Probe comp problem

[Steven69]

I was running a check on my CMM for the new production manager, and a wierd thing started happening. I am measuring a caliper calibration master with (3) different programs, and the same thing is happening in all (3) programs. It seems that the probe compisation is doubleing on some hits. I am using a 4MM probe, and I am taking a line (2 points) and then a single hit in the middle of the (2) for the line. The Z comes out pretty good for the line, but the point is reading 0.079" off in Z when it is compared to the line, or about 2MM. Anyone have any idea what is causing this? It is repeating these results. Also, the system has crashed twice tonight, which may have something to do with it, but I don't know. The code is as follows for one of the mysteriousreadings.

5.0001LINE =FEAT/LINE,RECT,UNBND
THEO/-0.5937,0.0138,-5.0001,0.9999981,0,-0.0019267
ACTL/-0.604,0.0264,-5.003,1,0,-0.0001974
MEAS/LINE,2,WORKPLANE
HIT/BASIC,NORMAL,-0.6071,-0.0001,-5.0028,-0.0019267,0,-0.9999981,-0.604,0.0266,-5.003,USE THEO = YES
HIT/BASIC,NORMAL,0.1926,0.0291,-5.0119,-0.0019267,0,-0.9999981,0.1722,0.0263,-5.0031,USE THEO = YES
ENDMEAS/
5.0001PT =FEAT/POINT,RECT
THEO/-0.0155,0.0183,-5.0001,-1,0,0
ACTL/-0.0899,0.0264,-5.0814,-1,0,0
MEAS/POINT,1
HIT/BASIC,NORMAL,-0.0155,0.0183,-5.0816,-1,0,0,-0.0899,0.0264,-5.0814,USE THEO = YES
ENDMEAS/

[Matthew D. Hoedeman]

I was running a check on my CMM for the new production manager, and a wierd thing started happening. I am measuring a caliper calibration master with (3) different programs, and the same thing is happening in all (3) programs. It seems that the probe compisation is doubleing on some hits. I am using a 4MM probe, and I am taking a line (2 points) and then a single hit in the middle of the (2) for the line. The Z comes out pretty good for the line, but the point is reading 0.079" off in Z when it is compared to the line, or about 2MM. Anyone have any idea what is causing this? It is repeating these results. Also, the system has crashed twice tonight, which may have something to do with it, but I don't know. The code is as follows for one of the mysteriousreadings.

5.0001LINE =FEAT/LINE,RECT,UNBND
            THEO/-0.5937,0.0138,-5.0001,0.9999981,0,-0.0019267
            ACTL/-0.604,0.0264,-5.003,1,0,-0.0001974
            MEAS/LINE,2,WORKPLANE
            HIT/BASIC,NORMAL,-0.6071,-0.0001,-5.0028,-0.0019267,0,-0.9999981,-0.604,0.0266,-5.003,USE THEO = YES
            HIT/BASIC,NORMAL,0.1926,0.0291,-5.0119,-0.0019267,0,-0.9999981,0.1722,0.0263,-5.0031,USE THEO = YES
            ENDMEAS/
5.0001PT   =FEAT/POINT,RECT
            THEO/-0.0155,0.0183,-5.0001,-1,0,0
            ACTL/-0.0899,0.0264,-5.0814,-1,0,0
            MEAS/POINT,1
            HIT/BASIC,NORMAL,-0.0155,0.0183,-5.0816,-1,0,0,-0.0899,0.0264,-5.0814,USE THEO = YES
            ENDMEAS/  :eek:

It looks like the vector for the point is incorrect for the line you are measuring it on. The vector for the point should be the same as for the hits of the line if you want the point 'comped' into the line.

[BOB PARSONS]

I think Matt has it right, good eyes Matt.

[Matthew D. Hoedeman]

I think Matt has it right, good eyes Matt.

Well, it was easier with the []code stuff, I did that in my reply, then before posting, I preview it so I could see it like I was looking at Pcdmis.

[code] goes before and [/code] goes after Pcdmis edit window code and it will look like Pcdmis

Then, once it looked right, I was able to spot that pretty quick.

[craiger_ny]

Nice useage ot the [noparse][/noparse] tags by the way

[Matthew D. Hoedeman]

Nice useage ot the [noparse][/noparse] tags by the way

bite-me, boogerboy. We all gotta learn something sometime. I thought about doing it the way I used to just to see if I could get a rise out of you, then I thought about poor Paul(a) and what happened to him/her when he/she got a rise out of you. By the way, has he/she started walking upright again?

[craiger_ny]

bite-me, boogerboy. We all gotta learn something sometime. I thought about doing it the way I used to just to see if I could get a rise out of you, then I thought about poor Paul(a) and what happened to him/her when he/she got a rise out of you. By the way, has he/she started walking upright again?

You flatter me

[rangerboat72]

Is this RESOLVED ? ? ?

[Steven69]

Is this RESOLVED ? ? ?

No, it is not resolved. Here are (2) morefeatures checked. Again, this is a 2 point line, and then a point is taken between the first 2. The vectors are the same, yet the 'X' positions are 0.078" off!

10.9973LINE=FEAT/LINE,RECT,UNBND
THEO/10.9973,0.1279,1.953,0,1,0
ACTL/10.9979,0.0699,1.9601,0.0001852,1,0
MEAS/LINE,2,WORKPLANE
HIT/BASIC,NORMAL,10.9981,0.1217,1.953,1,0.0002171,0,10 .9979,0.0699,1.9601,USE THEO = YES
HIT/BASIC,NORMAL,10.9979,0.8121,1.9529,1,0.0002171,0,10.9981,0.8809,1.9601 ,USE THEO = YES
ENDMEAS/
10.9973PT =FEAT/POINT,RECT
THEO/10.9973,0.4021,1.953,0,1,0
ACTL/11.0759,0.4316,1.9601,0,1,0
MEAS/POINT,1
HIT/BASIC,NORMAL,11.0761,0.4021,1.953,0,1,0,11.0759,0.4316,1.9601,USE THEO = YES
ENDMEAS/

[cmmguy]

Why is the vector of your point in the +Y direction?

The vectors on the hits reflect direction away from the surface.
The vectors on the feature represent feature direction.
Points dont have a feature vector but their hits do.
The point between the line should have a hit vector of 1,0,0 not 0,1,0

[Matthew D. Hoedeman]

No, it is not resolved. Here are (2) morefeatures checked. Again, this is a 2 point line, and then a point is taken between the first 2. The vectors are the same, yet the 'X' positions are 0.078" off!

10.9973LINE=FEAT/LINE,RECT,UNBND
THEO/10.9973,0.1279,1.953,0,1,0
ACTL/10.9979,0.0699,1.9601,0.0001852,1,0
MEAS/LINE,2,WORKPLANE
HIT/BASIC,NORMAL,10.9981,0.1217,1.953,1,0.0002171,0,10 .9979,0.0699,1.9601,USE THEO = YES
HIT/BASIC,NORMAL,10.9979,0.8121,1.9529,1,0.0002171,0,10.9981,0.8809,1.9601 ,USE THEO = YES
ENDMEAS/
10.9973PT =FEAT/POINT,RECT
THEO/10.9973,0.4021,1.953,0,1,0
ACTL/11.0759,0.4316,1.9601,0,1,0
MEAS/POINT,1
HIT/BASIC,NORMAL,11.0761,0.4021,1.953,0,1,0,11.0759,0.4316,1.9601,USE THEO = YES
ENDMEAS/

DON'T LOOK AT THE VECTOR of the LINE and the vector of the POINT, THOSE should NOT be the same, HOWEVER, the VECTORS of the line HITS and the vector of the POINT MUST be the same if the point is on the line, those are NOT.

[Steven69]

Why is the vector of your point in the +Y direction?

The vectors on the hits reflect direction away from the surface.
The vectors on the feature represent feature direction.
Points dont have a feature vector but their hits do.
The point between the line should have a hit vector of 1,0,0 not 0,1,0

I'll be Da*ned!! You are correct, sir. I went thru the code on three programs that I had written that night, and for some reason, the vectors on some of the points were switch!! Now I just have to figure out why they were switched. These were manual programs, so the hits I took were coming from the same direction, yet only a couple had the vectors messed up!!! Thanks much!!!