Recommended NI driver configuration for automatic tester

[sdavy]

Hi,

I am currently developing some automatic test scripts for some equipment we are currently repairing and would like to hear anyone's input as to a recommended software configuration for the hardware I am controlling. I am currently controlling the unit under test (at the moment power supplies) in addition to a DMM, Eload and in future a simple NI DIO device.

To give everyone a bit of background this task started off as a simple soft panel for GPIB control of one unit under test and has now grown to building test sequences in teststand, controlling the unit under test along with a DMM and an Eload from GPIB. We have recently purchased a newer DMM that naturally means changing some of the drivers and possibly small modifications to the teststand test scripts.

As you can see this little project has grown far beyond the origional scope and I would like to re-evaluate the curent test setup.

So far for any test equipment used, I have wrapped the available 488.2 and VISA drivers for the DMM and Eload and exported them to a DLL for teststand control. Our new DMM would naturally require a change to these drivers but could be done with fairly minimal modification. There is also an instance where an additional DMM (7.5 digit) may be required along with the standard DMM in use. I currently talk direct to the instrument, through the driver with no use of NI MAX.

The unit under test(s) also requires a driver written for it to be controlled via teststand and also for use with the soft panel that is also developed for local control / troubleshooting. Alot of the time these devices do not have a driver available, and are not SCPI instruments so the drivers must be written from scratch.

My question here is can anyone recommend an improved configuration to the one I have currently ended up with? I have looked at using IVI drivers but I am not sure how flexible this will be with instruments like the Eload for which there doesnt appear to be an Eload class or where I require quite specific control over the unit under test to verify all functionality is working correctly. I guess it is the partly the interchangability factor that I think would be benificial that can be found from using IVI drivers. Can anyone recommend a good scheme whereby I use something like the flexibility of NI MAX and an instrument reference "DMM1" (6.5 digit), "DMM2" (7.5 digit), "ELOAD1", "DIO1" to create something flexible in both instrument changeability and driver development and ease of implementation? There may be a different scheme between the instruments and the unit under test. Greater up front development is fine if the proceeding development of test scripts in teststand and/or soft panel in CVI are much quicker/improved.

I hope I managed to get my point across with this. I dont want to bore anyone with too much detail and want to keep my options open as far as possible.

[ColdenR]

Hello sdavy,

It sounds like you should be able to use IVI drivers for your DMMs and power supplies. Depending on the DMM and power supply you're using, the drivers might be available on ni.com/idnet, or you will have to code the specific version of the driver yourself. This document (http://digital.ni.com/manuals.nsf/websearch/2903634FD2A7EFD6862566B9007C4339) is a good resource for developing and using IVI drivers. I'm not sure if you've looked at that or not, but you should be able to achieve modularity through the scheme in that document.

Depending on the command set that the Eload device uses, you might be able to fit it into the one of the general schemes within the IVI drivers, or you might need to create your own driver. But you could always wrap all those drivers together using some functions that call the correct driver based on the device you're using.