Modeling Helical Gears for CMM 1

[AReed]

This is bit of a solid modeling or inspection question, which way do you model gears/worms for the most accurate output? We are outputting gears for CMM inspection programs and I am a little confused in my typical modeling style as to which way would produce a more accurate gear profile for inspection. Do you model the gear profile parallel/perpendicular with the cylinder axis or do you use a plane to draw the profile perpendicular to your line of helix before using a cut-sweep to remove material from the main body.

I'm having a brain-fart or logical argument issue either way, any insight would be helpful.

Thanks

 

[drawoh]

AReed,

Are CMMs accurate enough for gear inspection?

--
JHG

 

[BrianE22]

The tooth profile is defined on a plane tilted at the lead angle.

 

[spigor]

You might want to take a look at this:

http://www.gearsolutions.com/article/detail/5400/tooth-tips-william-crosher

 

[SwinnyGG]

@drawoh:

I don't know what specific equipment OP has access to, but we have a Nikon system that is accurate to 1.5 micron.

I would imagine that for measuring gear teeth you would want a non-contact system, which could get you to 5 micron or less.

 

[geesamand]

There are two questions here:
1) Per the gearing standards. Or you can buy a gear inspection software or gear modeling software and get going now.
2) Yes, CMMs are capable of measuring gears, with some limitations. We measure up to AGMA 12 quality gears on our full sized CMM. The gears are much smaller than the CMM table, so accumulated error over long movements is not a factor. It takes all of the available accuracy of the machine to do this. We confirm this with a master gear artifact that has a NIST certified inspection report.

Our system is driven by Zeiss Calypso and we use the GearPro software and rotary table options for gearing.

What AGMA quality are you attempting to measure? How large is the workpiece? What CMM do you have?

 

[tbuelna]

If you are comparing the CMM results from the finished gear against a digital CAD surface model, then it is important that the CAD surface model was created with a sufficient level of fidelity. If you don't have experience constructing very high fidelity CAD surface models, then you'd be better off using a software application designed to create the gear flank surface model for you.

 

[mfgenggear]

the software in CNC gear checkers like the Gleason or the Hofler are programed to to verify per the standards, AGMA & ISO, and can verify master gears. on the old AGMA 2000 it is a class 15. and both machines are calibrated to the NIST. that said depending on the critical nature of these parts. I think it would be be better to subcontract out the inspection of these gears. if this is developement and some how is reasearch. then OK. depending on the class of gears required, and application. this is not for the newbie to try and develop with out the proper knowledge. what is the requirements?