TM3 Vibration/Hunting problems

csharp · Jun 19, 2011

I have a TM3 with no enclosure. I have a thin piece of sheetmetal propped up with a vise grip on it to try to stop some coolant from getting on the floor.

I traveled the machine far enough to contact the sheet metal and bend it out of the way. Really seemed like no big deal. Now when I change directions by using the handwheel or pressing X- then X+ it will basically stop and start vibrating pretty severe.

Any suggestions as to what I have done?

I am in a real bad position as I have so much back log and needed to run.

Matt@RFR · Jun 19, 2011

Does the machine act normally when running a program? If it does, I think I would just run it. There are quite a few questions on the boards about the TM machines being 'jerky' in hand jog mode, so maybe you knocked yours into normal.

csharp · Jun 19, 2011

No does it either way in program or hand jog or push button jog.

csharp · Jun 19, 2011

What I did not explain clearly was that it continues to do even after it is moved in hand jog.

I had to shut down the machine to get it to stop once it started.

HuFlungDung · Jun 20, 2011

Maybe shut down the machine, remove the X axis way covers to get access to the ballscrew, and turn it by hand. Feel for roughness or tight/slack spots, or backlash in any motor to ballscrew coupling. It sure sounds like a tuning issue, but I cannot visualize how a mechanical overload would affect the tuning unless the ballscrew will no longer rotate freely and smoothly.

csharp · Jun 20, 2011

HuFlungDung said:
Maybe shut down the machine, remove the X axis way covers to get access to the ballscrew, and turn it by hand. Feel for roughness or tight/slack spots, or backlash in any motor to ballscrew coupling. It sure sounds like a tuning issue, but I cannot visualize how a mechanical overload would affect the tuning unless the ballscrew will no longer rotate freely and smoothly.

I know that is what was also baffling me. The TM's have the hand wheels and they rotate very freely.

The warm up cycle continued to run after it contacted the sheet metal. I never faulted the machine it actually ran for another 10 minutes.

It was not until I was ready to set machine zero and jogged the machine that I noticed it. Maybe they are not even related.

csharp · Jun 20, 2011

I guess I will leave this one to the Tech's.

I tried once more and as soon as I moved the x axis jog and stopped the machine x axis vibrated so bad I had to shut it down.

machineit2 · Jun 21, 2011

What you describe would almost certainly be an electrical or electronics issue not a mechanical one. A sticky ballscrew should not do that.

Check for loose wires, but it may be drive or motor issue, maybe encoder. I have a HA5C indexer that will sometimes search for the right spot when you jog it. You might want to jog to 90 degs for example and when you should be there the next click, the numbers start to fluctuate. I select another axis and then go back and make the adjustment, as it never settles on the correct number, and might read 89.987 or something. The fluctuations are not visible on the unit itself, just the readout. It is not an issue during operation, as it works fine then.

So while you have that cover off, check the wiring really carefully.

Just a thought, best of luck.

Mike

viper · Jun 21, 2011

Most likely either a servo amp or encoder problem. The encoder can many times be troubleshot by firing the machine up but leaving the servos powered down, then roll the screw by hand and see if the values on screen seem to change normally or act erratic.

If the numbers or values go up and down normally, record the gain values for the amp/servo in the control, then adjust and see what happens. I think it is rare that a servo gain just needs adjusted but never know.

This can also be caused by a bad termination from servo to amp and might be pretty likely being a Haas but also pretty likely that the servo card is out. To diagnose, swap amps between X and Y and see if problem follows the amp or the servo. That will tell you which direction to go.

csharp · Jun 21, 2011

thanks, is it difficult to swap the amps?

viper · Jun 22, 2011

nope, a few minutes. The tech would likely swap things around to diagnose the problem. Check your part numbers but X,Y,Z should all be interchangeable.

Tom3197 · Jun 22, 2011

i've had that problem on Z axis on one of my Mazaks. It was the encoder. I was told the encoder wears out, or you get a slight amount of bearing play in the axis motor and causes the encoder to slightly fight itself for position. The humming is the motor trying to correctly position itself because its seeing the slop.

csharp · Jun 22, 2011

viper said:
nope, a few minutes. The tech would likely swap things around to diagnose the problem. Check your part numbers but X,Y,Z should all be interchangeable.

Tried it but no difference still on the x axis

behindpropellers · Jun 22, 2011

I would find the encoder and make sure it is not loose.

csharp · Jun 22, 2011

behindpropellers said:
I would find the encoder and make sure it is not loose.

I will have a look. Thanks

csharp · Jun 22, 2011

Everything appears tight.

viper · Jun 22, 2011

If you did not take the cap off the servo, you don't know if it is tight. Did you turn the hand wheel like I mentioned and see if there are any glitches in the way the numbers move?

Another test you should perform is swap the X and Y cables from the Mocon board to the X servo amp. Then you would actually move the Y axis to get motion in the X. If it still jitters, you problem is either the encoder or the cabling. Encoders are actually pretty easy to test.

In my experience with Haas, wiring is not the best so don't discount cabling or terminations. I would pull the plug on the motor and test each wire in the cable independently.

Tom3197 · Jun 22, 2011

"Appears Tight" doesn't mean anything if you're talking about an encoder issue. Without the testing equipment to test the motor/encoder the best you can do is taking the encoder off and turn it by hand and put your ear up to it and really listen. If you can hear *anything" that sounds rough or dragging you have an issue. It should be silky smooth and damn near silent turning it by hand.

I'm not sure what the encoder count is on the haas stuff, but it should be in the 6 digit range if not in the millions... it doesn't take much at ALL as far as slop in the motor/encoder to make the encoder read a different position and fight itself to correct. It tries to correct to the right position, over shoots (because of slop or wear), tries to correct in the other direction and again overshoots... over and over again and now you have the humming/vibration.

If you swapped drives and the problem exists on the same axis then like viper said, you have a cable and IMO, an encoder problem.

Viper whats your easy way to test an encoder? I've understood that there really isn't? If ya got a trick let me know!

csharp · Jun 22, 2011

Yes I pulled the cap off the end of the servo. I pushed on the aluminum piece on the encoder. I could not detect any movement.

Yes I did rotate the hand wheel/ball screw and watched the x axis read out. It appeared to go through the numbers very smoothly.

I did switch the cable in to the amp. The one haas tech told me to switch the plug in at the top and the A,B,C (red,white,black)leads at the bottom of the amp. I restarted the machine and no change.

HAAS has indicated that it may be in the servo coupling. I do not really understand how it could be mechanical since at times it will continue with now operator intervention. I was told that if the coupling got bent that the servo would position and a bind in the coupling could cause the servo to backup or forward and the servo would try to re-position to hold in place.

I have turned the screw as slowly as I can and I can't feel any binding.

viper · Jun 22, 2011

If the little crash you had was a little more than you thought, yes, the coupler could be in question but would have to be pretty loose and pretty much broke. The only good way to test for that is to light the servos and try to turn the screw manually to feel for any slop. IIRC, the manual handles are opposite of the servo so pretty handy to test. Keep in mind though that the harder you turn, the more that servo will fight to hold position.

Regarding testing an encoder, what they usually do is lose a channel which makes the drive hunt for a position that is missing. The encoder will have A,B,Z channels along with their complimental not channels. By simply rotating the encoder and testing voltage switching, you can determine if they are working for the most part. I have never seen an encoder "sort of" work without mechanical damage which should be obvious.

Obviously "easy" might be debatable. The encoder wires all come back to the control so testing on those leads is easiest but have to determine the channel wires too. Anyway..

TM3 Vibration/Hunting problems

Stainless

Titanium

Stainless

Stainless

Diamond

Stainless

Stainless

Hot Rolled

Titanium

Stainless

Titanium

Hot Rolled

Stainless

Stainless

Stainless

Stainless

Titanium

Hot Rolled

Stainless

Titanium

Similar threads