cRIO AI issue

[ZoltanMTS]

Hi,

We have a strange issue during installation.

We have 2 cRIOs (9065 and 9067), which we checked out earlier and everything worked fine.
When we started to install them, we have an issue with the analogue input signals (9205 module) on one of them.
Various sensors are connected to the module and their voltages are not correct in the cRIO. (We measured the voltages on the pins also.) In fact, what we observed is that, when we connect the module, it shows correct values and then it drops to a lower value. 2.5V drops to almost 0V, 10V drops to around 7.5V, which suggests that, we have a 2.5 V offset.

Then we move the module to the another cRIO, where it worked fine. I also have the same setup on my desk, and they work fine. They setup for the AI modules are identical, calibrated +/- 10 differential voltage.

The AIs are handled in the FPGA (not in Scan mode) and we checked it using the development environment, and the wrong voltages are at the output of the FPGA block, which points to setup issue, but again, this module worked earlier, works in the another cRIO and still works on my desk.

As we also suspect short or ground loop, we have removed all other modules, except the AI module and then we powered it from the supply, which we used for the another cRIO, but we still have the same voltage drop.

I read about he COM pin, which should be grounded, but as the module works in one cRIO and not in the another, I do not expect that would be the cause.

As in one of the modules, we have a strain gauge module also and we learnt that, if it is handled by the FPGA (not in Scan mode), then we have to read out calibration data and calculate the engineering value from the incoming numerical value using the calibration data.
As the ‘calibrated +/- 10 differential voltage’ setting works on one of the cRIOs, I am not sure if this may be the issue on the another, but it seems something definitely has changed.

Currently, we are getting a new cRIO and try with that, as this module may developed problems (it travelled across the world), but any other suggestions would be welcomed, as we are running out of ideas.

Thanks,

Zoltan

[MitchBrown · Accepted solution]

Hey Zoltan,

If I were looking at this I would be looking closely at the cRIO which it is not working in. You have followed a good troubleshooting process, and from what you have said I would suggest the following:

1. Test the AI module in several slots in each cRIO, if it consistently doesn't work in one cRIO, and works in the other, then I would think the issue is with the cRIO itself.

If it is faulty in only one slot, then I would suspect an issue with the backplane or connection from it to the module.

Provided the module works in at least one slot, that would rule the module out as being at fault to me.

2. If you have an AO or DO module I would try performing loopback tests on the module, on different channels as well, as this will identify whether specific channels are affected, or whether the entire module is. Importantly, do connect all terminals up as stated in the manual, even if it has appeared to work without them (such as the COM pin).

If you do find any channels are not measuring correctly, then there may be an issue with the module.

If you do believe there is an issue with either piece of kit, please do reach out to your local technical support team who can help go through more troubleshooting steps, as well as help arrange a repair if needed. You can do this at ni.com/support.

Hopefully this helps!

Mitch

[ZoltanMTS]

The solution was the COM pin.

The lesson one can learn from this is that, with standard troubleshooting steps, you cannot eliminate this issue. One chassis works fine without the COM connected and the second worked earlier and as I wrote, moving to the another chassis it works fine.

Zoltan