10-10-2016 09:33 AM
I have a sequecne that runs great stand alone now we are wanting to add another test socket to the system. It is not as easy as it sounds here are the conditions and what we are wanting to do. I need to have either 2 instances of the program running or to configure the sequence in batch or parallel but I'm a little confused on how to do it properly.
System Specifics:
My issues:
Thank you for any input you can give me.
Solved! Go to Solution.
10-11-2016
05:56 AM
- last edited on
10-21-2024
01:55 PM
by
Content Cleaner
Hi Bill,
i think with Teststand 2016 this should be possible, because this versions is supporting multiple instances of TestStand.
https://www.ni.com/docs/en-US/bundle/teststand/page/teststand-environments.html
They are calling this feature multiple environments.
Synronisation between this enviroments is possible according to the help.
But when i am looking to my TS2016 lock step type i only see that highest snyc level is the excution. Maybe this is forced by starting my instance NOT in multiple environments mode or the blue ones knows more about that.
To my shame i have not tried this great feature because i see a great benefit to run a test while you are editing another one.
Regards
Juergen
10-11-2016 08:46 PM
You can have multiple instances of TestStand running in pretty much any version.
Here are a couple options:
1- run two UIs and use and asterisk to prefix your lock names (e.g. "*DAQmxDevice"). This lock will then be shared across engines. In fact any of the synchronization steps can use the * to preceed the name to share it across engines.
2- Create a custom UI with 2 execution managers and then have 2 sets of execution views on the front panel. When they execute you can somehow tell the sequence file (using a UIMessage) which execution they fired off. Then it will know which set of UUTs it is testing. Use the * trick for your locks.
3- What you should be doing- use the parallel model. It handles everything in your requirements right out of the gate. In other words the user can run which socket they want from the parallel dialog. Use normal locking mechanisms. I have attached an example.
You will get the best performance and control from option 3. Just saying.
10-12-2016 06:30 AM
Thanks as always Jigwax!
In Parallel mode can the operator start the parts at different times, for example the next day when parts are already running in the other station?
10-12-2016 06:37 AM
Yes! Did you run my example? You will see the parallel dialog. You can start either socket at your own whim.
10-12-2016 06:49 AM
@~jiggawax~ wrote:Yes! Did you run my example? You will see the parallel dialog. You can start either socket at your own whim.
One last question I promise! 🙂
I have a custom UI that I have for one instance using the simple UI provided by NI with a UI message callback added. Can I modify it easliy for use in the parallel mode or is there another UI example that would be better? what will I need to add to the UI if I modify this one and how do I show both executions on one UI?
10-12-2016
07:23 AM
- last edited on
10-30-2024
07:10 PM
by
Content Cleaner
10-12-2016 09:26 PM
bill_lewis wrote:I have a custom UI that I have for one instance using the simple UI provided by NI with a UI message callback added. Can I modify it easliy for use in the parallel mode or is there another UI example that would be better? what will I need to add to the UI if I modify this one and how do I show both executions on one UI?
Bill,
I'm not sure what the UI you have does or what the UI Message you are handling does. If it were me I'd just sue the Full-Featured UI that ships with TestStand and then add your UI Message handling to it.
With a little more detail I could point you to a better solution.
Regards,