Jumping X axis Mori seiki MV40b Fanuc 0m

GP Tech · Mar 31, 2015

I spent some time today checking the backlash on X and Y and found that the X axis is jumping .040mm when making .001mm adjustments with the handwheel back and forth.

I checked the backlash with MDI program of G00 G91 X50.; X-50.; Zero dial gauge Z50.;X-50.;X50.;Z-50.;
Backlash comp on X needed to go from 2 to 3 and Y had to go from 5 to 1.

I haven't worked how I'm going to check Z yet, I'll ask that question in another post.

I think this is a servo tuning issue, but I'm guessing and I can't find any info on how to tune the servos.

Also, how would I check the mechanical backlash? should I use a pry bar to get some back and forth movement?

Thanks for any help.

GP Tech · Apr 3, 2015

Here's a video showing the jumping axis.

Vancbiker · Apr 9, 2015

I don't have any way to see videos so can't use that to held diagnose.

Have you gone over the axis mechanically? Loose thrust bearings? Loose ball nut mount bolts? Coupling between servo motor and screw tight with no slop in it?

Since your pitch comp parameters were wrong, I wonder if the X axis has some incorrect parameters. Do you have an original parameter list? Compare the values in the 500-530 range at least but all would be better. If you do not have a list, then compare values between the X and Y axes for the 500-530 series of parameters. Don't worry about differences in between parameters that will be different like backlash or gridshift.

With the covers off, observe the rotation of the servo motor while moving 1 pulse in the forward direction. Then do 1 pulse in the opposite direction. Do this with no comp and then do it again with 1 in the comp parameter. Did the reversal with no comp and 1 comp show a big difference in behavior?

cncFireman · Apr 9, 2015

It is jumping with the MPG only?

GP Tech · Apr 9, 2015

Vancbiker said:
I don't have any way to see videos so can't use that to held diagnose.

Have you gone over the axis mechanically? Loose thrust bearings? Loose ball nut mount bolts? Coupling between servo motor and screw tight with no slop in it?

Since your pitch comp parameters were wrong, I wonder if the X axis has some incorrect parameters. Do you have an original parameter list? Compare the values in the 500-530 range at least but all would be better. If you do not have a list, then compare values between the X and Y axes for the 500-530 series of parameters. Don't worry about differences in between parameters that will be different like backlash or gridshift.

With the covers off, observe the rotation of the servo motor while moving 1 pulse in the forward direction. Then do 1 pulse in the opposite direction. Do this with no comp and then do it again with 1 in the comp parameter. Did the reversal with no comp and 1 comp show a big difference in behavior?

What's happening in the video is when there is a value (even with a value of 1) other than zero in the backlash compensation I get a jump of .02mm when an input of .001mm is made and this happens regardless of using the hand wheel or MDI.
I have rechecked the backlash and it ideally needs .007mm compensation.
I don't have the original parameter list. I now know that this is a bit of a problem, but the machine has been used with the parameters that are in it for years before I bought it, but this doesn't mean that the backlash problem didn't exist it was just that it wasn't a big enough problem for the previous owner to worry about.

Parameter DATA
500 20
501 20
502 20
503 20
504 17000
505 17000
506 15000
507 0
508 108
509 -4232
510 100
511 0
512 0
513 0
514 0
515 0
516 1000
517 3000
518 20000
519 20000
520 15000
521 0
522 128
523 160
524 128
525 0
526 0
527 5000
528 0
529 20
530 0
531 100
532 1000
533 100
534 400
535 2

I don't know what these parameter values mean, if anyone has a parameter list that they could forward me just so I know what the values mean that would be great.

I haven't gone over the axis mechanically. I can do this but I feel that the problem lies within a setting.

Vancbiker · Apr 9, 2015

Would be a good idea to look for someone with a similar machine that has an original parameter list. Just a quick look at the ones you listed tells me you have some issues. For example 522,523, and 524 should all match. These are accel/decel values. When they do not match you will get reduced contouring accuracy.

Did your machine come with the Fanuc maintenance manual? That's where the definitions for parameters are usually listed though sometimes there is a separate parameter manual.

GP Tech · Apr 10, 2015

Vancbiker said:
Would be a good idea to look for someone with a similar machine that has an original parameter list. Just a quick look at the ones you listed tells me you have some issues. For example 522,523, and 524 should all match. These are accel/decel values. When they do not match you will get reduced contouring accuracy.

Did your machine come with the Fanuc maintenance manual? That's where the definitions for parameters are usually listed though sometimes there is a separate parameter manual.

I did get books with it but they are for a MV-JR. I have the ladder book but no parameters.

I know who the original owner is and I just called them but their workshop has shut down for 2 weeks (I hope it's not permanent, Australian manufacturing is in bad shape) so I can't talk to anyone there about the possibility of them having backups for the machine, bit of a long shot but you never know.

What reason would make sense for 522-524 to not be matching?

Also parameter 509 is an odd one out, but it looks like it is for grid shift, Y axis.

machtool · Apr 10, 2015

GP Tech said:
What reason would make sense for 522-524 to not be matching?

Also parameter 509 is an odd one out, but it looks like it is for grid shift, Y axis.

Do you know what version of control it is? I know you said its an OM, but depending on vintage it could be a A, B, C or D modal.

522 -525. According to the first parameter manual I pulled off the shelf - modal B. 522-525 are described as "Time constant of accell / decell for rapid traverse". That will only apply to rapid moves. Not unusual for them to be wildly different. A Y axis thats carries both the Y saddle and the table, will have slower rapid accell rates than an X, that just moves the table or a counterbalanced Z, if they have used the same motors.

509 is just the gridshift of your Y axis. Once again 508 thru 511 arn't related as the are just shifts put in to correct an axis position. Max setting is +/- 32767. But you wouldn't normanly see them set that high. Any more than one pitch of each axis screw and you would move the dog for gross positioning, and fine tune from there with gridshift.

GP Tech · Apr 11, 2015

machtool said:
Do you know what version of control it is? I know you said its an OM, but depending on vintage it could be a A, B, C or D modal.

522 -525. According to the first parameter manual I pulled off the shelf - modal B. 522-525 are described as "Time constant of accell / decell for rapid traverse". That will only apply to rapid moves. Not unusual for them to be wildly different. A Y axis thats carries both the Y saddle and the table, will have slower rapid accell rates than an X, that just moves the table or a counterbalanced Z, if they have used the same motors.

509 is just the gridshift of your Y axis. Once again 508 thru 511 arn't related as the are just shifts put in to correct an axis position. Max setting is +/- 32767. But you wouldn't normanly see them set that high. Any more than one pitch of each axis screw and you would move the dog for gross positioning, and fine tune from there with gridshift.

It's a 1992 Mori MV40b Fanuc controller M4-MF, I don't know what version of the 0m controller that would make it.

GP Tech · Apr 14, 2015

I had a suggestion to check the gibs on the X axis so I'll do that, is this an obvious adjustment or is there procedure to doing this?

Thanks,
Grant.

GP Tech · Jul 28, 2015

Jumping SOLVED!!!

Ok, Just wanted to give an update to this.

I received a phone call from a chap in Melbourne, Australia. That had a similar problem to me on a Chavalier VMC with a Fanuc 0mc controller.
He had Fanuc techs working all day on his machine trying to find the problem, the techs told him about parameter 8103.5, 8203.5, 8303.5 that was for backlash acceleration and could cause jumping as it was not implemented properly by the machine builder.
Unfortunately for him it didn't solve his problem, but after seeing my youtube video he made the effort to contact me to tell me about the parameters.
I tried it and it looks like it has fixed my problem.
I made a quick youtube video showing the difference.

GP Tech · Aug 12, 2015

Hi Kevin,
I want to ask you a question by PM, but your mailbox is full.
Thanks,
Grant.

Vancbiker · Aug 13, 2015

Sorry, didn't know it was full. Cleaned up now so fire away.

Cncrotary · Aug 14, 2015

The jumping of the axis is not likely to be down to drive set up or backlash, as in particular, backlash should be a smooth transition in direction change. As the backlash settings are quite small, there may be a bearing issue or loose locknut somewhere causing the whole mechanical assembly to move (shift) when changing direction, you need to take a closer look at the mechanics.

Dave

GP Tech · Aug 14, 2015

Hi Dave,
In post 11 I go through what was the cause.
From what I've been told the earlier machines that have this parameter turned on weren't tuned for it and that would cause the jumping issue I had.

To me this had to be a servo issue as I was only putting in a .001mm move but getting close to a .010mm overshoot.

lowCountryCamo · Aug 14, 2015

I also have this issue. Fanuc omd here. I watched the video but was unsure which bit in that paramater you were focusing on. I guess you changed from 1 to a zero. Will you state here exactly what you changed? Thank you. This has been bugging me for a while.

GP Tech · Aug 14, 2015

lowCountryCamo said:
I also have this issue. Fanuc omd here. I watched the video but was unsure which bit in that paramater you were focusing on. I guess you changed from 1 to a zero. Will you state here exactly what you changed? Thank you. This has been bugging me for a while.

8303.5 is the parameter.bit, so the .5 means bit 5 and the bits are numbered left to right as 7,6,5,4,3,2,1,0. So the bit that gets changed is the third one from the left.

My video was a little confusing as I shot the sequence out of order, I filmed the parameter off first then with it on. I then swapped it around when I edited it.

Hope the clears it for you, if not let me know.
Thanks,
Grant.

PS
8303 is for Z, 8203 is for Y, 8103 is for X.
And yes 0 switches the parameter off.

machtool · Aug 15, 2015

Grant hunted me up a few days ago for a description of that 8*03 parameter.

I've turned up a blank. The OM maintenance manual, gives absolutely no description of them. Shows them but gives no description I'm guessing that its automatically assigned when you specify motor type at Param 8020.

You would need the digital servo drive motor manual which I cant find / stolen, misplaced or never had one.

Sorry I struck out.

Regards Phil.

Vancbiker · Aug 15, 2015

I've found a chapter in a GE Fanuc manual that describes the 8*03 parameters and the related parameters. It is a function that I do not believe should be needed by an MV40B. Possibly turned on by someone trying to compensate for an underlying problem with the machine. I'll try to scan it into a PDF. I'll post here when I get a good PDF of it.

GP Tech · Aug 15, 2015

Vancbiker said:
I've found a chapter in a GE Fanuc manual that describes the 8*03 parameters and the related parameters. It is a function that I do not believe should be needed by an MV40B. Possibly turned on by someone trying to compensate for an underlying problem with the machine. I'll try to scan it into a PDF. I'll post here when I get a good PDF of it.

Can you explain why you think it wouldn't be turned on for the mv40b and would you expect a MV-Jr to have it turned on?

Only reason I ask about the mv-Jr is I have a suspicion my parameters may have come from a mv-Jr.

Jumping X axis Mori seiki MV40b Fanuc 0m

Cast Iron

Cast Iron

Diamond

Hot Rolled

Cast Iron

Diamond

Cast Iron

Diamond

Cast Iron

Cast Iron

Cast Iron

Cast Iron

Diamond

Plastic

Cast Iron

Stainless

Cast Iron

Diamond

Diamond

Cast Iron

Similar threads