I am having a problem with a 7FA tripping at Base Load. The MarkV trip log viewer shows "Gas Fuel Hydraulic trip pressure low". Both Trip Oil dump valves and their associated coils have been changed as well as all three redundant trip oil switches. Could someone tell me the triggers for these trip oil valves to actuate and dump the hydrualic oil? Any help is appreciated.

 

Dear Genprospec,

it sounds as if your trip oil pressure is indeed low, have you verified it at the local gauge panel while running? With a tight system the trip pressure at the gas skid gauge should be around 50 -60 psi. When you say you have replaced "trip oil dump valves" would those be 20FG and 20FL? I would advise that you refer to your trip oil schematic, ML418 to see how trip oil is piped for your 7FA. My experience is with the 7ea, but I think the 7FA is the same. Trip oil circuit on our machine consists of trip oil to 20TV (IGV), 20FL (liquid fuel), and 20FG (gas fuel). To establish trip oil pressure to the gas and liquid fuel systems the first trip valve that must close is 20TV. So it is possible that you have a problem with 20TV or its related piping.

The alarm "gas fuel hyd pressure low" means the MKV has seen 2 of 3 trip oil pressure switches indicating low trip oil pressure when it does not expect it, it is commanding 20FG to close (L20FG) is true, but pressure switches 63HG indicate low pressure.

Control of the trip oil solenoid valves is controlled by the MKV main cores and the protection cores. You need to make sure that there are no other alarms or events that occur before the "gas fuel hyd pressure low". It sounds like there aren't from your post. Basically if the MKV sees 2 of 3 trip oil pressure switches indicating low trip oil pressure it will initiate a trip to the turbine, and you get the alarm you see.

If some other condition was occuring that made the MKV want to trip the machine, then it will deenergize the trip valves and trip the unit, but you should not see the alarm "gas fuel hyd pressure low".

Please write back and let us know what you find.

 

The Device Summary lists the pressure that the Dump Valves operate at, and it's usually about 20 psig, decreasing.

As with all alarms, you need to work "backwards" from the Alarm Drop Number, using the Alarm List and the CSP Crossreference and -Printout. Using the drop number, the Alarm List will tell you the logic signal name which, when it transitions to a "1" will cause the alarm to be annunciated.

There is always a LOT of confusion when reading Trip Historys, as there is usually about three seconds of alarms/data <b>*AFTER*</b> the trip condition. So, you need to find precisely when the trip occurred and what condition caused the trip to occur.

Also, many F-class units have Trip Oil Pressure transducers, and if yours does you can monitor it using the Mark V. You should also be albe to monitor the three trip oil pressure switches, as MIKEVI says, to see if any of them are actually indicating low pressure prior to the trip condition.

One of the many problems with F-class sequencing, which has become extremely "bloated" and needlessly complicated, is that some alarms which are "nuisance" alarms, in that they are "after the fact", are not "blocked" from being annunciated. In other words, conditions that occur after the "main" alarm condition should be blocked from being annunciated but are not.

It's not common but has occurred too many times that this particular alarm is annunciated after the real trip condition because the logic doesn't properly block the low trip oil pressure.

Again, the <b>LAST</b> alarm in the Trip History display is <b>NOT</b> the alarm which tripped the turbine, usually, because of the three seconds of post-trip data and alarms which are included in the Trip History.

So, from the Trip History Display, what are the last five seconds-worth of alarms that are annunciated that are listed in the alarm queue in the Trip History? Please list the alarms and times in the order they are listed in the Trip History.

What Diagnostic Alarms are being annunciated prior to and during the trip condition?

And, as MIKEVI says, you need to look at your L.O. and Trip Oil Piping Schematic drawings and determine all the solenoids that must be energized to maintain Trip Oil pressure. Many Trip Oil Systems require that the IGV Dump Solenoid also be energized, but not all. Some F-class units do; some do not. It was never clear what the driving reasoning was behind the decision to use the IGV Dump Solenoid valve in the Trip Oil System or not. Just one of those "generational" things that is not documented.

So, use your Piping Schematic Drawings and your Device Summary and your Alarm List and your CSP Crossreference and -Printout to thoroughly understand your Trip Oil System and components and their settings and the logic which annunciates alarms and trips.

 

Hi,

In addition to what has been suggested, I understand that it would be of some help to you to monitor the L63HL1L,2 & 3 (L63HG1L in your case) on the pre-vote data display on your <I> station.

From what MIKEVI, has mentioned about operating of 20TV, I am not clear about what should make 20TV act in case of 20 FL or FG action to trip. Is it for a faster dump action considering that 20FG (or FL) would always take some time to drain considering the line sizing. I think for building up pressure parallel closure of 20 TV & FG ( or FL) is okay upon L4 being true. Pl. clear and correct me, if required with details of sequencing.

 

Thank you for all the replies, they were indeed very informative. The problem was one of the three trip oil switches. They were all replaced as part of the initial solution but one inadvertently got put in that was also bad. Thanks again and sorry for the delay in responding.

 

Lessons learned. I am amazed that we still protect expensive equipment using switches. Their day should be well and truly over now that transmitter prices are close to switch prices (unless you use $50 commercial grade). I wonder how many industrial explosions have an undetected faulty switch lying crumpled in the rubbish of the disaster.

You only find out the hard way. Be interesting to know how much your trips cost you?

 

In this case, when GE started making heavy duty gas turbines, pressure transmitters were very expensive. And the electronics to make use of the signals were also very expensive to design and manufacture.

In addition, early pressure transducers were not always very reliable. Their calibration drifted a lot, requiring constant adjustment, and they simply failed a lot of the time. They just were not reliable and were costly.

So, GE used pressure switches, and temperature switches, and the like instead of a lot of analog transmitters. And, because the predominant failure mode of most of these switches was known, GE made use of that in designing their control systems to provide very high reliability and availability.

Now, GE is a very large corporation. The inertia in very large corporations is a very difficult thing to overcome. Almost impossible. Not as difficult as getting a power plant operator to accept a new operator interface without whinging (now there's a true definition of mission impossible), but very difficult. The amount of inertia (resistance to change) seems to be proportional to the size of the organization.

Add to that the fact that a lot of buyers and operators of GE heavy duty gas turbines are repeat buyers and some have large fleets of them. They get used to things being done in a certain way, and they don't necessarily like change very much either. They love it when GE tells them, "Our new control system is virtually like the old one, only better! It has color graphics and more bells and whistles, but the basic control and protection scheme, which we've been doing for decades, remains basically unchanged!" Customers just love that.

While I'm grateful for the feedback, I'm still very suspicious of the whole description of events and the explanation here. GE turbine control systems, especially those in use on F-class turbines, employ a lot of redundant instruments for reliability. In this case there are three switches all providing the same indication to the control system. Now, I haven't looked at the bloated and overly complicated sequencing of an F-class unit lately, but my recollection is that it would take two of the three switches indicating a loss of trip oil pressure to cause a trip on low trip oil pressure.

That's the whole purpose of using redundant devices: It takes more than one device indicating a problem before the unit is tripped. If only one device indicates a problem, there's usually an alarm to say there's a "sensor" trouble condition, but the units don't generally trip for any reason on a single sensor failure, especially when there are redundant sensors. And in this case, there are three switches.

Having said the above, the rocket scientists designing F-class turbines have surprised me many times. Why it's necessary to rewrite control schemes for systems that have been working fine for years on other Frame sizes and which are identical on the F-class units is just beyond comprehension.

Lastly, just about every time I've encountered a problem with a failed pressure switch, it's because the device hadn't been "calibrated" in years or it was obviously damaged (leaking diaphragm or broken capillary tube), two things which should have been noticed while power plant operators were making rounds. But, they don't think that rounds are productive uses of their time, and they think that rounds are just busy work and they shouldn't have to do them.

Switches by and large are excellent and reliable devices, if configured and employed and maintained properly. Like any other piece of equipment.

I know of one site where they were so keen to get rid of their "old technology" pressure switches, they replaced the two redundant low-low L.O. pressure switches with a single pressure transducer. And, instead of using a proper screened (shielded) pair cable for the signal, they used the old 14 AWG single conductors which had been used for one of the switches.

They continually had nuisance trips on low L.O. pressure, and the unit was unreliable. They ran it like that for a year, sometimes with the unit tripping more than once a week. They finally called GE and complained that it was a Mark VI problem. We temporarily ran a screened pair cable from the transmitter to the Mark VI, and the unit didn't trip once on low L.O. pressure in over a month.

Last time I checked, the "temporary" screened pair cable was still being used to get the transmitter signal into the Mark VI. I don't know about the low L.O. pressure trips, but that "temporary" cable was installed in 2002. And the last time I checked on the cable was in 2008.

So, one can have the latest and greatest whiz-bang, inexpensive instrument, but if it's not applied correctly, it's worse than the "old technology" devices being replaced.

Me, I'm about reliability. Sure, I'd like to see more transmitters for lots of things. Power plant operators would like to see more transmitters simply because they won't have to go take gauge readings (translation: they won't have to get off their arses).

By the way, what part of the automationa and control industry are you in?

 

Thank you for your detailed response. Appreciated. I am in Power Generation. Principal I & C engineer for a company currently a third of the way through building 7000 MW of generation. Current major activity is commissioning 5 off GE Frame 9E CT's, 3 of which are combined cycle, two peakers.

 

it's better you cheek :
1- trip oil and hyd. oil pass
2- trip oil filter
3- monitoring oil temperature

 

The Device Summary lists the pressure that the Dump Valves operate at, and it's usually about 20 psig, decreasing.

As with all alarms, you need to work "backwards" from the Alarm Drop Number, using the Alarm List and the CSP Crossreference and -Printout. Using the drop number, the Alarm List will tell you the logic signal name which, when it transitions to a "1" will cause the alarm to be annunciated.

There is always a LOT of confusion when reading Trip Historys, as there is usually about three seconds of alarms/data <b>*AFTER*</b> the trip condition. So, you need to find precisely when the trip occurred and what condition caused the trip to occur.

Also, many F-class units have Trip Oil Pressure transducers, and if yours does you can monitor it using the Mark V. You should also be albe to monitor the three trip oil pressure switches, as MIKEVI says, to see if any of them are actually indicating low pressure prior to the trip condition.

One of the many problems with F-class sequencing, which has become extremely "bloated" and needlessly complicated, is that some alarms which are "nuisance" alarms, in that they are "after the fact", are not "blocked" from being annunciated. In other words, conditions that occur after the "main" alarm condition should be blocked from being annunciated but are not.

It's not common but has occurred too many times that this particular alarm is annunciated after the real trip condition because the logic doesn't properly block the low trip oil pressure.

Again, the <b>LAST</b> alarm in the Trip History display is <b>NOT</b> the alarm which tripped the turbine, usually, because of the three seconds of post-trip data and alarms which are included in the Trip History.

So, from the Trip History Display, what are the last five seconds-worth of alarms that are annunciated that are listed in the alarm queue in the Trip History? Please list the alarms and times in the order they are listed in the Trip History.

What Diagnostic Alarms are being annunciated prior to and during the trip condition?

And, as MIKEVI says, you need to look at your L.O. and Trip Oil Piping Schematic drawings and determine all the solenoids that must be energized to maintain Trip Oil pressure. Many Trip Oil Systems require that the IGV Dump Solenoid also be energized, but not all. Some F-class units do; some do not. It was never clear what the driving reasoning was behind the decision to use the IGV Dump Solenoid valve in the Trip Oil System or not. Just one of those "generational" things that is not documented.

So, use your Piping Schematic Drawings and your Device Summary and your Alarm List and your CSP Crossreference and -Printout to thoroughly understand your Trip Oil System and components and their settings and the logic which annunciates alarms and trips.

Air i want to know that when we start MK-V OR MK-VIe gas turbine and suppose during startup flame not proved after purging, then to which pisition GT will come, mean it remain on crank speed or come to zero speed??

 

Secondly when ratchet device stop during GT start up. I mean at pick up of speed relay 14HR or at 14HM??

 

Air i want to know that when we start MK-V OR MK-VIe gas turbine and suppose during startup flame not proved after purging, then to which pisition GT will come, mean it remain on crank speed or come to zero speed??

Depends on your software, it could be either

 

Secondly when ratchet device stop during GT start up. I mean at pick up of speed relay 14HR or at 14HM??

14HR picks up at Zero Speed, ratchet will stop when 14HR drops out as soon as shaft moves

 

MSF,

It would be most helpful if you would specify which type of GE-design heavy duty gas turbine you are working on and what control system it uses. Not critical to the answer, but helpful.

Most ALL GE-design heavy duty gas turbine control software I have seen remains in the Master Control Mode selected when the START was initiated if the unit fails to fire on the first attempt. Most all GE-design heavy duty gas turbine control software I have seen has a little ... "feature" that GE refuses to remove that will often automatically attempt to re-fire the unit, and if that attempt is unsuccessful then the turbine control system often goes back to COOLDOWN (hydraulic ratchet; turning gear; slow roll; whatever the unit uses for cooldown operation).

So, let's say you are starting the unit after it was taken off COOLDOWN, (or you may put the unit on COOLDOWN after it was sitting for some time at zero speed), you would normally select either AUTO or REMOTE Master Control Mode and if the unit indicates READY TO START you would then initiate a START. The unit will go through a purge sequence and then attempt to fire by first energizing the ignitors ("spark plugs") and then admitting fuel. If no flame is detected during the firing sequence the fuel stop valve is closed and the ignitors are de-energized and often a Process Alarm "FAILURE TO IGNITE" is annunciated. Usually, the unit will remain at cranking speed awaiting some command from the operator. All that's necessary to re-attempt a fired start is to switch the Master Control Mode to CRANK and then back to AUTO (or REMOTE). That will re-enable the firing sequence for most units. As I wrote, some units will automatically re-initiate the purge sequence and then automatically attempt a second firing--and how which sequence is determined to be supplied on a control system is anybody's guess (there is no standard for the decision).

Some units, a few, were modified in the field to automatically cancel the START attempt and allow the unit to return to COOLDOWN operation. (Some were modified by the factory to do so, but that was even a smaller number.) Again, usually, on a failure to fire on a 'first' START attempt the unit remains at CRANK speed and in AUTO (or REMOTE), waiting for the operator to take some action. This is desirable, as it keeps the axial compressor spinning to push air through the turbine and exhaust to purge the unit of any combustible gases (and liquid fuel is the unit was being started on liquid fuel). Again, if the unit doesn't automatically attempt a second firing, all the operator has to do is switch the Master Control Mode from AUTO (or REMOTE) to CRANK mode, and then back to AUTO (or REMOTE). That will re-enable the firing sequence after the purge timer has completed.

On many units if the unit fails to fire a second time, the unit will usually go into STOP mode, shutting down the starting means and the unit will coast down to COOLDOWN.

Usually, any time the unit goes through a firing sequence and flame is not established the control system will annunciate the Process Alarm "FAILURE TO IGNITE."

As glemorangie said, it's very difficult to tell you exactly what will happen at your site without being able to see the software in the turbine control system at your site. Many times it's the typical software, as described above. Some times, the software was modified during commissioning or afterwards based on site demands, ..., er, ..., uh, ..., requirements (that usually weren't thought through very well because the software wasn't understood or was supposed to do something different than it did--based on very little thought and consideration or experience or knowledge).

But, your unit (whatever it is, with whatever control system it has) will usually fall somewhere in the description above. And, if it doesn't, then it was most likely modified from the above (again, based on (unreasonable) demands and perception, not experience and consideration).

 

Secondly when ratchet device stop during GT start up. I mean at pick up of speed relay 14HR or at 14HM??

MSF

If the unit has a ratchet, the ratchet is usually used to help the turbine shaft start turning if being started from zero speed. And, if the unit is "on ratchet" (COOLDOWN) when STARTed, the ratchet mechanism will usually start operating when the START command is accepted by the control system. The ratchet mechanism will (should) stop when L14HR drops OUT, because L14HR picks UP when the unit goes to zero speed, and drops out very soon after the shaft starts spinning.

Most GE-design heavy duty gas turbines equipped with ratchet mechanisms for COOLDOWN do not have starting means (diesel engine or electric motor or expander turbine) that can accelerate the turbine-generator shaft from zero speed. In other words, the starting means does not have to torque to overcome inertia and break the shaft away from zero speed all by itself--it needs an assist, and it gets that assist from the ratchet mechanism. All it takes is just a little "bump" from the ratchet to help the starting means accelerate the shaft from zero speed ("break the shaft away from zero speed") and get it started turning. Once the speed detection logic/sequencing/application code detects a small amount of speed the ratchet mechanism is (should be) disabled as it is no longer necessary to help keep the shaft turning (actually the shaft is turning MUCH faster than the ratchet could turn the shaft).

Hope this helps!!!

 

Thanx sir

 

Si

MSF,

It would be most helpful if you would specify which type of GE-design heavy duty gas turbine you are working on and what control system it uses. Not critical to the answer, but helpful.

Most ALL GE-design heavy duty gas turbine control software I have seen remains in the Master Control Mode selected when the START was initiated if the unit fails to fire on the first attempt. Most all GE-design heavy duty gas turbine control software I have seen has a little ... "feature" that GE refuses to remove that will often automatically attempt to re-fire the unit, and if that attempt is unsuccessful then the turbine control system often goes back to COOLDOWN (hydraulic ratchet; turning gear; slow roll; whatever the unit uses for cooldown operation).

So, let's say you are starting the unit after it was taken off COOLDOWN, (or you may put the unit on COOLDOWN after it was sitting for some time at zero speed), you would normally select either AUTO or REMOTE Master Control Mode and if the unit indicates READY TO START you would then initiate a START. The unit will go through a purge sequence and then attempt to fire by first energizing the ignitors ("spark plugs") and then admitting fuel. If no flame is detected during the firing sequence the fuel stop valve is closed and the ignitors are de-energized and often a Process Alarm "FAILURE TO IGNITE" is annunciated. Usually, the unit will remain at cranking speed awaiting some command from the operator. All that's necessary to re-attempt a fired start is to switch the Master Control Mode to CRANK and then back to AUTO (or REMOTE). That will re-enable the firing sequence for most units. As I wrote, some units will automatically re-initiate the purge sequence and then automatically attempt a second firing--and how which sequence is determined to be supplied on a control system is anybody's guess (there is no standard for the decision).

Some units, a few, were modified in the field to automatically cancel the START attempt and allow the unit to return to COOLDOWN operation. (Some were modified by the factory to do so, but that was even a smaller number.) Again, usually, on a failure to fire on a 'first' START attempt the unit remains at CRANK speed and in AUTO (or REMOTE), waiting for the operator to take some action. This is desirable, as it keeps the axial compressor spinning to push air through the turbine and exhaust to purge the unit of any combustible gases (and liquid fuel is the unit was being started on liquid fuel). Again, if the unit doesn't automatically attempt a second firing, all the operator has to do is switch the Master Control Mode from AUTO (or REMOTE) to CRANK mode, and then back to AUTO (or REMOTE). That will re-enable the firing sequence after the purge timer has completed.

On many units if the unit fails to fire a second time, the unit will usually go into STOP mode, shutting down the starting means and the unit will coast down to COOLDOWN.

Usually, any time the unit goes through a firing sequence and flame is not established the control system will annunciate the Process Alarm "FAILURE TO IGNITE."

As glemorangie said, it's very difficult to tell you exactly what will happen at your site without being able to see the software in the turbine control system at your site. Many times it's the typical software, as described above. Some times, the software was modified during commissioning or afterwards based on site demands, ..., er, ..., uh, ..., requirements (that usually weren't thought through very well because the software wasn't understood or was supposed to do something different than it did--based on very little thought and consideration or experience or knowledge).

But, your unit (whatever it is, with whatever control system it has) will usually fall somewhere in the description above. And, if it doesn't, then it was most likely modified from the above (again, based on (unreasonable) demands and perception, not experience and consideration).

Sir I am working at GE FRAME-V (MS-5001) GT.

 

Thank you

Depends on your software, it could be either

 

MSF,

Thanks for the partial clarification.

Now, is it a Mark V or a Mark V-to-Mark VIe upgrade?

 

MSF,

Thanks for the partial clarification.

Now, is it a Mark V or a Mark V-to-Mark VIe upgrade?

We have two GTs one has MK-V second one has MK-VIe