Wavy surface side milling

[billyghog]

I've been running a Hurco VM20i (2014) for a few years with good results and recently picked up a used 2007 VMX30 for a great price. The machine works really well, except it leaves a wavy surface finish when side milling. Face milling looks great, it's only side milling that has an issue. The machine interpolates a circle just fine (aside from the wavy surface finish) and it's to size and not out of round. It's definitely the machine/control as I've eliminated all other variables: I can take the same tool, holder, program, material, etc. over to my other Hurco and the finish is great. Plus this happens on all parts, different tools, holders, etc. The programable surface finish (G05.3) is set to 10 in this particular program, so that's not the issue. Any ideas?

 

[billyghog]

Oh, and to clarify, this happens in both the X and Y axis as well as any profiled (round, curved, angled, whatever) surface. I've also checked the spindle belt and it looks great and is properly tensioned.

 

[Milland]

Do you have a small granite precision parallel or right angle? If so, align it with the axis you want to test, then run a tenths indicator over it +/- and see if you get a tiny bit of needle movement corresponding with the waviness you're getting.

 

[BT Fabrication]

probably worn ball screws.

 

[bigjon61]

Does it do it in both conversational and NC? That will rule out any programming issues if the problem persists.

 

[Brian]

I'm going to go with ball screws or thrust bearings. What are your backlash settings?

 

[michiganbuck]

Might want to push-pull on the spindle of your older machine under an indicator..and on the new/use machine to see if the spindle or whole assembly has a big difference. Listen to the bearings to compare, and feel for any vibration difference.
A loose preload on a spindle often sounds Ok when not under load.
I'm, not a VMC guy, so just throwing out my thoughts. No experience at all with that machine.

 

[billyghog]

Do you have a small granite precision parallel or right angle? If so, align it with the axis you want to test, then run a tenths indicator over it +/- and see if you get a tiny bit of needle movement corresponding with the waviness you're getting.

I stuck a few things in the machine - a nicely flat machined piece, a (cheap) granite, etc. and ran the tenths indicator over them it and it's quite smooth. Not much of any bouncing, certainly less than a tenth.

The wavy surface finish is less than a tenth according to my indicator, though, so this doesn't really give me a solid conclusion.

 

[billyghog]

Does it do it in both conversational and NC? That will rule out any programming issues if the problem persists.

I don't do anything with conversational, it's all NC. I can take the exact program to my other Hurco and it has a great surface finish, so it's not that.

 

[billyghog]

I'm going to go with ball screws or thrust bearings. What are your backlash settings?

Not sure, where do I find those settings? Both my machines are the Winmax control (the used older machine was upgraded before I bought it). The machine in question is Winmax V09.01.250.33

 

[michiganbuck]

I have heard tell of people using a VMC for occasional grinding, that practice could sure loosen up the machine way tolerances.

 

[billyghog]

I have a Renishaw calibration tool that I stuck in the machine. Runout is essentially zero, no movement on the tenths indicator. I can grab it and pull pretty hard and it moves about 5 tenths. With this same setup in my new(er) machine, it's about the same.

The spindle sounds fine to me with and without load, but I'm not sure that means much of anything as I can't say I've ever heard a bad spindle.

The spindle turns nicely by hand and I don't feel any notchiness or anything strange. If I orient the spindle, there's a significant vibration that can be felt when touching the spindle (or tool). I get this similar vibration on the new machine, just to a much lower extent. I don't suspect that's part of my problem, but I'm not sure either.

I don’t know a whole lot about the machine, but from what I’ve been told it mostly cut plastic. Based on the chips all over the machine, that’s true, as I didn’t find much metal, mostly only plastic, so I don't think it had a hard life.

There's a parameter to monitor each axis, I suspect to adjust for backlash and adjust it out automatically? Those are all enabled. I found the servo parameters for each axis in the control too, but it's all gibberish to me. I can post photos of those params if that helps.

 

[swarfless]

Any difference in finish, 'conventional' milling vs climb milling? Apart from the normal.

 

[billyghog]

Any difference in finish, 'conventional' milling vs climb milling? Apart from the normal.

Haven't tried as we rarely use conventional on aluminum. The cut above was climb milling. Would that give insight into anything that could be wrong?

 

[michiganbuck]

Any axial movement with putting the indicator endways on the spindle and an axial lever push-pull applied?
Any difference in giving the table a lead hammer whack in various places compared to the other machine?
Is the new/used machine sure-footed, level and weight on all feet?

any motor or hydraulic device transmitting vibration?

 

[swarfless]

Haven't tried as we rarely use conventional on aluminum. The cut above was climb milling. Would that give insight into anything that could be wrong?

I reckon if the problem went away with 'conventional' milling it would indicate ballscrew/thrust bearing problem, as has been suggested.
Have to 'fess up .. no CNC experience here but I have seen such on a manual mill.

 

[Brian]

Not sure, where do I find those settings? Both my machines are the Winmax control (the used older machine was upgraded before I bought it). The machine in question is Winmax V09.01.250.33

I don't know where on the Hurco machines to find the backlash settings, could tell you on Milltronic. As others mentioned, if it goes away with a conventional cut the tensions is tight against the ball screws and thrust bearings, it's when you run climb cut and there is enough slop to have the end mill pulling the axis along that you start to see this sort of finish.

 

[mottrhed]

To happen in both x and y it makes ball screws less probable, unless severely loose and the whole table is movng/vibrating. Its more likely the spindle, check axial and radial play, drawbar force, test bar runout, and monitor vibration and heat. My guess is either drawbar issues or spindle bearings are failing/unloaded.

Its possible its in the control, but that being harder to diagnose, Id eliminate the spindle first.

 

[Richard King]

I suspect the spindle bearings tooo. There is a inexpensive way to test it. Go to a auto store and buy a Stethoscope and run the spindle up to where you get the vibration. If you can turn off the brake, listen to the bearings running and then switch off the spindle and let it coast to a stop. Bad bearings will make more noise as it slows down and stops. Yu can check top and bottom and the motor. I had a machine years ago have this issue too and iit was a bad spindle motor. It had a bad winding. If it was running good and all of a sudden it started happening I doubt it is the ball screws. Oh, Have you checked the gibs? Do you have a ground parallel you can indicate on the side with a .00005" indicator or what happens when you indicate in a vise jaw? If you don't see any movement it has to be the spindle. Push and pull the spindle like Buck said.

 

[Richard King]

Any news on this???