Results 1 to 20 of 25
03-30-2009, 07:00 PM #1
Turret Crash, need some help, will not index or lock
Have a customer that tried to push the tool block into the chuck on a machine, obviously not good.
It is a Hydraulic/Servo Turret, Fanuc 0i Mate CNC Control.
When I first got there, if you pushed the index button, (normally moves the turret 1 tool at a time) it would index, but it indexes to the wrong spot, and tries to lock, but it can't because its in the wrong spot. Push it again, it moves to that same incorrect position at the next tool, and so on.
With the machine power off, he moved the turret manually to the correct position, put the power on the machine, when it came up, it locked in properly in the right spot. Pushed the Index Button, at first it would go out like it was goint to move and then not turn, and then after a while it did turn one position, but went back to being in the wrong position, and that is where it sits.
Anyone have an idea of where to start? I am used to the electric turrets, or older turrets with the limit switches in them that you can sort of fool and get it working again, but this one is not one I have worked on before. Noting is bent or damaged that I can see, its just out of line.
My regular service guy can't make it until Friday, and they have people coming to the shop Wednesday, and they need it to be working by then so I am trying to help them out.
03-30-2009, 10:34 PM #2
The Key to
finding out why it isn't working, IS to know how it is supposed to work.
That sounds over-simplified, but I have gone into unfamilliar territory
many many times, and just kept exploring / asking questions / testing
basicly till I could use the process of elimination, to get to the heart of the matter.
Granted time is not on your side. But, if you have at your disposal a
blow-up diagram, that alone may help you find, the culprit that is
obviously quite "out-of-whack".
Whether it's mechanicly distorted, or something registering a false position,
that's been skewed.
03-30-2009, 11:31 PM #3
Sounds like either a param 1850 grid shift for the index motor or a slipped belt driving a geneva index.
Wouldn't have a clue what else it kood be?
Apparently they hit the chuck while indexing rather than just running into it eh?
Think Snow Eh!
03-31-2009, 12:10 AM #4
The servo gets it moving to the required spot then it usually checks for another external signal like a proximity sensor ,then when it gets this signal the plc commands the mechanical clamping operation ( maybe a curvic coupling).
when you had it indexed manually and took it to the correct spot , look for this external signal . ( ours is also of a similar type on the Takiswa cnc lathe - and at the back of the machine there is a green led that goes on when the clamping zone is reached....)
we had a similar problem I had it indexed it manually to this zone - by declamplng then moving the turret by hand to this zone . but I still had to reset the zero on the encoder..
(like OX mentioned)
-( I was getting '300 APC alarm' or something like that).
03-31-2009, 01:51 AM #5
M1M, yes you are exactly right, I tore into it a little today, Its not exactly what I am used to working on, everything I have done before always had what were pretty much limit switches inside the cover. I could fool the machine to think the turret was in position and OK to index by holding in the one switch, then it would start to turn, lock in the next position, and then it was done.
OX, yes you are right, hit the chuck while indexing.
Sudtechcnc, thats also a good suggestion, I will try to look for that.
I have a set on instructions that a great gentleman sent to me from CNC Zone, pictures are great, they were in Chinese but I translated it, (no I cant read Chinese, I copied and pasted to Babelfish Translator)
But they involve basically tearing the whole turret apart, it seems like it should be easier than that, but I dont know, maybe its not.
That one basically tells me to take the turret off, and then I think that Curvec joint that Sudtech is referring to is what would get reset, but I havent done that.
Its a tough situation. First, the customers are new to CNC (they hired an operator/programmer) so they dont exactly understand warranty doesnt cover a crash. I told them that I would do it for free, because they bought them from me, but the other service guy would cost them, because not a warranty thing. But I am trying my darndest to get them going, they are nice, but do not understand that #1, I am doing it for free because they were a customer, and #2, I just got a call at 2pm today and was there within an hour. I had a hell of a time working on it, because I dislocated my shoulder last Thursday, and I can barely move it. So I am doing this one handed too, trying to help them out.
And #3, I wanted to stay late this evening and try to fix it, but they close at 5, and there was no way I could stay late to work on it. It HAS to be fixed by Wednesday, so I was hoping it was something simple.
ANY more suggestions are greatly appreciated, hopefully they involve just something in the control, because I am only one handed for a while! (just kidding!)
03-31-2009, 04:56 AM #6
Has to be fixed by Wednesday, but they won't stay past quitting time when you're trying to help them recover from their own screw-up? My time of working free for someone like that would get real short, real fast.
And don't buy into their crap that they don't understand a crash isn't covered by warranty. If they slam their new car into a bridge its not warranted. If they jam a screwdriver thru their new toaster its not warranted. Or if they try to mow a boulder with their new Lawn Boy, that's not warranted either. People who try to play the dumb game just plain piss me off. When I jerked a part out of a power chuck and ruint my brandy-new $300 hardjaws, there weren't a soul in the world who had to tell me it wasn't a warranty item
03-31-2009, 05:00 AM #7
03-31-2009, 06:08 AM #8
You didnít say what the machine was. Does it have a Fanuc red cap servo to rotate the servo? Hydraulicsí for lift, lower/ clamp on the turret drum. Most of those are controlled by the Mate P side and the ladder. There just set up, so that from a home or #1 position, the motor does so many revolutions to index the turret 36 deg for a 10 station, 45 deg for an 8 station etc. Absolute encoders so they donít need referencing.
It gets very tricky to describe. If the manuals have a section about replacing the encoder, that should cover it, and how to establish the zero position.
What I normally do is to pull the plug on the servo encoder. Leave it off for 5 minutes or so. It will come up with a battery error, saying that it has lost position. Then itís a matter of finding the right parameter. Normally parameter 11 bit 0.(APCX) in the right group for that servo. There can be numerous groups if it has upper and lower turrets. But its on the Mate P side, not your normal machine parameters. Just as you get to the machine parameters, System - PMC-Parameters. Press the right arrow twice. That should bring up the Mate P parameters. If you push the solenoids to unclamp and get the turret drum to physically seat in the correct position, on the Curvic coupling, its just a matter of setting the encoder count to zero, by changing the APCX bit (11.0) to ď1í
You can do it mechanically by releasing the turret, and getting the back into line where the motor thinks it is. But is far easier to just to re-time the motor to the coupling by pushing buttons and resetting the encoder to the new zero.
I cant spend any more time on it on the chance that the motor isnít Fanuc and integrated with the control. But thatís the most common systems I see. And given your customer went home 3 hours ago. (Relatively speaking. Its 8.00p.m down here) I have 3 hours of catching up to do.
Any customer of mine that wouldnít stay back until 6 or 7 pm or put on beer and pizza for latter than that, would get a black ban for life. Iíd never darken their door step again. Just leaving a few hours left for tomorrow, totally stuffs the next day, and you play catch up all week.
03-31-2009, 07:30 AM #9
Appologies if it's been suggested as I only scanned the replies.
Have you referened the turret (ie zero returned it or homed it).
It's not a simple thing like it's lost its position?
03-31-2009, 09:53 AM #10
Yes, it is a fanuc Red Cap servo that rotates the turret, and hdraulic clamp/unclamp. You got it exactly right.
I agree with the warranty thing too, problem is, the owner is not a machinist, they have a guy who is, and he is pretty darn good, but I did not want to tell the owner its from a crash, because I dont want to get anyone in trouble. I said more or less its from hitting the chuck, from a mistake, but was sure to say it happens and you can't be mad at anyone for it, I am just trying to be nice to everyone, thats all.
I will go there today and try a few other things suggested here, and hope it works out. problem is and I will be the first to admit it, I don't exactly know what I am doing with this one, all the turrets I have worked on were different. Thats why I have a service guy, however, he cannot make it for a while, now they say next wednesday.
I will be there in about 1/2 hour, so ANY more suggestions would be great! (even an explanation on how to reference the turret or more into what Phil said, I GREATLY appreciate it!)
03-31-2009, 10:42 AM #11
setting the Turret
If the turret is put up correctly the first thing that gives in is the turret disk
(the one where you put all the tools in) there are normaly about 8-12 hex bolts inside that holds it in its place(open the turret front disk). This IS suppose to move if there is a bad crash
When turret is locked and spindle error is on the bolts must be loosend and the disk moved to the correct line with a masuring clock, it must be within +/- 0,01mm.
(you can do it on the turret disk itself, but best is to own a spacial tool for this prosses.)On the back side of this disk and on the turred (a ring next to the disk) is suppose to be a marking (a line that crosses both disk and turret) when in turret mode you can move the disk by hand so the line on bouth sides do match. this one has to be lined if you are in a situation when you cant move the turret (signl fault)(if the turret moves by its self and is in correct tool number but the pocket does not fit with the turret spindle and you have lined the turret)
Then you send me an E-mail (sorry about my spelling)
Hope this helps.
and ps. the reason for long time on repair is most of the time that this Mr. SOMEONE didnt do anything and nobody knows what happend so I have to track back all the things that could have happend in stade of Knowing just what happend.
Last edited by Nordic; 03-31-2009 at 01:11 PM.
03-31-2009, 01:53 PM #12
My $ is still on "Grid Shift Param" 1850 or the one that machtool mentioned...
Think Snow Eh!
03-31-2009, 04:17 PM #13
The importer should have a step by step procedure for re zeroing the turret servo. It probably has absolute pulse coder on motor (being so new). Procedure varies by OEM but basically they have a method of unclamping turret, jogging or MPGing into #1 position then setting that position. Without the OEM procedure, I believe you will be hard pressed to resolve this issue.
03-31-2009, 07:00 PM #14
Well, to find out if the turret is "normal" axis, but PLC controlled, look at any axis parameter - if there is one more to X&Z, then play with 1815 - when the turret is down, reset bit 4 to zero, power off/on, 1815.4=1, power off/o, alarm 300 must disappear and it should go.
As far as you said "0i-Mate T", if so, then a.m. is not possible, the drive should be I/O link drive. First, with hydraulic on, CNC off, align the turret at any position, but down clamped well. I/O link control screen, as Machtool said, can be found at [System], right arrow until find in the middle "PMM". Push it, there is "message" and 'system" buttons. After system, PARAM. Set prm 11 bit 0=0. On PMM/message, alarm 000 and/or 224 will appear. Power off/on. Now you should be able to make one turret move in jog. If not aligned, repeat with manual alignment. Keep rotating until find tool No.1 Power off/on. Go again to system/pmm/system/param and set prm 11.0=1. power off/on. Now there should be nor 000, nor 224 alarm on PMM screen.
That should be all. Look at the program numbers, look for o20 or o8020 - inside you must find all the PMM parameters. If not, or no parameters inside, rename these to any other number, go again to system/pmm/system/param, push OPRT, right arrow, READ!!!! Be carefull, READ means drive -> to CNC memory, punch means CNC memory TO THE DRIVE. After confrming direction TO CNC memory, push EXEC.
Save a copy of everything on a card, with such a customer you definitely will need this copy.
Mode MDI, offset/setting, PWE=1, CAN+RESET, I/O DEVICE=4. Put the memory card in.
set prm 3202.4=0 (to save all macros), return to 1 at the end.
SYSTEM >> ALL I/O PRGRM OPRT PUNCH -9999 O-SET prg.all F-SET EXEC
< PARAM OPRT PUNCH prm.txt F-SET EXEC
< > MACRO OPRT PUNCH mac.txt F-SET EXEC
< PITCH OPRT PUNCH pitch.txt F-SET EXEC
SYSTEM PMC > I/O M-CARD WRITE LADDER EXEC
I/O M-CARD WRITE PARAM EXEC
Set 3202.4=1 and PWE=0 and save the memory card till next call.
04-01-2009, 10:37 AM #15
THANKS Guys, I got directions from the factory, I just emailed them to the customer, but they can't get to this screen.
LT-400 turret setting for 0i MATE TC
1. Check K0.0=1, select 0i MATE TC (PMM)
2. Set K1.0=1
3. Under HANDLE mode
4. Press FEED HOLD and SPINDLE STOP buttons simultaneously. The turret will be unclamped.
5. Press direction keys , to rotate turret to tool 1
6. Press RESET button to clamp turret
7. Set K1.0=0
8. SYSTEM [PMM][SYSTEM][PARAM]Set parameter 0011#0=1power on again.
Any more ideas? The customer said this screen will not come up when they try to get to it.
04-01-2009, 06:51 PM #16
friends and folks that have done me a good deed somewhere along the way.
Maybe some-one broker than me, etc.
SO, I make it very very clear, that I know d*** well that a problem is the result of
X or Y or Z senario, and some-one better start talk'n if they want the bennefit
of my low-cost assistance. Ever again !!
04-01-2009, 07:13 PM #17And #3, I wanted to stay late this evening and try to fix it, but they close at 5, and there was no way I could stay late to work on it. It HAS to be fixed by Wednesday, so I was hoping it was something simple.
HAS GOT TO!
Werds that really should be removed from the English language...
Obviously in this case he didn't even want it running REALY REALY bad even. Not enough to alter his evening plans enyhow. That's a reasonable guage to measure by...
The sun came up this morning in your part of the werld I assume?
Think Snow Eh!
04-02-2009, 06:30 AM #18
Sorry Iíve not been in the office. Iíve barely had time to scratch myself down here. Iíve got machines in pieces all over town. And two truck loads of new machines in crates that hit a customer today that I forgot about.
As it turned out. Your turret is controlled by the PMM side. (Thatís Power Mate Management). Lots of the machines I see have the servos for Live tool drive and turret index hidden away there. You donít find the set up for those in the normal system parameter list. It took me a long time to find the first time.
Iíd originally thought, like Ox, that it would be the grid shift or the axis origin, just like setting a home position on absolute axeís. But thereís a totally different manual for that side of things. And I think thatís written worse than the normal manuals. This bit Iím talking about 11.0. Itís described in the manual as ďAbsolute encoder is availableĒ Iím thinking ďYeah I know its got one of thoseĒ thatís why you donít reference it in the morning, and thereís no switches for a zero or number one position.
What it doesnít say is if you toggle that bit from a zero to a one. Thatís your new zero position. If you have the coupling clamped. Call the servo zero at that position. Its going to rotate the turret correctly from there.
Getting back to your client. The problem with finding the right screen. When they say they cant find the right screen. Is it the Keep relay screen or the PMM one? Iíve never had to enable the PMM by using a relay. Thatís a new one on me. But thatís in the general area where you find the parameters. Right next to it is Timers.
What I think they wont be seeing is the PMM page. I think they missed a step in those directions. They are saying to be in handle mode. Thatís fine to get into the tricky two button clamp / unclamp Ė rotate stuff. I think to see that screen you have to be in MDI maybe. Or another mode other than handle.
I needed to do what your doing on Wednesday. I pulled a lower slide out of a Miyano BNE 42 for refurb. I had to disconnect the turret motor for that. Thatís a 4 axis with sub machine. It still took me 5 minutes to find that PMM page. That trick to that was you had to press the button for Head #1 only. Any other combination and it wasnít there.
Iíd be getting them to try looking for it in other modes. I still donít know your machine, but make sure its on the main head / slide if itís a multi.
Thereís been some great input in this thread. Now that you have seen the toothed curvic coupling. Nordic is correct. Ultimate position comes from that coupling. Once those teeth lock together the servo does nothing to affect the position. So if the tools are too high or too low, or your drill doesnít look down the centre line of the spindle you physically have to manipulate that.
But you had said that the turret was rotating when it took the hit. Because it was unclamped and the teeth werenít engaged it shouldnít have moved the coupling halves.
It would have just slipped on the centre drive shaft that turns it. Thatís why the conflict with the servo. Its turning it to a wrong spot that isnít over the curvic teeth to bring it back and clamp it sweetly.
The F-Bu guy from Bulgaria. He explains that procedure better than I can. Iím a nuts and bolts guy. Iím printing that out to carry around with me. He makes a good point about the normal 1815 bit 4 parameters. Thatís on the conventional parameter side. Where the grid shifts and backlash are.
Thereís a higher than better chance that if you have hit the turret hard enough, you will have also moved the true X & Z. Its always nice to have the turret set at a nominal dimension especially in the X axis. Radially or up and down (imaginary Y axis), you have to re-set the curvic coupling. But the X & Z you have to do by re-zeroing the axis. Either with 1815.4 or the 1850 gridshift parameters.
04-03-2009, 12:21 AM #19
I just wanted to tell you guys THANK YOU for the help. It is back running again. I don't think the shear pins are gone, but so far its working.
You guys are AWESOME with the replies, it helped a whole lot, and it is working. It was real hard to get to the screen, but then at one point it just worked, I know something was different, but to be honest I dont know what but when I originally did it, the screen was not there, after many tries I guess the right screen came up.
Anyway, THANK YOU again.
Also just to add to the thread, I have posted pictures of step by step to fix the shear pins, just for the future, IF this happens to anyone. Just may help someone out down the road who searches for this.
04-03-2009, 12:22 AM #20