Cylinder probe calibration issues

The clue here is "calibrating ABOVE the equator". What you are seeing is a difference between the RADIUS ground on the end of the stylus versus the diameter of the cylinder. I have three cylinder probes and all have LARGER diameters ground into the RADIUS than that of the cylinder. I was able to confirm this on an optical comparator at 50x. In all three cases the diameter of the radius is about 30 to 50 microns larger than the diameter of the cylinder. In all likelyhood... the center of the radius won't exactly align with the center of the cylinder. I haven't checked it on mine. I only use them on a manual Cordax 1805 with Measuremax. I NEVER use them on our FX7.10.7 under PC-DMIS. The results aren't accurate.

If you are only interested in Z readings, why are you using a cylinder stylus?

I'm using the cylinder probe because the surface that I have to check is VERY narrow. The engineer(s) want to make sure that the probe is hitting on the high point of that surface and not deflecting off.

That sure explains it. I have a part where the alignment datum is a knife edge. I have to use the cylinder probe with a radiused end to align the edge and then find the location of a cylinder parallel to that edge. I hope you find where that 1mm discrepancy is coming from. I have a five way stylus adapter that we bought years ago that isn't the "standard" length defined by Renishaw. I didn't discover it until I needed the configuration with A=45. It wouldn't calibrate correctly. Had to custom design it in PC-DMIS. Neil is right. There's a length problem somewhere up the line from the stylus. 

N8VES said:

Had to custom design it in PC-DMIS.

When you say custom design do you adding/removing extensions until the length matches your actual probe. Or, can you actually design your own probes and accessories in PC-DMIS?

You can actually do a custom probe in PC-DMIS using the custom probe builder.

Yeah. The custom probe builder is nice but can be a PITA. I can draw it in CAD faster than designing it in PCDMIS. If you have EXACT dimensions on this custom probe, you can account for some of the length errors Neil is talking about. Years ago I bought some 2x20 ruby stylii form a certain Q manufacturer. They had drilled the shaft hole all the way through into the wrench hole. You know, the hole you put the special tool in to remove or tighten the stylus to the probe. After a few hundred hits in the Z direction it would lose its original length (shorten) until it bottomed out in the wrench hole. By that time is was almost 4mm shorter. I didn't pay attention to it until I rotated the head 45° and it erred with an unexpected hit in the hole it was supposed inspect. Kept wondering why I had to re calibrate the cluster before every inspection. I bought new stylii from a manufacturer here in Dayton and the problem went away. I did keep that stylus and custom re-designed it in the custom probe builder as a back up should I ever need it.

Hexagon apps engineer advised against playing with custom probe builder because according to him it's cumbersome, complex & unreliable piece of junk. 

Then that apps engineer needs to be careful that they don't become an ex apps engineer.

IF you are old enough, and IF you were a computer dabbler back in the old DOS days, the probe definitions are almost identical to the old method of drawing on the screen using DOS.  Colors RGB, 0 to 255, size & location of geometric shape you want to draw.  The only REAL addition is the HOTSPOT stuff.

Yes, I fit both of those.

I just open up the PROBE.DAT file and quickly type it all in.

I also prefer the probe.dat method. Granted I have to get some paper and draw it all out. But have never put the effort into getting the probe builder to work either.  

I also prefer the probe.dat method. Granted I have to get some paper and draw it all out. But have never put the effort into getting the probe builder to work either.  

have you seen my Rubics Cube PH9/PH10?

ITEM:HEAD-PH9-PH10T-RUBIK-SOLVED ARM
comment lines of faces of block
color 0 0 0
comment front face lines
line -10 -30 0 -10 -30 -60 1
line 10 -30 0 10 -30 -60 1
line -30 -30 -20 30 -30 -20 1
line -30 -30 -40 30 -30 -40 1
comment left face line
line -30 -10 0 -30 -10 -60 1
line -30 10 0 -30 10 -60 1
line -30 -30 -20 -30 30 -20 1
line -30 -30 -40 -30 30 -40 1
comment right face line
line 30 -10 0 30 -10 -60 1
line 30 10 0 30 10 -60 1
line 30 -30 -20 30 30 -20 1
line 30 -30 -40 30 30 -40 1
comment rear face lines
line -10 30 0 -10 30 -60 1
line 10 30 0 10 30 -60 1
line -30 30 -20 30 30 -20 1
line -30 30 -40 30 30 -40 1
comment top face line
line -30 -10 0 30 -10 0 1
line -30 10 0 30 10 0 1
line -10 -30 0 -10 30 0 1
line 10 -30 0 10 30 0 1
comment bottom face line
line -30 -10 -60 30 -10 -60 1
line -30 10 -60 30 10 -60 1
line -10 -30 -60 -10 30 -60 1
line 10 -30 -60 10 30 -60 1
comment THE FOLLOWING 6 face CREATE THE BOX OF THE PH9
comment top face
color 0 0 255
face 4 -30 30 0 30 30 0 30 -30 0 -30 -30 0
comment right face
color 255 0 0
face 4 30 30 0 30 30 -60 30 -30 -60 30 -30 0
comment front face
color 255 255 0
face 4 -30 -30 0 30 -30 0 30 -30 -60 -30 -30 -60
comment left face
color 0 255 0
face 4 -30 30 0 -30 -30 0 -30 -30 -60 -30 30 -60
comment bottom face
color 255 255 255
face 4 30 30 -60 -30 30 -60 -30 -30 -60 30 -30 -60
comment rear face
color 255 100 0
face 4 -30 30 0 -30 30 -60 30 30 -60 30 30 0
comment THE FOLLOWING CREATE THE LED
ring 0 -30 -30 0 1 0 20 4
ribcount 10
color 255 0 0
cutsphere 0 -30 -30 0 1 0 8 0 -10000
comment THE FOLLOWING STARTS THE ROTATING SPHERE
autojoint 0 0 -77 0 0 1 0 7.5 -180 180 b angle
color 30 30 30
ribcount 10
cutsphere 0 0 0 0 0 1 50 18 -10000
autojoint 0 0 0 -1 0 0 0 7.5 0 105 a angle
color 142 142 142
ribcount 10
cylinder -22 0 0 -25 0 0 13
connect -25 0 0 1 0 0 M8

This rubiks cube looks awesome! Cool build! Thanks for posting the code!