Ge 9fa Unit During Running Got Tripped Due To Mark Vi Chanel R And T Failure


Thread Starter


In my plant GE 9fa unit during running got tripped due to Mark VI channel R and R failure (2 out of 3) also in HMI Mark VI rack status page r&t i/o state rack status was showing "0" red color instead of 106 (green color).

Observation after unit tripped.

1. All drives / motors got started except 88hq 1&2 with lift oil, (lube oil aux p/ps 1,2 emergency lube oil p/p, emergency seal oil p/p were running). Hence same time unit RPM was 3000 with decreasing state without lift oil.

2.In HMI Mark VI rack status page r&t i/o state rack status was showing "0" red color instead of 106 (green color). Hence Pos volt 60.4 vdc neg volt -57.7, diff volt- 2.7vdc ( normal). In alarm page diagnostic alarm slot 7 VTUR was persiting in same channel

3. After rebooting of the Mark VI channel r & t (controller) from PDM all drives got normal with hydraulic p/ps, lift oil, turning gear motor, unit reached to 6 rpm.

Could you please describe what could be the reasons for the failure of Mark VI channel r & t (controller)?

Note: Same unit 2 times tripped by same reason in a week.
A single Diagnostic Alarm WILL NOT result in a turbine trip (on a TMR system). BUT, combinations of Diagnostic Alarms CAN and WILL result in turbine trips. That's why it's important to understand and resolve every Diagnostic Alarm--because combinations of Diagnostic Alarms CAN and WILL result in turbine trips.

And, because there is NO chart or documentation about which combinations of Diagnostic Alarms will result in a turbine trip, it's an excellent idea to resolve, or at least understand, EVERY Diagnostic Alarm--as soon as it is annunciated. They are NOT nuisance alarms; they indicate problems with the health of the turbine control system (hardware and/or software).

And the VTUR card is one of the most important cards for the control and protection of the turbine. It controls the primary trip relays for the fuel stop solenoids, so if it's not working and is having issues then it's likely there are going to be other issues.

You have NOT provided enough information about the VTUR Diagnostic Alarm(s), nor about the conditions surrounding the event. It would SEEM from the DC voltage splits you described there is no 125 VDC Battery Ground problem--but we don't know if that's continuous measurement or one-time measurement.

As for all the "drives" coming on when two of three control processors are lost--that's normal. GE uses a 'drop-out-to-run' philosophy for all critical motors and when two of three control processors don't say they should be off, they run (to protect the unit). So, instead of all the critical motors shutting off when there's a serious problem with the Mark*, they all start and run--to protect the unit. Better they start and run when they don't need to be running than they stop or don't start when they need to be running. That's the reason for the drop-out-to-run philosophy.

I don't know what else to say, based on the little useful information provided. Except to say that just because a unit can be started doesn't mean it should be started. If there are Diagnostic Alarms present, and Process Alarms, prior to or during a START, they should be investigated and resolved, or at least understood. Understanding Diagnostic Alarms, again, is key to reliability and availability.

There was a TIL (Technical Information Letter) many years ago about problem processor rack power supplies. We don't know how old the Mark VI is at your site, nor if any or all of the TILs have been implemented. We also don't know the condition of the batteries or the battery charger(s) powering the control processor power supplies. Battery output capacitors do fail and cause large spikes and chops in the output to the battery, which can affect the control processor power supplies. AND, if there is a <DACA> (AC-DC converter) providing back-up to the battery AND if the power input to the <DACA> is not clean (if it has spikes or dips or is excessively choppy, which can happen from some inverters/UPSs) the <DACA> will make the problems WORSE on the 125 VDC output.

Finally, what is the condition of the coaxial cables which interconnect all of the control processors and the <P> processors--the IONET? The BNC connectors used for the IONET cables can get corroded and cause intermittent communication problems. They should be regularly inspected and cleaned, if necessary. And, that goes for just general housekeeping in the Mark* control panels. Dust and dirt, along with humidity, are BAD. As is high compartment temperatures. And, LOW compartment temperatures in a humid environment is VERY bad.

Again, there's simply not enough information to be of much help. And, just because a unit can be started doesn't mean it should be. Tripping F-class turbines is NOT good for the turbine. NOT good.

Please write back to let us know how you fare in resolving the problem(s). But, be sure to include ALL Diagnostic- and Process Alarms when you do if you want additional help. They are likely the best indicator(s) of what's going on, even if they just seem like nuisances. And tell us if there are storms in the area, electrical storms, and what the condition of the grid is when the trip occurs (stable; unstable; etc.).