MKV Simplex <C> Core Point Toggle

On the MKV simplex a logic point which is also a <C> contact output is written to by a normally closed bigblock output. The bigblock is also in <C>. Modbus reports this contact output point changing state fleetingly, however the contact output does not physically change state. The logic can be forced and behaves normally. The toggle does not happen all the time, and seems to occur at random times, unrelated to machine operation or activity on the station. The logic can set and reset every second or so when it does toggle. The Modbus has been checked thoroughly and operates correctly on all other points. There are no LCC or communications errors.

Any thoughts or pointers to help resolve this would be appreciated.

SIMPLEX panels do seem to have some idiosyncrasies not shared by their TMR brethren.

When did this problem start? What happened before the problem started?

My guess is, based on my understanding of your post details (and I'm not sure I quite understand exactly what is going on) that there's some issue with the way the operator interface (<I> or GE Mark V HMI--not sure which you're using) is handling the data.

You say the relay output (contact output) isn't physically changing state. Have you tried using the Logic Forcing display to see if the logic value is changing state when MODBUS is reporting that it is changing state but the actual relay on the TCRA card is not changing state? Just observe the value to see what is being reported; don't force it.

You could also use VIEW2 to capture values for the point to see if <C> is toggling the signal, or, if as I suspect, <C> is stable but something is amiss with the MODBUS ends' handling of the signal. This is probably the best way to troubleshoot whether or not the issue is in the Mark V, or in the operator interface or the other end of the MODBUS link (DCS or ???). The fastest data rate for VIEW2 is 32 Hz (32 times per second), but, again, it's probably going to capture any changes in what <C> thinks the values are, without any delay caused by the monitor/video card on the operator interface.

It would be helpful if you would tell us what the signal name is, and what the output does. (It would be helpful to me, at least...)

My guess is there's something amiss with either the MODBUS definition file on the HMI, or something amiss with the MODBUS definition file at the other end of the MODBUS link.

You could try changing the assignment of the signal to a different relay (in IO.ASG). That would require compiling the CSP and downloading to <C> (and <R>), as well as re-starting the operator interface. You would also have to move the wires associated with the output. I don't think it would require a change to the MODBUS definition file(s).

Again, my limited experience with these kinds of MODBUS issues is that they are related to problems with MODBUS definition files (for example, the same signal name assigned to to different registers/coils in one or more MODBUS definition files).

Please write back to let us know how you fare.