Holes on an angled surface

[JohnCTB]

Greetings,
When measuring a vertical hole that comes through an angled surface. Using CAD data and auto circle. When I select the hole on the cad it places the circle hits on the same plane as the surface. The problem is, the nominals (X&Y) are not correct. (because it is an egg shaped hole when looked at from the angle)If I change the normal vector to 0,0,1 and move the hits down into the hole using the depth feature I can measure the hole OK but the nominals are still incorrect. The only way I have found to get the correct nominals is to pick my hits down in the hole at the next lower wire frame line and then move them up in the hole to measure. However, that is not always possible.
Does anyone know of another way to do what I want with PCDMIS?

Using Ver. 3.7 mr3

Any and all input is appreciated!
Thank You,
John

[Wes Cisco]

Have you tried rotating your alignment to the angle of the hole so your vector will 0,0,1 or 1,0,0 or 0,1,0 ? I find it is much easier, especially with compound angles to rotate the alignment for the measurement then recall your previous alignment for dimensioning. BTW I almost never get a CAD model, so I can't help you on that aspect of this. HTH

[JoeRZ]

Try Cyl-Pln intersect

or

Level to angled surface, set zero then auto circle.
Be careful with nominals to input for auto circle when angled surface is set level (may have to do some trig or probe hole to find new noms).

[Matthew D. Hoedeman]

No, from what I understand from your question, there is no way to do what you want to do with Pcdmis, it will not 'see' the round circle as data when it pierces the surface at an angle not perpendicual to the surface. You will need a wire frame circle created by your cad guys inside OR a different piece of trimmed surface (like the inside of the hole) that does make a true circle. When you pick the elipse of an off-angle cylinder intersecting a surface, you will not get the true, correct circle. You might be able to get the correct XYZ IF INDEED the intersection of this cylinder is through a totally flat piece of surface data (not square, but flat) by using the elipse option, but I have never used the elipse option.

You might also try finding a cylinder, if you have the 'side' data down inside the hole, then once you have the cylinder picked out in autofeatures, change to circle.

[JohnCTB]

Thanks for your help. You've given me some things to try. Often if I'm not sure DMIS has picked the correct nominal I will go to the data (VISI) and check the feature myself and then if I have to I enter the correct nominal before outputting my results. As many of you know, the quality (inspection) dept., in my case that is ME alone, is held to a higher standard then those doing the machining. In other words if I tell the EDM dept. that their holes are 3 thou off location I better know what I'm talking about!

Thanks again,
John

[mchough1]

John,

One other suggestion, you can take a hit next to the hole then select your hole on the CAD then set your nominals to what ever they are (I'm assuming that the nominals are in the same axis of your hole).


P1 =AUTO/VECTOR POINT,SHOWALLPARAMS = YES
THEO/6.888,2.7456,11.642,-0.9269885,-0.3693438,0.0654031
ACTL/6.8736,2.7421,11.635,-0.9273874,-0.3683044,0.0656083
TARG/6.888,2.7456,11.642,-0.9269885,-0.3693438,0.0654031
THEO_THICKNESS = 0,RECT,SNAP = NO,$
AUTO MOVE = NO,DISTANCE = 0.5

Then assign that point as your surface reference.

ASSIGN/DP1 = (P1.X-P1.TX)*P1.I+(P1.Y-P1.TY)*P1.J+(P1.Z-P1.TZ)*P1.K

Then set your depth

C1 =AUTO/CIRCLE,SHOWALLPARAMS = YES,SHOWHITS = NO
THEO/6.8318,2.722,11.9589,0.7071068,0.7071068,0,0.76
ACTL/6.8318,2.722,11.9589,0.7071068,0.7071068,0,0.76
TARG/6.8318,2.722,11.9589,0.7071068,0.7071068,0
THEO_THICKNESS = 0,RECT,IN,CIRCULAR,LEAST_SQR,ONERROR = NO,$
AUTO MOVE = BOTH,DISTANCE = 0.5,RMEAS = None,None,None,$
READ POS = NO,FIND HOLE = NO,REMEASURE = NO,$
NUMHITS = 4,INIT = 0,PERM = 0,SPACER = 0,PITCH = 0,$
START ANG = 45,END ANG = 405,DEPTH = 0.005+DP1,$
ANGLE VEC = 0,0,1

This works the same as using an initial point but the measured value follows the axis of the hole and is not projected to the surface so you won't see any error based on where the surface is located. the deviation will be based on the theoretical pierce point.

In other words your deviation is based on a 2 dimensional measurement (X and Y) instead of a 3 dimensional deviation forced by PCDMIS when using an initial point.

P.S. Note: add the depth reference if your measuring from the front of
surface subtract the depth reference if your measuring
from the back of the surface.

[Tested]

John,
I run into this same thing many times. When ever I can "on such a hole" I select the bottom of the hole and set the depth up to where I want to measure it. Much like you are doing. Most holes have wire frame at top and bottom. I know this isn't alot of help but working off the bottom of the holes has been my only means of getting many holes like this.

[Rick Coleman]

As many of you know, the quality (inspection) dept., in my case that is ME alone, is held to a higher standard then those doing the machining. In other words if I tell the EDM dept. that their holes are 3 thou off location I better know what I'm talking about!

I wonder how many that applies to in this forum? I got the same situation.

[Matthew D. Hoedeman]

Bingo!

[cmmguy]

....if I tell the EDM dept. that their holes are 3 thou off location I better know what I'm talking about!

I would hope this applies to everyone... isnt that our jobs - to measure parts, give correct answers, and understand and explain how we got them???
Or am I missing something here.

[John Kingston]

I would hope this applies to everyone... isnt that our jobs - to measure parts, give correct answers, and understand and explain how we got them???
Or am I missing something here.

Yes, that is our job. I think he is bringing up the point that CMM results (and Inspection results) get questioned so often. If we are wrong 1 out of 100 times, they throw that in our faces everytime we say a part is out of tolerance.

[cmmguy]

Yes, that is our job. I think he is bringing up the point that CMM results (and Inspection results) get questioned so often. If we are wrong 1 out of 100 times, they throw that in our faces everytime we say a part is out of tolerance.

Ok. and your right, the guys in the shop can put holes in the wrong location but you better never be wrong about your answer..

[John Kingston]

Ok. and your right, the guys in the shop can put holes in the wrong location but you better never be wrong about your answer..

Egg-Zachary.

[Tony.Hirotec.Aus]

We have similar jobs come up regularly. If you level to the plane that the job theoretically lies on and create an auto vector circle that then you can remove the alignment that does the levelling, leaving you with your auto circle with a compound vector