GT Frame-V machine runs with MARK-Vie control system. Since last 2 Years, SOE alarm appears in GT Running condition for 1 ms and reset. Sometimes trip input SOE alarms appears in running machine variable like: L4ct and we have physically looped on card terminal end.

We are facing different type of SOE alarms frequently in GT Running and stop condition.

Please advice why its appearing and what will happened in GT running condition if SOE appears frequently.

Is it safe for runs machine on SOE appears?

 

Check your dc supply voltage splits, should be plus minus 65, and make sure no ac component is present. Never seen this on our mkvie but have seen it plenty of times on mkv.

 

GT DC voltage checked and found plus minus 62. No AC voltage found.

Please advice further..

> Check your dc supply voltage splits, should be plus minus 65, and make sure
> no ac component is present. Never seen this on our mkvie but have seen it plenty of times on mkv.

 

Vimal Gupta,

When did this problem start? Has it been ongoing since commissioning of the Mark VIe? Or has it just gotten progressively worse over time, since a lightning-induced trip, or some other event (excessive rains, etc.)?

Exactly how was the AC voltage checked? What kind of device was used to check for AC "ripple" on the DC supply? The method/device may not be capable of accurately measuring AC ripple.

Can you associate the SOEs with any particular event? Such as, the energizing of 20CB-1, for example? Or the closure of the generator breaker during synchronizing?

Was this this Mark VIe provided new with a new turbine, or is a retrofit/upgrade of an older control system on an existing turbine? If it's a retrofit, are you sure the panel is properly earthed/grounded for the earthing/grounding system at your site? The Mark VIe is capable of two separate grounds--detailed in the Mark VIe System Guide, GEH-6721: a functional- and a protective earth. If both earths are available in a plant, then it's very common for other equipments/systems to be improperly connected to the wrong earth system and cause problems.

If the Mark VIe was an upgrade/retrofit, it's very common for older solenoids to cause problems when energized/de-energized because of high inductive kicks. The supplier/packager of the Mark VIe should have provided suitable "snubbers" for the older solenoids.

What Diagnostic Alarms are being annunciated by the various I/O Packs in the Mark VIe control panel?

 

We have checked AC voltage by simple putting multimeter on DC supply voltage. But no AC Found.

We have upgraded MARK-IV system with MARK-Vie with old turbine. Problem started since last 2 years after 6 month of MARK-Vie commissioning.

After 2-3 days GT running normally on full load, SOE starts.
We have provided two individually earth at system and panel. Earthing resistance was checked and found 0.2/0.3 Ohm.

SOE alarm list:-
1. L33TD-1; Liquid valve transfer in distillate position (It was forced from toolbox)

2. L26QT1A;

3. L63EA1

4. L26BT1

IF ANY FURTHER INFORMATION REQUIRED,PLEASE LET ME KNOW. PLEASE FURTHER ADVICE.

 

Vimal Gupta,

Yes; more information is required.

Usually, it's necessary to ground one lead of the multimeter (set to AC) and then measure each of the DC leads with respect to ground. Only when the AC voltage is excessive will an AC voltage measurement from the positive DC lead to the negative DC lead reveal any AC ripple. And, it's best to use a True AC RMS multimeter when checking for AC ripple. An oscilloscope can also be used to check for AC ripple on the DC supply in a similar fashion.

When you say there are two earths/grounds, it's not really clear what you mean. Do you mean that the control panel is solidly grounded in two places? If this was a Mark IV Migration (where the existing Mark IV control panel enclosure was gutted and Mark VIe modules were then installed in the panel, there should have been a copper ground bus bar in the bottom of the Mark IV panel which should have already had at least one suitable ground conductor already connected to it--and if not, one should have been added between the ground bus bar and the station earth/ground grid. Grounding is very important.

Grounding is extremely important to any control system, including Speedtronic Mark* turbine control panels.

If the Mark VIe is a completely new control panel that replaced the existing Mark IV control panel then during installation/commissioning the earthing/grounding should have been reviewed and properly configured to match the existing installation. Most power plants of Mark IV-vintage did not have two separate earth/ground grids; they only had one and all earths (control system AND protective) were all connected to earth/ground via the same ground grid. If the plant only has one ground grid, then there would have been a flexible, braided lead connecting the functional (control system) earth and the protective earth <i>of the Mark VI control panel</i> which should have been left in place--because there is only one earth/ground grid. Only if the plant has two separate earth/ground grids (one usually dedicated to control systems--the functional earth--and the second for personnel and equipment protection) should that flexible, braided strap have been removed.

You did not answer the question about Diagnostic Alarms. What Diagnostic Alarms are being annunciated by the various I/O Packs?

The signals you listed--are they all connected to the same discrete input terminal board? If they are--and there are Diagnostic Alarms being annunciated by the I/O Pack(s) on that terminal board, then there may be a problem with that terminal board and/or one (or more) of the I/O Pack(s) on that terminal board.

A lot of problems similar to this one are related to poor construction practices--in this case, many years ago when the plant was constructed. If cables are not properly segregated from each other to prevent induced electrical noise from being generated in control cables/circuits this problem can occur. Wires and conductors can act like capacitors, and take a while to "charge" before problems begin, so, problems seem to be very intermittent. But the root cause is almost always improper cable segregation, which is very difficult to "undo" decades after the plant was constructed.

I have seen wiring problems experienced during commissioning weaken and cause eventual failures of I/O Packs and input/output terminal boards, also.

Another known source of AC ripple (which is another way to describe induced electrical noise on control cables/conductors) is from old battery charger output filter capacitors. Capacitors need maintenance and replacement, too.

Do you use SOEs in monitoring the operation of your plant? I have seen ALL discrete inputs have their SOE set to "enable" from the factory, even ones which are unused--which is just insane because it just causes so many events (not alarms--events) to be logged to the Alarm Logger, wasting printer ribbon and paper. You could have someone go through the discrete inputs and choose only the most important ones, and disable all the others.

Troubleshooting and resolving this problem can be very frustrating--especially if poor construction practices were used during the original construction of the plant. Mark IVs weren't so susceptible to induced electrical noise on discrete inputs; Mark Vs were VERY affected, and to a certain degree so are Mark VIs and Mark VIes. Patience and a thorough, logical troubleshooting process will eventually resolve the problem--but it may also include re-routing some cables/conductors.

As for the answer to your question about what will happen if the SOEs toggle (also called "dither") when the turbine is running, well, that's pretty hard to predict, unfortunately. Most control systems "read" their inputs periodically, and then use those values in the execution of the program (the application code in the case of the Mark VIe). Then they write to their outputs, have a little idle time, then read start over by reading the inputs again, executing the program, etc. So, if the input is determined to be in an alarm/trip state when it is read by the control system, it may initiate a trip. I don't know how long the Mark VIe reads its inputs (if it's more than several milliseconds to try to prevent a nuisance toggle/dither from initiating an alarm/trip--but it seems to). I'm sure the length of time the Mark VIe reads its inputs is NOT adjustable, so that would not be an option. The fact that you say the SOE toggles/dithers in a millisecond and the turbine has not experienced nuisance Process Alarms (an SOE is not a Process Alarm) or trip seems to suggest that the Mark VIe is somewhat immune from extremely fast dithering/toggling of discrete input states. But, that's simply an educated guess on my part--based on experience and the information you provided. Only GE or the packager of its products could really answer that question with any degree of certainty--and the answer would definitely have to come from the factory where the Mark VIe is designed and the I/O Packs are programmed (the "firmware").

Please write back to let us know how you fare in resolving this problem, or if you can provide the requested information we might be able to provide more help or direction.

 

We will inform AC ripple voltage in next reply.

We have 2 No earth grids and connected at MARK-Vie as below:

1. Connected at bottom below copper bus bar that is functional earth.
2. Connected at bottom below panel body earthing.

There will be no diagnostic alarms appearing during SOE appearing by various I/O pack. These SOE are connected with 4 different no of TDBT card.

Yes, we are using SOE for monitoring in system by "enable" in tool box only for used variable.

All unused variable is already "disable".

Even L33TD was forced in software but it's alarm was also appeared.

We have contacted to BGGTS to check problem and they suggested the field cable problem. We have replaced the 63EA-1 and 26BT-1,63QT-1A cable replaced from switch to MARK-Vie panel. After 1-year, no alarm appeared of these inputs. But recently these was also re-appeared.

One more thing we want to share with you. During commissioning of MARK-Vie some pack was sparked due to poor grounding. we have called GE to resolving the problem of sparked the system. GE replaced the damaged card and healthiness made of panel. GE was given clearance to panel commissioning.

After successfully commissioning of MARK-Vie, after 1 year SOE alarms problem appeared and after that continue coming.

IF ANY FURTHER INFORMATION REQUIRED, PLEASE ADVICE