Hi,
Thank you in advance.

I am facing an issue MARK -IV Control system. During electrical breaker modification work time, we shut down the Mark-IV panel for 1 month after we restarted MARK-IV.

During GT trail startup, the gas turbine was tripped due to 20FG (Gas trip oil solenoid valve) is not energized. Then I checked the solenoid valve and the field side was found ok. Again started Checking the L20FGX1=1 in software but 20FGX1 relay found not energized. then again tripped.

20FGX1 relay to energize we need master protective relay is must active 4R, 4S, 4T, 4X-1, 4X-2, 4X-3, 4X-4, 4X-5 should energize then only 20FGX1 get power P28VR-4 for the relay coil. then the following list is checked.

1. All power supplies are healthy,
2. Relay power supply module P28VR is healthy
3. All RST and C core are running healthy.
4. Diagnostic alarm checked and found no alarm.
5. Checked all Fuses in the Power distribution module and found ok.
6. After starting, I checked in software L4X1=1 is active, But relay 4R, 4S, and 4T are not energized.
7. Checked 45FTX=0 (fire fighting relay) - output NC contact is getting 28VDC.
8. 4R, 4S, 4T - I Changed the new relays and tested, found the relay ok.
9. 4R, 4S, 4T - relay coil terminal 11 is getting 28VDC and 10 terminal connected to Relay drive module <RDM>
10. After shutdown GT, Then I forced L4X1=1 into the software, then I checked relay 4R, 4S, and 4T is not energized.
11. After shutdown GT, then I force L4X1=1 in software showing all <R>, <S>, <T> core showing value 1 but relay 4R, 4S, 4T not energies.
12. Then I suspected the relay driver module. <RDM> may have an issue, then HRDB Slot 'A' module interchanged between HRDB Slot 'B" after I tried it's not working and the same issue.
13. I forced L20FGX1=1 only, then I manually pressed relay by hand 4R, 4S, and 4T during that time 4X-1, 4X-2, 4X-3, 4X-4, 4X-5, after that 20FGX1 relay is active.
14. I completely restarted 2 times of MARK-IV, but the problem still exists.

Note: during relay testing time I completely isolated pumps and starting device etc..

At last, the main master protective relay 4R, 4S, and 4T are not active after the start command given, Kindly give me a suggestion to solve the problem.

Main master protection page attached for reference.

 

I'm not to familiar with the 28Vdc on the Mark IV. I assume the P28VR is the return of the 28 Vdc and it is ungrounded. I'm not sure what you are using as a reference on your measurement or if your measurement are positive or negative 28 volts. Just be carefull that you don't fool yourself into thinking you have the proper voltage when what you are reading is the opposite polarity coming from a closed path. My concern is that you have a remote permissive contact and the ESD button on the OPM which could be opening the return path. If you have the proper voltage at the 45FTX N/C contact, you should be good but not sure what polarity you are measuring. I know that's pretty basic but in the "heat of the battle" its an easy mistake to make. Good Luck.

 

I have spent a couple of hours looking over my old Mark IV prints (we upgraded to Mark VI many years ago so I dont have ant hardware to look at). I now understand that P28VR is the redundant source of 28 volt power. Based on item 9 on your list you dont appear to have a problem on the positive side (unless there was a high resistance contact that drops the voltage when load is applied) so the Relay Driver Module would be a good suspect as you said. Apparently the RDM switches the PCOM to pull in the relay. My problem is that I cant find in my documentation where the PCOM is connected to the RDM. Since the 28 volt source is redundant, there must be a path from each supply to the RDM. Good luck and please let us know what you find.

 

Mr. MWO.
Thank you for your sharing you knowledge.
P28VR is a redundant source of 28V, There i mentioned 28VDC is a positive checked with Ground. We assumed the PCOM getting from RDM (Relay driver module) is correct ( i checked with another working relay PCOM getting from RDM). You may see the diagram which i have attached. RDM power supply from P28V1 and P28V2. Its all from Power supply redundant module <PSREL1>, <PSREL2>.

 

ONE of the very good things about GE Mark* IV and Mark* V turbine control systems is that COM is ground is earth. That goes for PCOM, DCOM, COM, ACOM, earth, ground--they're all the same. NOT LIKE SOME control systems that have one or three or more grounds they reference their control system voltages to. So, when measuring DC voltages (5 VDC, +/+12 VDC, 28 VDC, 24 VDC, 125 VDC) attach one lead of the meter to a proper earth/ground and you can measure any voltage with respect to ground and get a meaningful reading. No guessing; no maybe's; no questions. Full stop. Period. So, that answers one question. 'Nuff said.

If the 4R, 4S & 4T relays aren't picking up (either during a normal START (with a READY TO START indication)) or when forcing them then the problem is likely that 28 VDC (specifically P28VR) IS NOT getting to the common side of the three relays. Per the drawing sheets you provided, that means that either the panel-mounted Emergency Stop Push-Button is latched in the TRIP position or it's not working and the contacts haven't closed when released from the TRIP position, or the PERM contact isn't closed. I don't believe you mentioned the state of the PERM relay in your description of the problem and the troubleshooting and troubleshooting results you provided.

The PERM contact was often bypassed with a wire jumper between RTB-8(6) and RTB-8(7) if nothing was used to actuate the PERM relay (such as a remote E-Stop P/B in a Central Control Room or somewhere else in the plant--there may even be two or more E-Stop P/Bs in series to the PERM relay). If there's no jumper between RTB-8(6) and RTB-8(7) then something has to be connected between RTB-1(10) and RTB-9(6). SOMETIMES another permissive contact was connected between RTB-9(7) and RTB-2(10). (The drawing sheet shows a jumper between RTB-9(7) and RTB-2(10), but there might be a contact connected in place of the jumper--maybe from a DCS or BOP or gas compressor controller, who knows. Unfortunately, the drawings weren't always properly marked-up in the field and returned for as-builts.)

If the PERM contact is closed (meaning the PERM relay is energized) OR there is a jumper between RTB-1(10) and RTB-9(6), you should be able to read 28 VDC with your meter at RTB-8(10) and RTB-8(11) when 4R, 4S & 4T are de-energized (dropped out). If you can manually push the buttons on relays 4R, 4S & 4T and get 4X1, 4X2, 4X3, 4X4 & 4X5 to pick up (energize), then the problem is that there is a wiring issue in the backplane of the <RELAY1> module. To find and fix that you need to remove the relay module (no easy task) and have a copy of the Speedtronic Panel Connection Diagram and check every connection between the relay socket bases for that circuit. A combination of Termipoint and wire-wrap methods were often used to wire the relay module backplanes, so repairing them (if necessary) is going to require one or the other. Wire was usually provided in the Start-up Kit that was provided with the Mark* IV when it was new.

It might be a problem with the three control processors (you haven't really told us that you have a READY TO START when trying to START the turbine, and you haven't told us precisely WHAT Process Alarms are active when trying to START the turbine...). But that's pretty unlikely.

There are some tests that are run when the processors are booting up (<C> MUST BE powered-up and running FIRST before any of <R>, <S> or <T> can start--they have to go to <C> for information and if <C> isn't running then <R>, <S> and <T> ain't gonna get nothing they need for initialization). There are also some diagnostic tests (briefly mentioned in the Mark V Maintenance Manual) which can be run for additional checks. But, it's really difficult to imagine how 4R, 4S & 4T can ALL not be picking-up if there is a READY TO START and both of the PSRELn power supplies are working and the E-Stop P/Bs and any external permissives are made up (closed)--unless there's been a serious problem in the backplane of the <RELAY1> module, and that would have resulted in some smoke and a sickening smell that someone should have noticed. It DOES require two of <R>, <S> or <T> to energize the 4X1, 4X2, 4X3, 4X4 & 4X5 relays, and you say none of them are picking up (energizing)--so that probably means a wiring issue or a P/B/permissive issue. A LOT of people use a bicylcle lock in the panel-mounted E-Stop P/B to prevent STARTing during a maintenance outage, and sometimes that has caused a failure of the E-Stop P/B when some care wasn't used when inserting and removing the bicycle lock. (It is an industrial P/B--the one that was originally installed in the Mark* IV--we don't know what's in there now--but it will only take so much abuse and that panel is no spring chicken.)

That's all I can offer without being on site, and that's not happening in this century.

Best of luck. Please write back with results and what you discovered and how you fixed it.

 

ONE of the very good things about GE Mark* IV and Mark* V turbine control systems is that COM is ground is earth. That goes for PCOM, DCOM, COM, ACOM, earth, ground--they're all the same. NOT LIKE SOME control systems that have one or three or more grounds they reference their control system voltages to. So, when measuring DC voltages (5 VDC, +/+12 VDC, 28 VDC, 24 VDC, 125 VDC) attach one lead of the meter to a proper earth/ground and you can measure any voltage with respect to ground and get a meaningful reading. No guessing; no maybe's; no questions. Full stop. Period. So, that answers one question. 'Nuff said.

If the 4R, 4S & 4T relays aren't picking up (either during a normal START (with a READY TO START indication)) or when forcing them then the problem is likely that 28 VDC (specifically P28VR) IS NOT getting to the common side of the three relays. Per the drawing sheets you provided, that means that either the panel-mounted Emergency Stop Push-Button is latched in the TRIP position or it's not working and the contacts haven't closed when released from the TRIP position, or the PERM contact isn't closed. I don't believe you mentioned the state of the PERM relay in your description of the problem and the troubleshooting and troubleshooting results you provided.

The PERM contact was often bypassed with a wire jumper between RTB-8(6) and RTB-8(7) if nothing was used to actuate the PERM relay (such as a remote E-Stop P/B in a Central Control Room or somewhere else in the plant--there may even be two or more E-Stop P/Bs in series to the PERM relay). If there's no jumper between RTB-8(6) and RTB-8(7) then something has to be connected between RTB-1(10) and RTB-9(6). SOMETIMES another permissive contact was connected between RTB-9(7) and RTB-2(10). (The drawing sheet shows a jumper between RTB-9(7) and RTB-2(10), but there might be a contact connected in place of the jumper--maybe from a DCS or BOP or gas compressor controller, who knows. Unfortunately, the drawings weren't always properly marked-up in the field and returned for as-builts.)

If the PERM contact is closed (meaning the PERM relay is energized) OR there is a jumper between RTB-1(10) and RTB-9(6), you should be able to read 28 VDC with your meter at RTB-8(10) and RTB-8(11) when 4R, 4S & 4T are de-energized (dropped out). If you can manually push the buttons on relays 4R, 4S & 4T and get 4X1, 4X2, 4X3, 4X4 & 4X5 to pick up (energize), then the problem is that there is a wiring issue in the backplane of the <RELAY1> module. To find and fix that you need to remove the relay module (no easy task) and have a copy of the Speedtronic Panel Connection Diagram and check every connection between the relay socket bases for that circuit. A combination of Termipoint and wire-wrap methods were often used to wire the relay module backplanes, so repairing them (if necessary) is going to require one or the other. Wire was usually provided in the Start-up Kit that was provided with the Mark* IV when it was new.

It might be a problem with the three control processors (you haven't really told us that you have a READY TO START when trying to START the turbine, and you haven't told us precisely WHAT Process Alarms are active when trying to START the turbine...). But that's pretty unlikely.

There are some tests that are run when the processors are booting up (<C> MUST BE powered-up and running FIRST before any of <R>, <S> or <T> can start--they have to go to <C> for information and if <C> isn't running then <R>, <S> and <T> ain't gonna get nothing they need for initialization). There are also some diagnostic tests (briefly mentioned in the Mark V Maintenance Manual) which can be run for additional checks. But, it's really difficult to imagine how 4R, 4S & 4T can ALL not be picking-up if there is a READY TO START and both of the PSRELn power supplies are working and the E-Stop P/Bs and any external permissives are made up (closed)--unless there's been a serious problem in the backplane of the <RELAY1> module, and that would have resulted in some smoke and a sickening smell that someone should have noticed. It DOES require two of <R>, <S> or <T> to energize the 4X1, 4X2, 4X3, 4X4 & 4X5 relays, and you say none of them are picking up (energizing)--so that probably means a wiring issue or a P/B/permissive issue. A LOT of people use a bicylcle lock in the panel-mounted E-Stop P/B to prevent STARTing during a maintenance outage, and sometimes that has caused a failure of the E-Stop P/B when some care wasn't used when inserting and removing the bicycle lock. (It is an industrial P/B--the one that was originally installed in the Mark* IV--we don't know what's in there now--but it will only take so much abuse and that panel is no spring chicken.)

That's all I can offer without being on site, and that's not happening in this century.

Best of luck. Please write back with results and what you discovered and how you fixed it.

Dear Engineer,

Thank you for your great support,

Once start command given, the turbine is running crank mode normal condition without any alarm, after turbine speed reach Fire mode, then Gas trip pressure low alarm appear, after that hydraulic pressure low, then tripped. During running I saw physically 20FG relay not energize, I followed 20FG relay there need a coil power, P28VR-4 is required to energize the 20FG, then I problem has been found this master protective relay is not active. (4R, 4S, 4T) & 4X-1, 4X-2, 4X-3,4X-4,4X-5 is not active. For Master protective relay testing I done two mode,

1. I forced in software L4X1, Then i checked master protective relay is not active.
2. Then I pressed manually 2 relays 4R & 4S, then 4X-1, 4X-2, 4X-3,4X-4,4X-5 is active,

The permissive relay is healthy, I checked voltage across all terminal RTB-8(6), RTB-8(7), RTB-8(10), RTB-8(11) is getting +28VDC across with ground.

As you given suggestion MARK-IV bootup, may I mistake to bootup procedure . I try the step and update you soon,

Once again thank you for your support.

 

Dear Engineer,

Thank you for your support.

Problem has been resolved.

May I mistake during MARK-IV startup. I followed by below steps.

1. <C> core started wait until healthy.
2. <R> core started wait until healthy.
3. <S> core started wait until healthy.
4. <T> core started wait until healthy.

All system healthy. then i went to force

After I forced L4X1 and found master protective relay is active (4R,4S,4T, 4X-1, 4X-2, 4X-3, 4X-4, 4X-5).
then force L20FG and the relay is active. Problem resolved.


Conclusion: MARK-IV Startup procedure to be followed.


Many thanks to all comments.

Thanks & regards,
Selvalagan. A

 

alagan85,
Glad to hear your problem solved, appreciate the feedback. I am curious though, in your original list item 3 said R,S,T and C cores were running healthy. When you rebooted them the last time you waited until one processor was up and healthy before you did the next one. I have a vague memory of this requirement. Was this simply by chance that you did it differently or did you gain some insight from someone to do this. Were there any telltale signs that they were not synced up. As i recalled this was a noticed improvement when we upgraded to the Mark VI. As I recall, on the Mark VI there is an LED that flashes. Once they are all sync up, they flash in unison. Again thanks for sharing the solution to your problem.

 

Thank you

alagan85,
Glad to hear your problem solved, appreciate the feedback. I am curious though, in your original list item 3 said R,S,T and C cores were running healthy. When you rebooted them the last time you waited until one processor was up and healthy before you did the next one. I have a vague memory of this requirement. Was this simply by chance that you did it differently or did you gain some insight from someone to do this. Were there any telltale signs that they were not synced up. As i recalled this was a noticed improvement when we upgraded to the Mark VI. As I recall, on the Mark VI there is an LED that flashes. Once they are all sync up, they flash in unison. Again thanks for sharing the solution to your problem.

 

The Mark* IV was--and still is--an excellent turbine control system. It does, however, have a serious flaw that occasionally rears its ugly head. The four processors ARE NOT synchronized. I don't know if this is the cause of the problem in this thread, and I don't think it was. But, it's something that anyone who's still fortunate enough to be working on a Mark* IV should know and understand. It's one of the common causes--if not the most common cause--of Voting Mismatch Diagnostic Alarms.

The Auxiliary Display can be used to check Process- and Diagnostic Alarms in each individual processor. Sometimes, it's the ONLY way to see what's really "going on" in that "mind" of a particular processor. Process Alarms aren't annunciated by the Mark* IV (via the <C> processor) unless two of the three control processors agree that a Process Alarm exists.

You, alagan85, said that there were no Diagnostic Alarms. That seems pretty odd. It also seems odd that a turbine START could be initiated (in any mode, actually), given that it was impossible for the 4 relays to be energized.

Another thing to remember about the Mark* IV is that it was one of the first digital turbine control systems--ever. It was brilliant for its time, and still is a very good control system. It just lacks some simple troubleshooting abilities--at least we think they're simple today. But, back then--when i80186 and i80286 microprocessors were what was available--and RAM was hella expensive, it was damn hard to get into the control system. And desktop computers were very expensive and not very portable.

Anyway, glad to hear the problem is solved. Mark* IV boot-up (initialization; "start-up") is very precise--in that <C> MUST BE initialized and ready to communicate with the control processors. They are "dumb" at boot-up--they have to get their instructions from <C>.

Over and out.