FRUSTRATING FINISH PROBLEM WITH VMC. HELP ID SUSPECT

[bob1112]

I was asking on another forum but thought I might take a shot here to figure out where to go on this machine. Machine is an older Haas. Finish problems seem more prominent in the Y axis. I decided to do some testing cutting to try and create a control cut for repairs. I have loaded a pic that shows X on the right of the block, and Y on the left. Very obviously difference. I can pretty much eliminate a spindle issue there just because in most cases, finish would suck in both directions.

Drawbar force = 1200#
Taper = fine
Typical tool runout = .0005-.001
Backlashes, X=.0004, Y=.0005, Z=.002 (needs work on Z screw) X and Y sound like new and very smooth.
Can push and pull head inline with X .001 in either direction and total loss of motion is .0005
Machine has linear guides throughout
Tool holders new

Test cut = 6061 block, .010 finish pass, 7500rpm, 60ipm, 3fl/.500" cutter new, .400 DOC.

Spindle stays cold to the touch, slight vibration in machine (gear box machine), Finish in Z looks very nice, finish in Y seems to be the only issue. Cannot rule out thrust bearings or even servo problems BUT, was able to get a nice cut in Y after slowing everything down to 4k rpm and 30ipm in test cut.

Any help would be greatly appreciated. I have rebuilt about everything in machines over the years and feed confident in troubleshooting but this one is beating me...May be a combo of things.

[HuFlungDung]

I have a similar problem with my older Haas '96 VF3: When I interpolate circles with a heavy cut, the tool vibrates as it approaches and departs the Y axis. The rest of the way round, its quiet and cuts quiet. So far, I still get excellent finishes on a light cut so I have not determined the problem.

Sometimes the spindle cartridge is shimmed to improve the tram on the table. If there is only one shim used, it may destabilize the cartridge mechanical bearing onto the head casting. There may even be a preferred direction of oscillation because the cartridge is practically sitting on top of a ridge making two points of contact, if you ignore the effect of the capscrews. You might try loosening some of the capscrews in some sort of sequence and seeing if the change in tension is reflected when you take further test cuts.

I trammed my cartridge, but I made sure I used two shims about 120 degrees apart from each other and from the non-shimmed side. I also ran some loctite under there to fill the gap. But I just did this as a preventative measure when I redid my spindle bearings, so I'm pretty sure this is not the problem on my machine.

I've been advised to look a the bearing preload on the Y axis as being a bit low. I will still do that sometime.

[bob1112]

Which bearing preload are you referring to? The thrust bearings?


I am hoping someone has seen this type of finish and a story behind what fixed it. At this time, I am reworking all the ballscrews to remove some backlash even though it is minimal in X and Y. reworking Z will only degrade quickly because it needs reground but at least it will give an idea if what is going on. I am shocked that I can get a decent floor finish with .002 lash in the Z.

Pushing on the headstock in the Y direction yields nothing. If I push in the X direction, I am able to deflect it .001 in either direction plus .0005 of lost motion I assume through the guides. I am not in a position to replace them right now so hoping the screw rework will help. I would take "better" at this point.

[Glacern]

Climb or conventional milling? Do you get the same results either way?

Also, try increasing your chipload. But instead of increasing the feedrate, keep the feedrate at 60IPM but lower the spindle speed, as you may be getting spindle-induced vibration. At a very light 0.010 stepover for the finish pass, your chipload is only about 0.0007 IPT (60IPM, 3 flute, 7500 RPM). Cut the RPM in half, keep the feedrate at 60IPM, and see what happens.

-Sol
Glacern Machine Tools

[bob1112]

Thanks but I think we found our problem... Trans and spindle motor are transferring vibes to the spindle. No amount of feed, cutters, etc will fix this. We are addressing the Z linear rails now because they are likely not damping the vibes. We also have improving trans isolation and got improvement in finish. FINALLY found it!