Y Axis Shudder

[alloyspec]

Hi guys.
First post on this sight and I need a point in the right direction. I run a 95 model 6030 with glass scales and a 4th axis. It has an 88HS control and just recently has developed a shudder in the Y axis on rapid positioning at 100 percent rapids. After reaching position it has a slight shake that shows on an indicator for 5 seconds. It has no shake at 50 percent rapids.
Here is the kicker. If I use a G1 to position it has :
The same 5 second shudder at F150.
A 3 second shudder at F100.
A 2 second shudder at F75.
A 1 second shudder at F60.
No shudder at F50.
This happens if you move it 3 inches from the check point or 15. Now I could run most of my parts at 50% and be a little less productive, but some of the parts I run at F80. regularly, and the finish suffers badly. If I slow down the feeds it will help, but really slow these parts down. This is one of the few production jobs we have run for years without these problems.
So I pulled the brushes from the motors and blew out the carbon with compressed air, still there.
I checked the backlash leaving the scales connected and it is good, even with moves of .050. There is the usual spit-second hesitation on the last tenths of backlash, but that has always been normal on this machine.
I loosened the resolver for y with the machine on ( as if I was adjusting the cold start position ) ,and it did not move the Y axis. I think this is normal as it is reading off the scales. It says very little about the scales in the manuals, and or technician does not fully understand how much the scales interact with the control. The manuals are slim in this area.
The balance on the gain for A axis is off a little as the balance chart is 3 to 5 on A. The balance on X,Y, and Z are all within normal range (-1 to 1 ).
We are leary of disconnecting the scales to check it without them , as one time in the past they had trouble getting the machine to cold start without them.
The only other possibly related problem this machine has is a severe shudder/ shake in rigid tap that come and goes. We quit using rigid tap for now. It just starts shaking the whole machine as if it was a lathe cutting corners off a square block. When it starts doing it it might shake 1 hole out of 30. Then on a different hole next part, or none at all.
We do most of our own maintenance (with the help of a service in the next state over the phone), as he is the best we have found with Fadals. If it is a big problem we have him come to repair it, but usually he can tell us what to check over the phone.
We may try switching X and Y boards tomorrow. This is the perfect time to work on it as I have about 2 days the lathe dept. is waiting on gauges to show up.
This machine is the Caddilac of the 4 Fadals we have and I hate to see it having problems.

Any help would be appreciated
Alan

[DareBee]

My experience with that scenario is a worn ballscrew.
The wear in the ballscrew causes the control to think the screw has gone to the requested position, then the control looks to the scale for verification and the scale says WRONG so the control starts hunting for the correct scale position.
The quick fix for this is to unplug the scale and run without it.
My uncle has been running his Fadal for a year this way. It is out enough (doesn't take much) that it hunts but not enough to be detrimental to his work considering it is his "roughout" machine.

I dont know if he has ever had cold start issues, but his machine never gets turned off either.
I suppose you could plug your scale back in before you restart from a power-down situation.

[gus]

sounds more like tach adjustment to me

[Factory Guy]

It sounds like you are describing a chatter situation while cutting, and an oscilation when finding position. It might be 2 different problems but it sounds like something has wear or is lose. If it is oscilating at the end of a move or position, the servo is serching for the scale position. I work with Fanuc controls and this can be fixed with gain adjustment. Is the servo using a belt or direct drive? Check for lose coupling or belt. Check the scale and reader head bolts/brackets. Make sure nothing is lose.

I'm not familiar with the machine you are working on, but I would start looking at the basics. Check the ballscrew for lost motion with a tenth indicator if you can jog the axis incrementally. Check the lost motion when direction is changed. If you have more than a couple tenths, check the ballscrew support bearings and the nut on the end. Also check the location where the ballscrew nut bolts to the Y-Axis. Get an indicator out and start doing a pry test until you determine what is lose. Not familiar with this machine, are there gibs, keepers, type of bearing rails/surfaces? Has the machine quit using lube?

Also, just because you are seeing chatter on Y-Axis does not mean the problem is in that axis. Look at the X and Z also. I wish I had a better understanding of the machine you are working on. Good luck, and I hope some of the above makes sense.

[Factory Guy]

I just read your post again, and what a description!

"It just starts shaking the whole machine as if it was a lathe cutting corners off a square block." lmao

I build lathes that turn diff cases and I've heard that same noise, lol.

It sounds like something is very lose, and you can see this on an indicator. Where are you locating the indicator to see this? Start working your way back with the indicator until you find the source. It doesn't sound like an electrical problem to me, but mechanical. How are you locating the tools? If by a drawbar, how do you check the retention pressure? Just another thought on your interupted cut description.

Let us know what you find. Good luck.

[alloyspec]

Thanks for the replies,
We talked to Fadal directly, and they steered us in the same direction. We first checked the gibs with a tenth reading indicator in the spindle against the jaws of a vise, then the proper way the manuals say to. There was a tremendous difference between the two so that led us to think it is a mechanical problem. We then did the pry test with the indicator at the end of the ball screw and found a full thousandth slack in the thrust bearings on both X and Y, so we are in the process of replacing them. We will probably do Z also if we have the time. This machine is 10 years old.
Also a tip for you Fadal owners out there: we had a 4020 we replaced all the thrust bearings in a couple of years ago and we did not see a whole lot of difference in one axis. Tore that axis down again and found the aluminum block the bearing presses into was about a thousandth deeper than the width of the bearing. I guess after twelve years of use or so the aluminum block got a little hammered out. I just faced off the cap to capture the bearing rigid and it then checked decent. We have had no more problems on that machine.
By the way, do a pry test with an indicator on the thrust bearing block of a 4020 and you will wonder how it will ever make a decent part to begin with. It will make you think about replacing the aluminum blocks with thicker steel ones.
Thanks for all the ideas, I will try to keep you up to date.

Alloyspec

[SCHEMATICS LTD.]

hi, i am a fadal service technician(9 years)
you must first power down.
unplug the scale from card 10 in the nc rack.
power on, and cold start
remove 4 screws holding the front way cover
put an indicater at both the end of the y ballscrew and in the spindle against the table
at control enter bl,2,0,0,0- zero backlash comp
move y axis .001" zero both indicaters
move y back to zero
if the table shows .005 and the balscrew indicates .004 then your thrust bearings have .004 slop and your ballnut has .001 slop
your glass scale will cause shaking if your combined backlash is greater than.0015
if you have any questions, email me at jamie.riberdy@sympatico.ca

[alloyspec]

Thanks again for all the replies, everyone.
After replacing both X and Y axis thrust bearings the thrust bearings are down to .0005 to.0006 clearance. We updated the backlash values before hooking up the scales.
After plugging the scales back in there was no longer any shudder on the Y axis. We will start using the rigid tap in the next few days to see if that corrected the severe shudder in that mode also.

Alloyspec