Our gas turbine frame 6 ,Mark-V control TMR application was shut down on 05.01.2011 at 21:00 PM for offline water wash.

The following sequence occurred at 07:48 AM at 06.01.2011 while water wash job was under progress

07:35:23:000 1 L1START CPB
07:35:23:062 1 L1X
07:35:23:062 1 L20FG1X
07:35:23:062 1 L20FGX
07:35:23:062 1 L3MCSTR
07:35:23:062 1 L4
07:35:24:000 0 L1START CPB
07:37:30:812 1 L28FDA
07:37:30:875 0 L28FDA
07:37:38:187 0 L14HR
07:37:38:187 1 L20TV1X
07:37:41:093 1 Q 0380
07:37:54:062 1 L14HT
07:38:08:093 1 Q 0400
07:38:33:312 1 L14HM
07:39:33:187 1 L2TV
07:46:37:312 1 L28FDA
07:46:37:375 0 L28FDA
07:48:24:812 0 L1X
07:48:24:812 0 L20FG1X
07:48:24:812 0 L20FGX
07:48:24:812 0 L2TV
07:48:24:812 0 L3MCSTR
07:48:24:812 0 L4
07:48:24:812 1 L4T
07:48:25:093 1 Q 0093
07:48:25:093 1 Q 0378
07:48:25:750 0 Q 0093

AT THIS JUNCTURE m/c tripped however the i/o comm alarm was coming since 31.12.2010 intermittently(thrice) & got reset immediately without tripping the m/c. Flame detector A is toggling so, this point can be nullified.We are not able to track why L4T appeared. Can communication failure be a cause for tripping? if so, how to rectify the same.

 

You need to tell us what what Process Alarms were annunciated prior to the start of the Off-Line Water Wash, and during the Off-line Water Wash.

For example, what are:

>07:37:41:093 1 Q 0380
>07:38:08:093 1 Q 0400
>07:48:25:093 1 Q 0093
>07:48:25:093 1 Q 0378

Without access to the Alarm List for your unit, we can't know what those specific alarm drops are annunciating.

The GE convention is that any condition which results in a Trip is supposed to have an alarm and alarm text message associated with it.

Without being able to examine the logic running in the Speedtronic it's very difficult for anyone to say exactly what the sequence is or should be. There are many variations of water wash sequencing. We don't know if there's some timer associated with the water washing procedure that stops the unit after some period of time.

It's possible that a Loss of I/O Comms Diagnostic Alarm can cause a turbine trip, but if you have updated PROMs that shouldn't happen.

Lastly, the SOE information you provided may have included inputs configured to log a change of state to alert the reader to a condition. For example, a low water was flow indication might not have been configured as an SOE so a change of state of that input wouldn't be logged.

We need to know what alarms, primarily Process Alarms in this case, were annunciated. Not just the Alarm Drop Numbers; we need to know the text messages to try to be of any help.