I've been chasing an issue at one site for a while and haven't had much success myself or with GSC (yet). I'm wondering if anyone else has seen this.
We have a handful of DI points where the DI block will periodically generate a bunch of "Input Failure/Bad PV" errors. We are alarming on this (DI_BAD_ALM), so it generates alarms to the operator. Sometimes we'll see this happen a couple times and then go away, other times it will happen 3-4 times per minute for an hour or more and then go back to normal. We'll go days without any alarms, and then it will come back. After a few weeks, the errors will start showing up on a different DI point on the same CIOCs, and the previously alarming point will be fine.
This is only happening on CIOCs where the scan time is set to 50 ms. We have other CIOCs at 250 ms and don't see this issue. We have 2 CIOCs set to 50 ms and it is happening on DI points on both of them. These CIOCs are in different buildings and assigned to different controllers and both exhibit the same problem.
The scan rates of the offending modules is 500ms or 1s. I tried setting a few to 200ms just for fun and still had the same issue.
The CHARM bus statistics indicate 0 for 'Charm No Comms', 'Acyc No Response', 'Acyc Invalid Response', 'Charm Retries', and 'Failed Send'.
This site previously had ACN communication issues (certify your cables people), but in the last 40 days the relevant controllers, CIOCs and switches show only a single CRC error.
We are not getting errors on the other CHARMs on the CIOCs (e.g. AI).
All of the offending points are using CHARM type '24 VDC Low Side'.
Linefault detection is not enabled.
The field devices are simple level / flow switches.
We opened a call with GSC in September but haven't made any progress toward a resolution. It's now being escalated to product engineering.
CIOC and controllers software version is 13.3.1.6375.xr
Has anyone else run into this before? Any suggestions?