probe cal revisited

[Perry Fisher]

I have searched the forum but did not find anything about prehit and retract on probe calibration. If someone has addressed it I missed it.

How critical is the prehit/retract on calibration matching the prehit/retract in the program?

Thanks

[Matthew D. Hoedeman]

The same as it is for measuring your part. It must be a high enough value for the machine to 'get in the groove' before it touches the sphere. I use 5mm for prehit/retract both in the program and in the probe calibration. I will use a higher value at times without worry, but using a smaller will, unless I am checking something with a big tolerance (sheet metal) like +/-0.5mm minimum for no worry, otherwise, the touch speed gets slowed down for smaller prehit/retract.

[Underspec]

It is not critical at all. Only the touch/hit speed, force, etc.

[Mike N.]

I try to keep them both equal as much as possible. It would seem that some machines apply probe comp (for diameters) better than others based on my experience and suggestions from others in this forum. This suggests that more prehit is usually better if possible to allow the axis to "settle" on a required vector prior to touching the surface thus allowing for better comp and dia calculation. Prior to running this Global Image machine I had no luck with diameters being reported correctly.

just my opinion and 2 ct's

[Big Al]

pre hit and retrack distance dose not matter it is touch speed that matters in calibration and programming, there you want to keep them the same. I use 2.54mm or .1 pre hit and retrack distance. Have a nice day.

[Matthew D. Hoedeman]

If your prehit/retract distances are not enough, then the machine will not settle down into the groove of the vector, unless you ahve one he!! of a lot better controller than most of us do. B&S even recommend a certain amount, but I am not sure what it is, to allow the machine to settle to the correct location, direction, etc.

[Sean Harris]

I believe .118" or 3mm is the recommended distance to allow for machine settling. Anything greater just adds time to the program and anything less could induce error due to machine motion. ( for the cal )

In your programs you need to work with the parts you have obviously.

[Perry Fisher]

Thanks...As always some informative input from replys.