AS-i to Westinghouse Ovation Trouble! Help!

[HighPotential]

440 Megawatts of power dropped off line today during commission testing of a new power plant. The problem, using historical data, is largely suspected to be communication failure between an AS-i 2.1 gateway and a Westinghouse Ovation RLC module. At the start of the failure sequence both the gateway and the RLC module alarmed at the loss of communcation within one second of each other. Plantwide, we are using Bihl+Wiedermann AS-i 2.1 / Modbus gateways to connect each of several AS-i segments to the Westinghouse Ovation RLC modules. The connections are "1:1" using a striaght RS-232 cable connection carrrying the MODBUS intermediary protocol. The problem seems to not be limited to this segment and this time, it hit a critical segment. It is infrequent but has been encountered on virtually all of the other segements. Today for example, we had trouble on two segments about one hour apart.

We are also suspecting a Westinghouse RLC Module "SLEEP" or "QUIET" mode (they call it a feature) to have complicated the situation. Westinghouse has resisted disabling this loss of communcation "time-out".

Upon removing and reinstalling the Westinghouse RLC module, comms were NOT reestablished with the AS-i gateway until the gateway was itself cycled.

All cable lengths are less than 100ft.

I know its a very technical question but any suggestions would be greatly appreciated.

Thanks!

HP

 

[buzzp]

Just a thought not being real familiar with RS232.
I thought that 232 systems can only go 50 feet. I must be wrong or there is some retransmitting going on for something so easy to be overlooked.
Time out has caused issues on our RS485 systems as well. Still no real solution except to increase the "time out" time.

 

[HighPotential]

Yeah, sorry. The RS-232 length is only about five feet.

 

[lukner]

We've tested ASi segments with DeltaV and found that as long as the proper cable and power supplies / repeaters are used, it's an extremely reliable bus. However, use the wrong cable and power supplies (or lengths for that matter) and you see the kind of failures your observing. What kind of cable and power supplies are you using on the ASi bus? Is the cable you are using certified or has it otherwise been checked to meet specs? Have you checked the ASi signal on the cable?

Anyway, if you need more help, let me know.

Sincerely,
--Ralf

************************************
Ralf B. Lukner, Ph.D., P.E. (TX)
President
Momentum Technical Consulting, Inc.
PO Box 142906
Austin, TX 78714-2906

512-451-2863 main
512-451-1770 fax
lukner@momentum-inc.com

"Reliable Industrial Automation"
************************************

 

[gaharvey]

From a Power Genration Commissioning Engineer's Point of View:
During Power Plant operation, operation using non-current information can lead to disasterous consequences in relation to human life and equipment damage. Extending the timeout timer on communications involving control and in particular protection systems eg Burner Management, Turbine and Generator Protection systems should be detected immediately and action should not be delayed. The cause of the problem should be determined and corrected not masked by extending a timer. Are "Sleep" and "Quiet" modes appropriate for this type of plant? Can they be disabled without detriment to the Control System Performance? Depending upon the possible senarios ie the communications between which elements of the control system have been lost is the action taken appropriate in all possible circumstances?
From experience the flagged error might not be the real cause because error flags are usually responces to limited logic which assumes all other elements of the system are operating correctly at the time.

GAH

 

[HighPotential]

Thanks for all of the suggestions. At this point the contractor could not find a definitive cause of the trouble. Even though the AS-i segments only control "balance of plant" equipment (such as boiler feed pumps, condensate pumps, transformers), the contractor, to their credit has agreed to and in fact started, removing all of the AS-i system and replacing all control with a direct connection to the DCS.