Best Practices to prepare CAD model for CAM 3

[sluzzer]

Are there any best practices to follow regarding CAD model preparation for CAM?

Nowadays, most of the CAM are feature based CAM, which is going to use the CAD model for tool path generation. Hence, the accuracy of CAD model affects output of CAM.

So regarding this, I have the following questions:
1. What are all the best practices?
2. Suppose if my part tolerance as per engineering drawing is 50 microns, what should be my model tolerances?
3. Should I increase the no. of poles / control points using the X-form or I-form to increase the accuracy of surfaces? (Note: Assume the model we get is in unparametric form)
4. If we use different CAM package (other than NX CAM), then what is the best neutral data format(IGES or STEP or parasolid) to have minimum data loss? Is there any setting which we should take care while exporting the model?

Any ideas are much appreciated. Thanks.

 

[jerry1423]

I always check the solid model for errors before passing it on for CAM.

MENU -> ANALYSIS -> EXAMINE GEOMETRY

If any of the seven Checks within the red rectangle give a negative result they need to be fixed.
The other four Checks depend on the application.

Jerry J.
UGV5-NX11

 

[Tingsryd]

Perhaps drop the default OOTB modeling tolerances a little. I had a manufacturing tolerance issue a while back and GTAC determined the modeling tolerances were at fault. I seem to recall the variance was in the .0002"-.0004" range. On the large end that is approximately 25% of your 50 micron tolerance so perhaps using slightly smaller modeling tolerances could help.

I second Jerry's recommendation for Examine Geometry which will give you a good idea as to model quality although you could go directly to using Optimize Face in Synchronous Modeling to clean things up. File-Export-HealGeometry can fix a lot of anomalies too. Not sure about other cam software but NX seems to be very forgiving with dirty models with only very rare problems rising from anomalies. If your models are complicated then I recommend cleaning them up as part of your prep workflow.

I haven't heard that increasing surface poles/control points accuracy impacting cam. It really comes down to what you require for model accuracy; not so much for machining; it will give you what the model is.

As for data format obviously native is best; feel free to delete model parameters for IP considerations. Parasolid is a very commonly used format and STEP behind that and lastly IGES, IMO. Some software have issues with IGES. If you don't mind creating multiple files I would make a Parasolid file and a STEP file. Another consideration is the Parasolid version. You could run into a rare version issue if someone is using an older cam software. I wouldn't worry too much about it; most companies keep their software up to date but if you want you can easily output Parasolids a couple versions old.






NX 12.0.2
Testing series 1872
EVP's