parallelism of bore axes

[[Ja...]]

Hello,

I have a cylindrical housing with a smaller bore at the bottom. The small bore should be concentric with the housing cylinder datum.

The concentricity was off by 0.003", but we wanted to check the parallelism of the two bores also.

The CMM tech made a program to make a cylinder of the housing and a smaller cylinder of the bore. Then the axes of the two cylinders were compared for parallelism. The calypso report gave the result as 0.00007".

My question is, what is the 0.00007 dimension? How does that tell me how parallel the axes of the two bores are?

Is this number just a vector, with a value of zero meaning the two 3D lines are perfectly parallel?

[[Ow...]]

The bore could be 0.500 off center to the other bore and still be parallel to it.
How are you measuring the cylinders?
If measuring with a cylinder feature, make sure you are using at least 3 scan paths evenly spread throughout the length of each cylinder and make sure your filters are set correctly with enough points to satisfy the filter and use outlier removal of at least 3.
With out going into a lot of detail (short on time) I'd avoid using concentricity and use runout instead or measure both bores as cylinders (with 3 paths) and use coaxiality.

[[Ja...]]

Ok thank you for the pointers. We used two circle traces for each cylinder.

We also used coaxiality which gave a result of 0.0027", and we know the bores are off concentrically, but we specifically want to know how parallel the bores are seperately from the concentricity and coaxiality.

Do you know what calypso is reporting when computing parallelism of 3d lines?

[[Ri...]]

The result of parallelism is the attitude of each axis to each other.
They can be perfect or not, but it wont tell what's causing runout, etc.
Coaxiality will aid in results.

[[Er...]]

Another trick to see what is going on is to use report True Position with Axis End points turned on.

[[Jo...]]

You can project the bores onto a theo plane in each axis to get a "simple" 2d angle, not compound, to help troubleshoot.

[[Ja...]]

Please sign in to view this quote.

Thank you for the response. Can you give more detail about the attitude? I am not comprehending the value reported. How is the parallelism between two 3D lines defined by a single number? Is that number even a dimension in inches?

If the axes are very close to parallel, then that means our spindle and cutting tool are staying perpendicular to the table, and perhaps our work piece is vibrating laterally causing the concentricity to be off.

[[Da...]]

One of the cylinders is being used as the base and the other is being compared to it, this is why only one number is reported. I would suggest looking at the point cloud as a better way of seeing what is going on with these features.