macro - it would fault out every time eh? It has not been faulting out on static toys at all.

I am thinking checking whatever prox is associated with the meshing of the bevel gears. Possibly that is warming up and throwing an alarm - but then why does it reset and continue on? And also - why then would it throw the overspeed code on the amp if it was indeed a prox? :skep:

It is NOT messing up during a slide move at all. (Other than a few times on the OTHER turret on fri night for whatever reason? Sympathy pains maybe?) Bared wires should not wait two hrs every morning and should only fault out during an axis move of some sort - so I don't see that one... Once I finished replacing the encoder drive belt on that turret a cpl yrs ago I did set the backlash to the live toys on the tight side - but then - come to think of it - when I check that the toys are meshed - I have a bit of lash, so ...

It is not dooing it only on one specific live toy either. There is 6 on that turret, and since my last post I think it has messed up on 4 of them. Not at all on the first one in the cycle (roughing mill) - which is also the longest use, Once by the next toy (finish mill) with the 2nd longest cycle. Only toy that hasn't tripped it yet would be the first one back on that turret after a slight break from lives on the other turret and a drill/ream cycle on this turret. This appears to show a little cool down cycle.

Once again - I keep comming back to an overtemp amp - yet that's not the alarm - and I HAVE gotten that alarm a few yrs ago on a nother job. However - that machine IS sporting a diff ladder than it had back then... :crazy:

Maybe I'll doo like I did on that other job and throw in a cool down cycle just fer grins?


----------

You guys gitt'n enough wind this last week?
Ox
 

Perry Harrington

Titanium
Joined
Oct 7, 2006
Location
Klamath Falls, Oregon
Just a silly question, but is the overtemp associated with the motor *attached* to the amp, or is the overtemp from the *heatsink* on the amp? My bet is the motor. Sounds kinda like you're hitting the duty cycle for the motor.
 

Ox

Diamond
Joined
Aug 27, 2002
Location
West Unity, Ohio
Yes - that is what it sounds like - but doesn't seem to be.

Motor is cool.

And the A/C didn't seem to help the amp at all.


However - I am NOT getting an "overtemp" alarm. That was a few yrs ago on a nother project. The reason that I added that was to show that this machine DOES have an actual O/T alarm - and that is NOT what I am getting.

I put a 90sec dwell in an already 8 minute long part and not sure if it has helped much. I think maybe some...
 

Ox

Diamond
Joined
Aug 27, 2002
Location
West Unity, Ohio
Since I played musical amps/encoders - I have not gotten a runnaway or lost turret since. However the same turret is still throwing the bevel gear alarm at the completion of the index.

This is all with turret 2 live toys werking on spindle 1.

The alarm at the amp level is 7 (Overspeed) and it is on the main spindle drive amp. This would Shirley be used in the C axis manipulation...

However - the alarm on the screen comes up on turret 2 - not spindle/turret 1.

That all said - I am controlling C1 with the turret 2 control at that time.

I doo have three live toys on turret 1 as well and am not getting any issues there.

Again - a simple reset and go is all that is needed. When I had the lost turret I had to power down and reboot.


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

Think Snow Eh!
Ox
 

Ox

Diamond
Joined
Aug 27, 2002
Location
West Unity, Ohio
The 90 sec dwell did seem to help.

I upped it to 120 sec and I have not had a "bevel gear" alarm since.

I don't know if the new/current ladder has had some changes in the alarm messages?

I still hafta wonder why - if this IS a heat/duty cycle issue - then why didn't the A/C seem to make eny real diff? :fight:

Oh well...


I guess then the next question begs now:


If I am running into a duty cycle issue - and I have plans to continue this schedule - can I upgrade simply to a bigger drive? If so - would it fit in the "single" slot position? Current (pun?) amp is A06B-6078-H206#500.

Would this be a plug/play - or is there some sorta "tuning" involved in these? I assume these are the "digital" drives? And if so - are they on the order of "self tuning"?



EDIT:

After a full day of "MAXX" setting - the A/C tripped out (Lucy not involved) and the cabinet was 29*c inside and machine was running fine and STILL not throwing alarms! :wrong:

I just don't git it? :willy_nilly:

I was planning on adjusting the temp up once I got a baseline enyways...


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

I kood'a tripped out easy - but I've _ I've changed my ways
Ox
 

Perry Harrington

Titanium
Joined
Oct 7, 2006
Location
Klamath Falls, Oregon
Uhh, are you *sure* it's not the motor that needs a rest and not the amp? Seems like it's not in the control cabinet since you put an arctic chill in there. That would lead me straight to everything between the amp and the motor, including the motor. Do you have any way to monitor the motor parameters in real time, like a diagnostic screen for that axis drive?

Have you contacted Hardinge yet and posed this error to them?
 

machtool

Diamond
Joined
Jun 22, 2004
Location
Melbourne Australia
Have you contacted Hardinge yet and posed this error to them?
Is it just me or is this the first time its been mentioned that’s its your Hardinge 65?

Ox.

Would it do your head in, if I suggested it’s probably mechanical. Its been a while since I’ve been in one. I’d have to be standing in front of it, so it come flooding back to me. But I’ve retimed them before. I spent 3 days last week in a Mori ZT 1500Y, with similar control / mechanism that was doing very similar things.

Correct me where I’m missing here. You have One servo motor in that turret that takes care of both turret indexing & live tool drive. There’s a bevel gear like a toothed clutch engagement with prox’s that has to change between index or live tool drive.

In several places with in that motion train, there will be compression couplings or span rings, that lock the live tool drive or turret rotation shafts. No splines or keys, these things are prone to slipping. That creates a conflict between the position the live tool drive needs to be so that the drive tang lines up, and doesn’t clash with the holder ring as the turret rotates, and the physical position the turret needs to be on the curvic coupling.

My best guess is that the over speed / abnormal current alarm you some times get is that when its shifting between the tool drive and the index functions. If some thing in the drive has slipped, you get a clash of meshing which the motor sees, that moves the motor when it doesn’t have a command. That’s possibly why you have also seen the “bevel gear” alarm. Some thing is twisted up tight there, so they don’t mesh with out torsionally winding up the drive train and fighting the servo drive.

Something like being out 1 tooth on that belt you changed would be enough to do it, but just a high cutting load or a snapped tool will move the live tool drive. Hitting the E.stop while the turret is rotating is enough to do it. The inertia of the turret at the rate it is geared will throw them out.

This silly Mori that I was playing with last week, Which every one had deigned had a problem or a crash. The very first guard I took off, and the last to put back on, was a shroud around the tools so that swarf didn’t come back towards the door. (Lower turret). Never noticed it when I took it off, but there’s was a dint from a tool that was too long to clear, from a straight on tool. The tool had span around from where it was loaded and the tool has hit the guard while the turret was indexing. That slipped the drive a few degrees. Engage the curvic coupling and the servo sat there for 12 seconds at 297% load, before it bombed.

If you have a cold motor and a cabinet at 29deg, & swapped drives, I don’t think you have servo / electrical problems. Turret #1 is fine. Putting dwells & timers into your cycle that’s just extending the time you see the problem, I fear.

You can play with what is a grid shift, but that will only move both the live tool drive and the turret, as its coming out of the same spindle / turret servo. If one or the other has moved relative to each other. No amount of parameter changing will sort it out. You have to release the coupling hubs and align the tool drive to the index drive so it’s not fighting the motor when it goes one to the other.

I don’t have drawings of your machine here. I’m fairly tight with the local agent. I can take a look on Monday.

Regards Phil
 

Ox

Diamond
Joined
Aug 27, 2002
Location
West Unity, Ohio
Is it just me or is this the first time its been mentioned that’s its your Hardinge 65?

Yes.


Correct me where I’m missing here. You have One servo motor in that turret that takes care of both turret indexing & live tool drive. There’s a bevel gear like a toothed clutch engagement with prox’s that has to change between index or live tool drive.

Yes - well no actually.

Yes - one motor does both, but the bevel gear would be (AFAIK) the one that the live toys mesh with when they seat. HOWEVER - Yes - there very well could be a bevel gear at the engagement point as well. I have a mental pic of a straight on clutch type unit - like would be on an old B&S maybe - that does the engagement - but I have no basis for that mental pic as I simply don't remember w/o looking either at pics of the breakdown drawing... But there is a possability that this is the point that it was actually "thinking' it's problems were based? But I have found there to be just too many inconsistancies in this whole thing for me to believe this. A good example would be when the turret is still roaming around - lost - and still throwing a bevel gear alarm.... That just makes no bloody sence. And even if so - how did my dwell change this effect if it was a mech problem?


Something like being out 1 tooth on that belt you changed would be enough to do it, but just a high cutting load or a snapped tool will move the live tool drive. Hitting the E.stop while the turret is rotating is enough to do it. The inertia of the turret at the rate it is geared will throw them out.

When I played musical encoder packs I had to reset the grid shift on each of them. If it is off by 5 or 10 the turret will not seat properly of course.


If you have a cold motor and a cabinet at 29deg, & swapped drives, I don’t think you have servo / electrical problems. Turret #1 is fine. Putting dwells & timers into your cycle that’s just extending the time you see the problem, I fear.

With 120 sec dwell it ran all day yesterday. Only hicup was a reamer bottoming on chips in a blind hole. (Z axis thrust alarm)


You can play with what is a grid shift, but that will only move both the live tool drive and the turret, as its coming out of the same spindle / turret servo. If one or the other has moved relative to each other. No amount of parameter changing will sort it out. You have to release the coupling hubs and align the tool drive to the index drive so it’s not fighting the motor when it goes one to the other.

It has a been a cpl yrs since I was deep enough in the turret to remember how fine of a tooth arangement there is on the engage/disengage coupling, but after I had it all apart that time - as well as swapped encoders the other day - I have not had eny issues like you are referring to. Mori must have a slightly diff config? The turret has it's own encoder pack that is only spun with actual movement of the turret shaft. A slight hicup in the engagement of the motor should have little to no effect on turret timing near as I can tell. (Not werking from a motor mounted encoder.)


Uhh, are you *sure* it's not the motor that needs a rest and not the amp?

I aint 100% sure of enything - but when I was having those issues and I checked the motor temp and it was cool to the touch. I don't think the motor is the issue.




My conclussion that I have come to at this point is:

As some of you may know - the Fanuc amps have a heat sink that protrudes through the back wall of the cabinet. On my machine at least - these are out in ambiant air. With this in mind - the A/C would not have effected the heat sink - but the cold morning would have.

The power supply unit has an enclosure around the heat sink fins and a fan on top drawing the air through. The spindle amp does not have this shroud and no fan on top. The Hardinge part of the machine has a whole bank of bigger fans underneath the row of amps blowing up on them in assistance. (All of them - not just this amp)

I have finished that particular part last night and will start a much bigger setup that will take some amount of time and still require a few tools/holders yet. (Figger out the balance as I go at this point) I don't expect to be "running production" for a week. Although I will be chipping away at the setup. ;) Job requires more tools than I have positions.

I think that if I have issues on the next job (summers comming!) then my first action will be to figger out how to git some more fans on that amp. I believe that I am just running a higher duty cycle than it was designed to offer currently. I just don't understand why I was getting the alarms that I was, rather than a "overtemp" alarm like I had in the past on a diff project? Oh well... :crazy:

The nice part of these troubles is that you know your equipment a little better every time and future issues are such a setback. :) It also semi requires that you git another machine like you already have and know so that you make the most of your experiences. (As well as the tens of 1000's of $ in live tools!)



PS:

I have found that while typing info like this in here that I can sometimes have a :scratchchin: moment and git an idea on my own to try next. Call it "therapy". :gossip:


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

Think Snow Eh!
Ox
 








 
Top