What's new
What's new

One Machine Acting Differently

NOStrum

Plastic
Joined
Sep 25, 2019
Hello,

We have eight Haas VF2 vertical mills, with TR160 trunnions. We feed every one of the mills with the same CAM software, using the same machine definition. This allows the operators to grab a job and put it into any open machine. When comparing machines, every trunnions platter is within .005" from the centerline of A rotation, from one machine to the next. Every one I've measured is within Haas's specifications. We do most of our milling with the platter perpendicular to the table. When one machine indexes ninety degrees from there, (platter parallel to table) its Z is off, and it mills holes through the units. If we take that code and put it on any other mill it runs fine. The only immediate solution I see is to assign another WCS, but that would create 100% more work for the programming department. We are using HyperDent, and utilizing feature based machining. This is VERY scaled down CAM software that doesn't give the programmers control like Mastercam offers. So if we were to implement another WCS that would be written into every program. The programmers would then have to combine NC files because they would be too large for HyperDent to handle. This is not a viable solution for us. Where would you start trying to diagnose this problem?
 
....WHAT?

It would help if you could clearly define the actual problem? This makes no sense.

every trunnions platter is within .005" from the centerline of A rotation, from one machine to the next. Every one I've measured is within Haas's specifications.

Sounds fine so far. Assuming you're talking about the y-center of platter in relation to tilt axis?

We do most of our milling with the platter perpendicular to the table.

So at A90.?

When one machine indexes ninety degrees from there, (platter parallel to table) its Z is off, and it mills holes through the units.

What? 90° which way? A0? A180? or you trying to say A90 or C(B?)90?

Holes through what units? The rotary unit? Workpiece? What?

Are you really talking about the rotary platter, or are you referencing some kind of fixture?


Perhaps some pictures would help
 
Imo.
It´s a specific bug/feature related to your setup.

Use any form of bump by offset of something until it goes away.
The bug is not necessarily related to rotaries, could be tools, or tooltip compensation, or anything.
Try adding 0.001" to everything-comp and see if it changes.


Hello,

We have eight Haas VF2 vertical mills, with TR160 trunnions. We feed every one of the mills with the same CAM software, using the same machine definition. This allows the operators to grab a job and put it into any open machine. When comparing machines, every trunnions platter is within .005" from the centerline of A rotation, from one machine to the next. Every one I've measured is within Haas's specifications. We do most of our milling with the platter perpendicular to the table. When one machine indexes ninety degrees from there, (platter parallel to table) its Z is off, and it mills holes through the units. If we take that code and put it on any other mill it runs fine. The only immediate solution I see is to assign another WCS, but that would create 100% more work for the programming department. We are using HyperDent, and utilizing feature based machining. This is VERY scaled down CAM software that doesn't give the programmers control like Mastercam offers. So if we were to implement another WCS that would be written into every program. The programmers would then have to combine NC files because they would be too large for HyperDent to handle. This is not a viable solution for us. Where would you start trying to diagnose this problem?
 
Correct Bob. The Z is off when the trunnion goes to A90.
We aren't milling aerospace. .005" is plenty for us. It's overmilling by over .1" though.



Five thou is a lot but how far off are you blowing the Z by at flat?
+90 and zero are the only used positions?
XY are okay at both and only Z the problem?
Bob
 
From centerline of A to platter face (when parallel with table) are all within .005" from one machine to the next. So this is not our problem. Although grinding everything the same would be ideal.
 
I can't speak directly about your machines, but on our Hermle machines with Heidenhain controllers there is a function to adjust the kinematics of the machine so that the controller calculates correctly regardless of the actual position of the individual machine components. So if you had a hard hit you can run this test to bring the machine back to it's maximum accuracy. Perhaps Haas has something similar?
 
This is way above my pay grade, but if I were faced with a similar issue the two things I could think of doing

1] download all the machine parameters and since they ought to be text files there are several ways to compare them. Look at the results and consider what you see. IS there a orientation or Z related difference?

2] look at the way the tool length offset is set. Is it in some way different on this machine than the others? For instance what is the actual distance from the spindle nose to the trunnion surface with tool zero, or no compensation compared to the other machines. I cannot for the life of me think of why it would be different in one orientation than another, but something is different in different orientations

It would seem to me that it is either machine side, IE parameters, or user side IE tool setup. If it is something else, you are in deep doo doo...
 
Anytime I see an error of .100” I assume someone screwed up with an edgefinder :)

How were the machines set-up originally? I generally sweep the platter flat for A zero. Then rotate to A90 and edge find my first Y. Rotate to A-90 and find my second Y. The difference between these two Ys times .5 is the difference between platen height and the A axis of rotation. On the TR160, the A axis center of rotation should be ~.120” above the platter. The way I program, that is how I set my Z0 (a single G54 for all operations).

However, if the Z offset was incorrect, you would have the same error on the operations that you do at A90... holes and features would be too deep (assuming you are using a single G54).

I almost wonder if that one machine has Dynamic Work Offsets and Tool Center Point Control turned on and that is causing the conflict. I don’t use it, so I’m not familiar with the ins and outs... other than it alters the interpretation of the code to alter the position in space.
 
Ok you can rule out the cad or program, and start looking at parameters or taking a closer look at the fucked up machine/trunnion. Either that was initially set wrong, it's a different height, different location, or you have a parameter different than the rest.
 
Sounds to me like the Y offset for that platter's work offset is incorrect. That becomes a Z error when the platter is rotated.
 
ya that's an offsetting problem. Are you using dynamic work offsetting? have you checked the platter with an MRZP calibration? I'm assuming these machines have probes...
 
When you say you measured the point of rotation and the platter height, is that relative to the table or to machine zero? Are the machines set up with the machine zero at the same height off the table?
 








 
Back
Top