Robodrill Haberle Trunnion Offset X axis

[Theck1983]

My shop has 4 Robodrill D21MiB5adv with 4 Haeberle DTTIB470 trunnions. In a nutshell the trunnions are all built with a 20MM offset in the X axis. For some unknown reason we cannot get a part to run on nominal when milling a feature kicked off a 3° In my mind the X axis being offset works more like a cam when moving the B axis since it doesnt share a pivot point. When going to the positive 3° the feature is oversize by .0025" when going to the -3° the feature is undersize by .0025" Its like the machine itself doesnt know how to compensate for the cam motion while working in TWP. Doing the trig, at 3° I would have to do an offset in Z by about .015" to compensate for it which is totally unrealistic. I dont believe it is a post issue. The code should be the same no matter what. The kicker is we have 2 more Robodrills with older Haeberle trunnions that will run the same program and make a nominal part using the same posted code. Anyone have any experience with offset trunnions and how to troubleshoot them? Ran ball-bar test on Robo and it is within .00004" Axi-Set program with a probe only fluctuates by .0001" every time we run it. I dont think its the machine itself or the CAM software, I think it is TWP not knowing how to pickup the 20MM offset properly and compensate for it

 

[sakis]

Are you using clamp codes while running axi set and in your part program?

 

[Verney&Trecker]

Out of Curiosity have you ran the axis set data exported to the thumb drive thru the axis check software to identify any kind of error that may exist In the B axis?

 

[Theck1983]

Sorry

Are you using clamp codes while running axi set and in your part program?

Yes. It clamps and unclamps when moving in B and C axis similar to how it runs the part program.

 

[Theck1983]

Out of Curiosity have you ran the axis set data exported to the thumb drive thru the axis check software to identify any kind of error that may exist In the B axis?

Oh yes. Every time it runs it exports it to a thumb drive. We have weeks of data and all remain consistent. Very little fluctuation in XYZBC if any at all.

 

[dandrummerman21]

No experience with your machine or how to calibrate/offset the errors it's calculating, etc. But it wasn't clear to me which way you're off.

Are you off +/-.0025 in Z or X when it rotates the B?

And as I'm unfamiliar with the machine and some of this newer terminology (I use a 5 axis trunion in an old 0m machine, where the trunion is controlled by an external box; I have macros written to output new origin locations to work offsets when i rotate B and C to different angles), is the Axi-set program where the machine probes a ball or other known feature at different angles, to determine centerlines for the A and B axis (or B and C, whatever)?