Intouch/Compact logix heartbeat problem

[farmerj8]

Hello,
I have an HMI running an Intouch 9.5 application. It is communicating with a compact logix PLC via an external I/O server. The Intouch application is reading the PLC seconds clock (OIT_Handshake_TimeOut)from the I/O server and then via a "while showing" window script (a header window that is always showing)performs the following; OIT_Handshake_TimeOut = OIT_Handshake_TimeIn;

The PLC logic compares the two tags looking for a value difference of 3 seconds or greater for a period of 10 seconds to latch the heartbeat alarm bit.

The symptom is an intermittent (several times per week)heartbeat alarm. Interestingly enough, I do not get I/O server status alarms. Looking at my trended tags, the "OIT_Handshake_TimeOut" value from PLC looks normal (counting up) but my "OIT_Handshake_TimeIn" value that is written back intermittently freezes randomly for various time periods. It stops incrementing from less than a minute to several hours with about 4 or 5 minutes being the average. I have never seen a windows script used in this fashion and was looking for any feedback as to why I might see this "stalling" in the
"OIT_Handshake_TimeIn" value. Thank you very much.

 

[daba]

Hello,
I have an HMI running an Intouch 9.5 application. It is communicating with a compact logix PLC via an external I/O server. The Intouch application is reading the PLC seconds clock (OIT_Handshake_TimeOut)from the I/O server and then via a "while showing" window script (a header window that is always showing)performs the following; OIT_Handshake_TimeOut = OIT_Handshake_TimeIn;

The PLC logic compares the two tags looking for a value difference of 3 seconds or greater for a period of 10 seconds to latch the heartbeat alarm bit.

The symptom is an intermittent (several times per week)heartbeat alarm. Interestingly enough, I do not get I/O server status alarms. Looking at my trended tags, the "OIT_Handshake_TimeOut" value from PLC looks normal (counting up) but my "OIT_Handshake_TimeIn" value that is written back intermittently freezes randomly for various time periods. It stops incrementing from less than a minute to several hours with about 4 or 5 minutes being the average. I have never seen a windows script used in this fashion and was looking for any feedback as to why I might see this "stalling" in the
"OIT_Handshake_TimeIn" value. Thank you very much.

An "over-the-top" approach to heart-beat monitoring IMHO.

Simplify the HMI script to just copying in IN BOOL tag, to an OUT BOOL tag...

symbolically that is...

  from PLC       to PLC

---| |------------( )---

get the PLC to invert the tags...

  from HMI       to HMI

---|/|------------( )---

The execution of this logic cause a self-propagating "bit-chase", which cycles at a rate determined by comms updates. If any bit in this logic "freezes" for too long, then comms (or script execution!) has failed

PLC uses "change-of-state" logic to reset a watchdog timer, the timer "done" setting an alarm which holds processes if necessary.

HMI uses similar to annunciate comms fails.

Note that the HMI might not display any comms alarm generated in the PLC if comms is down for any reason. That is why heartbeat monitoring should be done at both ends.

 

[farmerj8]

Thank you very much for taking the time to help and offer your input on how to simplify and improve this heartbeat task! I will take your advice and implement the changes soon and hopefully eleviate this intermittent nuisance alarm.