Jump to content

CMM won't measure the feature!


Recommended Posts

I have a simple base alignment. A cylinder in XY is both spatial Z and XY origin, planar rotation is done with a 3d line, constructed from two radii. I probe the start alignment, the CMM changes probes and heads towards the part, but stops (red light), and status window displays attached.
150_87af607444d64e149251029d12077907.png
Link to comment
https://qualityforum.zeiss.com/topic/5590-cmm-wont-measure-the-feature/
Share on other sites

Remove the planar and you will be fine.

The cylinder removes 4° of freedom 2 rotations and 2 translation the only thing left to do is stop the last translation.

Example:
If you measure a Plug Gage using the a cylinder for the OD and a Plane for the front the alignment is:

Spatial - Cylinder
Planar - NONE
X - Cylinder
Y - Plane
Z - Cylinder

You don't need planar because the Plug Gage can spin about the cylinder axis and you should get the same result.

If you try to do a planar to stop the last rotation with a feature that is in the same as spacial it throws the values way off into space because it doesn't understand what it stops.
The part is fixtured near the center of the machine, and the closest
axis to a limit is Z, there is more than 6" of Z travel before the limit.

This was a problem when this same program was running on the 322
O-inspect. That is part of the reason for transferring it to the Contura.
I have had my CMM travel off into space when using a cylinder as a BA when programming from CAD model if the model is not in the same orientation as the BA. I found that if I plan to mount the part different than the original CAD model I would start the program by selecting a plane and then create a BA with just the plane in the spatial direction I am looking for. After the direction is established then I would select my cylinder and use it as spatial after.
So far as the original error is concerned, IMO the likely cause is that the CMM is determining the alignment of the part is at an angle in relation to the CMM axes. This cause Calypso to calculate the next clearance position based on the newly determined alignment. The problem stems from the fact that the projection error from the alignment to the other side of the machine makes the next clearance position be calculated as somewhere outside the machine envelope. The machine sees this position and determines - "hey I cant get there" and produces the hardware end position error.

How to fix it.

A - Verify part Alignment
B - More stable base alignment.

Part A is self explanatory - make sure the part is not actually off on some angle, this can be done with a fixture / a level or by making an alignment to the CMM axis and probing the part and looking at the error in relationship to that alignment.

If the part is already aligned correctly or you have now got it aligned - the next step is base alignment.

3D feature for spatial - from the picture the OD as a cylinder looks like a good start.
2D feature for spatial - one of the circles on the teeth would work - no reason to complicate it with a 3D line

X od cylinder
Y od cylinder
Z perhaps center of your spacial - or something else you can probe for even a single point.

Then loop the alignment for best results.
I just tried running the base alignment manually, that produced the same error.
I tried using just the +Y radius for the planar rotation instead of the 3d line, that didn't work either.

Derek's explanation makes me think this has something to do with the fact that this program was originally written and ran an O-inspect, and somehow the Contura is looking at the program as if it were still on the O-inspect? I'm baffled.
The base alignment must be flipped and you should see that the in the Plan, Navigation, Blocked Edges window, the black circles will be on top.
Like others, I suspect the 3d line is what is flipping it.
Go to resources and scroll down to, utilities and then at the bottom select 'set base alignment to zero'.
Then run a manual alignment and be sure to measure the two radius's in the order you have them recalled from, front to back or vise versa. Measuring the back radius when it's supposed be the front radius first (if that's what you recalled in the line 1st) will flip the alignment upside down.
  • 2 weeks later...
  • 5 years later...
I've created a point that follows a rotational pattern. I can execute the feature itself by hitting F9, no problem.
When I try and run the whole program it skips that feature. Any idea why is that?
I have it set to run from features list.
I attached some screen shots of the feature and its settings. 6365_7d30e8a37ee93149e9849416dc6b94a4.pdf
×
×
  • Create New...