BX-08AD-3 pulsating input signals

[Kristjan]

PLC: BX-DM1E-M-D
I/O: BX-16ND3 / BX-16TR / BX-08UT / 2 x BX-08AD-3 / BX-08DA-3

All BX-08AD-3 inputs are configured for 4-20mA. One of the analog signals on the second BX-08AD-3 is stable and correct at ca. 16mA for a while (8-24 hours) and then the signal falls to 6mA (incorrect reading) and shows short 12mA pulses every 4 seconds. Cycling the PLC to PROGRAM mode and back to RUN mode fixes the issue and we get correct readings for a while but then the signal fails again. See attached trend logs.

The input is reading from a DL24-00 level sensor and we are pretty confident about its configuration and installation. Based on the fact that RUN-PROGRAM-RUN transition fixes the issue, I suspect there is something wrong on the BX-08AD-3.

Is this a known issue with BX-08AD-3? Any hints about what could be going on?

[Bolt]

Can you move the channel in question to the other AD-3? What confuses me about your trend, is that the line spikes to past 16mA, drops to 4mA for a while, then jumps to before jumping back up to 16mA. When toggling to PROGRAM mode, it keeps reading/logging the input while not in RUN (at least for onboard AI/DI?) When did you toggle it? You could add DST10 to the graph other axis for a more complete overview or RUN/PROGRAM mode.

It's been years since I messed with it, but I had a DL14 once that I could not get stable reading out of either, it would be fine for a while, and then not. I don't remember the timing/details, and it was not in an ideal application either. I gave up and went to a submerged pressure transducer. Radar is also an option.

[Kristjan]

Thank you Bolt. There is a couple of pulses after I set the PLC to PROGRAM mode, then upon RUN mode transition I have a spike - zero - reading.

The input has now been stable for about 20 hours, I am crossing my fingers that it's OK now but somewhat uncomfortable not knowing why. If it fails again, I'll try moving to a different AI input and as a last resort changing to a submersible level sensor.

[Kristjan]

Just a quick update for future reference.

We moved the sensor to a different AI input and the sensor has been stable for over a week. We'll settle for this solution for now even though I am not convinced that the new input arrangement is the only relevant factor. It is a somewhat challenging installation in a very remote site. I was kinda hoping that the signal cable and a power cable (VFD to a deep well pump) were too close but unfortunately that is not the case, hence not a relevant factor.

Thanks again Bolt for your input.

[rlp122]

No idea if it might be the case or not, but the pulses used to be indicative of a bad common mode R/C circuit on that channel.