Batch Run For Repeatability Study

Advanced techniques applicable to any version of CMM-Manager
Post Reply
JNorton
Regular User
Regular User
Posts: 19
Joined: Wed Jul 25, 2018 9:40 am

Batch Run For Repeatability Study

Post by JNorton » Wed May 22, 2019 4:57 pm

I am having some troubles with a batch run on one of my programs. It's not an issue of getting the program to run repetitively it's an issue of the features repeating on a reporting level. I have set up and ran the same part on a 10 cycle batch 5 separate times in the last two weeks and each time I can get 16 of 20 dimensions to repeat within 5% of each other(difference of min/max deviation over total tolerance). The remaining four features repeat within 16%-40%.

The part does not move between runs, my alignments are solid and I use a Recall Alignment to direct the machine to the part every run. I have verified that the points are programmed where they need to be on the cad model and I have features that repeat within 5% indicating that.

One of the strange things I have noticed is a diameter measuring the Root Diameter on the OD of the gear measures lager and larger by a few tenths on each run creating an upward trend, the same holds true for a Surface Profile on one end of the part, yet an identical Surface Profile on the other end of the part repeats within 5% and does not deviate more than .0003 from run to run. I've looked at each feature and I am not experiencing shanking or seeing anything that would indicate probe deflection. The part has soaked to 70° and that temp only varies ±1° throughout the day.

I am at a loss of what to do to get my program to repeat on these runs.

Any help would be greatly appreciated.

User avatar
CrashN8
Site Admin
Site Admin
Posts: 367
Joined: Thu Feb 09, 2017 4:07 pm

Re: Batch Run For Repeatability Study

Post by CrashN8 » Wed May 22, 2019 5:05 pm

What kind of system? DCC CMM, Manual CMM, Vision system, what controller, what probing, etc.

Does it repeat if you run the program individually or is this truly isolated to Batch Run?

Can you send the program and results to Help Desk?

JNorton
Regular User
Regular User
Posts: 19
Joined: Wed Jul 25, 2018 9:40 am

Re: Batch Run For Repeatability Study

Post by JNorton » Thu May 23, 2019 11:54 am

3.8 SP1 on Windows 10 running a Mitutoyo Bright-A916 with a Renishaw Controler. The program was repeating on an individual run basis but not on the batch run. After my initial post I found a very small error in my alignment rotation that for some reason was stacking from run to run. After fixing that error and doing another 10 run batch everything repeats within 3% from run to run. It's confusing because I remeasure the alignment on each run so for the software to stack the error is strange. The only thing I can think of is that the error was transferring on the initial Recall Alignment at the beginning of each cycle. Just a thought though as I have no way to prove it.

User avatar
CrashN8
Site Admin
Site Admin
Posts: 367
Joined: Thu Feb 09, 2017 4:07 pm

Re: Batch Run For Repeatability Study

Post by CrashN8 » Thu May 23, 2019 12:04 pm

If you save and then recall the SAME Fixture Alignment then, yes I would expect the behavior that you describe. Fixture Ref Frame should be reserved for Recall at beginning of program only. Also, note you can skip initial steps of program for Batch Run if needed.

JNorton
Regular User
Regular User
Posts: 19
Joined: Wed Jul 25, 2018 9:40 am

Re: Batch Run For Repeatability Study

Post by JNorton » Fri May 24, 2019 11:27 am

Just to clarify. I should be using the Fixture Alignment as follows.

1. Run through the initial alignment steps and save & fixture that alignment.
2. Delete/Disable the Save Alignment line for said alignment.
3. Recall said alignment at beginning of program allowing me to skip the steps used to create that alignment.

If this is correct then, as long as the part doesn't move, my alignment will transfer correctly from run to run within the Batch.

Do I have this right?

User avatar
medupriest
Moderator
Moderator
Posts: 408
Joined: Fri Mar 13, 2015 11:59 am

Re: Batch Run For Repeatability Study

Post by medupriest » Fri May 24, 2019 2:38 pm

That all sounds correct.

JNorton
Regular User
Regular User
Posts: 19
Joined: Wed Jul 25, 2018 9:40 am

Re: Batch Run For Repeatability Study

Post by JNorton » Fri May 24, 2019 2:52 pm

Thanks!

Post Reply