Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations SDETERS on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

The result of TA from CETOL and CAD

Status
Not open for further replies.

chelseasea

Mechanical
Oct 25, 2023
1
Hello everyone,
This is my first time posting on forum, if any bother or unclear information, please bear with me.

I meet a question when using CETOL, picture is on attachment.
I set dimension as 11mm height, 10mm width and 45+/-1 degreee, the target is the gap between one plain and other plain impacting by the bevel.
The result of CETOL is different from CAD.
Does anyone know how to set the CETOL to let the result meeting the CAD? Or how could we explain the difference?

Many thanks.
 
 https://files.engineering.com/getfile.aspx?folder=28b2338d-827d-4aca-b81f-b665f42a4058&file=02.jpg
Replies continue below

Recommended for you

How CE-TOL works should be covered in the user manual/documentation that came with it. If not, ask Sigmetrix or your VAR for more details and see if the following is still correct.

It used to be that CE-TOL would use the CAD modeler to recalculate/regenerate the model by varying one dimension at a time to determine the ratio of the change in each dimension to the change in the measured result. Since angles aren't perfectly linear it would produce a slight difference over their direct combination.

It would not do combinations as those could easily number to the millions for complicated assembly models and they distinguish themselves as being fast. 20 dimensions gets to 1 million combinations. 32 dimensions is 4 Billion.

I see the following statement:

Changes to tolerances or input statistical distributions immediately update the results.

from
By having individual sensitivities they don't need to rerun the analysis, just add and subtract the distributions, but it also means interactions won't always be correctly accounted for.

You are getting 3 significant digits in the result which is as good or better than Quality Assurance will be able to extract from actual measurements. There won't be a setting to get an exact match because CE-TOL never evaluates it the way you have in the model.

One area that it used to not work right is that, because it depended on the model,it could not account for a change in parallelism due to thickness tolerance.

The main competitor is VSA and a VSA offshoot that creates a model evaluation that is compiled and can run millions of variations rapidly in a manner that duplicates manufacturing - it creates transformation matricies, the sort that can be done by the hundreds of millions per second. It could handle changes, such as form tolerance variations, that produce orientation effects. However, it is not typically licensed as a module available licensed through CAD software companies.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor