GE frame V Gas Turbine Starting Device Trip

S

Thread Starter

SANJAY MALLIK

why our GE frame V gas turbine trip due to starting device trip at 2200-2300 rpm during startup where as starting diesel speed is 1900 rpm. After firing its average temperature (TTXC) increases and when it reached to 507 TTXM degree Celsius, it control mode change from startup to temperature control mode and FSR was 18.2%. At 2200 rpm turbine get trip due to starting device trip. CPD value was-0.12 bar, FSRG was 5.57%, FSG was 14.57%, FPRG1 was 6.07 bar, FPG1 was 6.63 bar. CSRGV was 42.0 deg, CSGV was 41.9 deg. Our control system is MARK-IV speedtronic control system.

Please give us some idea what might be the probable cause of this tripping.

Thanking you
email id is : [email protected]
 
SANJAY MALLIK,

You need to look at the rung for the Starting Device Trip alarm to determine what condition might be causing the Mark IV to trip the unit. I think the signal name might be L4SMT, but if I recall correctly there is more than one signal which can cause the trip--so you need to determine which one of the multiple signals that can actuate L4SMT is actually tripping the turbine.

The starting means (whether it's a diesel engine or an electric starting motor or an expander turbine) is NOT directly coupled to the shaft. Virtually no starting means could develop the torque required to break the shaft away from zero speed to accelerate the shaft for starting. That's why the torque converter is there--it's a hydraulic "coupling" (actually it's a hydraulic pump driven by the starting means, and a hydraulic motor which drives the torque converter output shaft which is coupled to the Accessory Gear which is coupled to the turbine shaft through the Accessory Coupling). The starting means can spin at a very high speed to develop a high torque while the torque converter output shaft and turbine-generator shaft spin up at much lower speeds. Sometimes, depending on the type of torque converter, at higher turbine-generator shaft speeds the speed of the starting means is higher or lower than the turbine-generator shaft speed (it just depends on the design of the torque converter). So, diesel engine starting speed is almost never going to be the same as turbine-generator shaft speed--ever, not during break away of the turbine-generator shaft from zero speed nor when the turbine generator shaft should be accelerating on its own (self-sustaining speed during acceleration on the way to FSNL).

I'm going to guess that because the unit is going on Temperature Control during acceleration that the fuel is being limited--so the turbine can't develop enough torque to add to the torque of the starting means and help the unit achieve self-sustaining speed and continue accelerating to FSNL without the torque converter assist. Some times when this happens the turbine-generator shaft speed will actually start to decelerate--and that's NOT good for the machine, not at all. There is logic called "bog down" logic which senses the decreasing speed and will trip the unit to prevent damage caused by insufficient fuel to keep the unit accelerating during starting. But, there is usually an alarm to that effect ("STARTING DEVICE BOGGED DOWN" or something similar).

If the torque converter has not been recently refurbished it could just be that the torque converter is "tired" and can't transmit the proper torque. Also, diesel engines need maintenance, too. I have seen diesel fuel filters get plugged because of water in the fuel and starve the engine of fuel--so it might be trying to put out full power but not enough fuel can get through the fuel filter. I have also seen diesel engine air filters that were so dirty they couldn't pass enough air to the engine for it to develop full torque. And, also sometimes the fuel rack mechanism just gets worn out and needs adjustment or replacement and won't let enough fuel get to the engine to develop full torque.

So, you need to find out why the unit is tripping. I'm presuming the only Process Alarm you are receiving from the Mark IV is "STARTING DEVICE TRIP." You need to find out what condition is causing that alarm. I don't know if you have Rung Display on that Mark IV (it wasn't on every Mark IV), or you will have put all the elements of the L4SMT rung (if that's the correct logic name!) on the Logic Forcing Display and watch to see which one is picking up and causing L4SMT to pick up. And, then resolve that issue.

It's either a worn out and tired torque converter, or a worn out and tired diesel, or a plugged diesel fuel filter, or a dirty diesel air filter--or some combination of the above that is causing insufficient torque to be developed or transmitted to the turbine-generator shaft. OR, it's a problem with excessive fuel flow trying to assist with insufficient torque that's causing the unit to go on temperature control during acceleration and limit the fuel. Or, it could be come combination of all of the above (happens on older machines too often).

Please write back with details of what you find. And, if you write back for more assistance, please tell us when this problem started. After a maintenance outage? After a trip from load? Did it just start?

Hope this helps!
 
Sanjay

Probably more information is required to know what was wrong at your site. Can you provide the list of process and diagnostic alarms before or after the trip? What are the maximum RPM at your site? Do you have separate speed pickups for turbine and starting diesel engine as you said turbine speed was 2200 to 2300 RPM and diesel engine speed was 1900 RPM? Did L14HA pick or unit tripped before L14HA was picked up?

Below is the sample code for "Starting device Trip" alarm.<pre>
L4DE L14HR L4 L14HA L14HSX L94X L3SMT
---|/|----|/|----| |----|/|------|/|-------|/|------------( )
| | |
L33CSE | | |
---|/|-- | |
| |
| |
L3SMTX | |
----| |--------------- |
|
|
L60BOG |
---| |------------------------</pre>
If I were you, I would have troubleshooted like this. L3SMT must be "1" for "Starting device Trip" alarm to occur.

If L14HA is picked up the only cause that can trigger "Starting device Trip" is L60BOG. L60BOG will be "1" the unit starts to decelerate. If we look at the L60BOG block, it says that TNH must be lower than (5% or more) the previous value of TNH.

If L14HA is NOT picked up i.e. "1", then either L3SMTX must be "1" or BOTH L4DE AND L33CSE must be "0". L33CSE is "0" when clutch is not engaged. L4DE will be "0" if diesel engine run permissive is not present. In this case "Starting device Trip" will occur if starting clutch is not engaged AND diesel engine start permissive is not present.

If L3SMTX goes to "1" one of the following alarms may appear.
1. Starting device lockout
2. Diesel failure to start

Please post the process and diagnostic alarms to narrow down the problem. Hope it will help.
 
S

SANJAY MALLIK

It has been observe that around 2100 to 2200 rpm after firing the turbine tripped due to starting device tripped (L3SMT picked up). Diesel speed was 1900 rpm, and torque converter oil pressure was 2.8 bar. And most importantly temperature control not occurring now during acceleration as it happens earlier after fuel nozzle cleaned and few replace with new one. What might be the probable cause of tripping?

Thanking you.
 
SANJAY MALLIk

CSA has written in the above post

<I don't know if you have Rung Display on that Mark IV (it wasn't on every Mark IV), or you will have put all the elements of the L3SMT rung (if that's the correct logic name!) on the Logic Forcing Display and watch to see which one is picking up and causing L3SMT to pick up. And, then resolve that issue.>

Have you considered to look at Rung Display or Logic forcing display, which contact is making L3SMT "1"?
 
Esoteric _Stone,

You wrote:

>If L14HA is NOT picked up i.e. "1", then either L3SMTX must
>be "1" or BOTH L4DE AND L33CSE must be "0".

Because L4DE and L33CSE are in parallel, if EITHER L4DE >>OR<< L33CSE are Logic "0" >>OR<< L3SMTX is a logic "1" AND L14HA is a logic "0" then L3SMT will be a logic "1" and the turbine will trip.

Because the unit speed is below TNK14HA1 (usually 60% for a Frame 5 with a diesel starting means) the most likely cause of tripping is either the jaw clutch is disengaging (causing L33CSE to go to logic "0") or the diesel engine has tripped for some reason (which would also cause L33CSE to quickly go to logic "0").

If the jaw clutch teeth are worn the jaw clutch could easily disengage with any little bobble in torque from the starting means.

But, SANJAY >DID NOT< report what was observed on the rung display or in the Logic Forcing Display, probably because someone insisted they continuously monitor the Main Display instead of looking at the rung display or the Logic Forcing Display, which means they aren't going to get the required information to solve the problem. TNH an PN_PR can be added to the Logic Forcing Display to monitor the speeds as can TTXM. And if I recall correctly, three signals and their values can be added to the bottom of the rung display (TNH, PN_PR and TTXM) to be monitored while watching the rung display. But without data it's going to be extremely difficult to help with troubleshooting the problem.

And, yes, diesel starting means have a speed pick-up to measure diesel speed.
 
S

SANJAY MALLIK

it has been observed that at around 2200 rpm of turbine speed and diesel rpm was 1875 rpmò jaw clutch is disengaging (causing L33CSE to go to logic "0"). Thus tripped due L3SMT "1". Now what might have gone wrong in order to troubleshoot the problem?
 
This is normal operation for diesel shutdown. The turbine is producing more HP than the diesel, the clutch faces start to open, and the clutch opens. This should the diesel to idle rpm and cooldown before diesel shutdown.
 
2250 RPM is approximately 45% speed. If the jaw clutch is opening (L33CSE goes to logic "0") before L14HA picks up then L3SMT is going to pick up and trip the turbine.

I think I made a typographical error for the typical value if TNK14HA1 (the value of TNH at which L14HA goes to logic "1"). I meant to type 50%--but I don't have a CSPEC for a Frame 5 with a diesel starting means to be sure.

I believe the jaw clutch should open after L14HA goes to logic"1." So L33CSE should remain a logic "1" until after L14HA goes to logic "1." That way L3SMT will remain a logic "0" and the diesel will go into it's cooldown normally.

BUT, until we get good, actionable data from the original poster we won't know for sure. But, based on the information provided L3SMTX is not picking up, and we don't know about L4DE dropping out (which it shouldn't be), so it follows there is insufficient torque from the starting means to keep the jaw clutch halves engaged and L33CSE is dropping out before L14HA is picking up causing L3SMT to pick up and trip the turbine. Based on the information provided.

We need more--and better--information. Without better information we are all just guessing (as is the original posters).
 
S

SANJAY MALLIK

But the starting device trip happens at around 2200 rpm before self sustaining speed achieve by the turbine.What we have observe is before turbine self sustaining speed achieve the jaw clutch get disengaged and thus turbine tripped.Now is that diesel torque problem or some thing else.

>This is normal operation for diesel shutdown. The turbine
>is producing more HP than the diesel, the clutch faces start
>to open, and the clutch opens. This should the diesel to
>idle rpm and cooldown before diesel shutdown.
 
CSA

>Because L4DE and L33CSE are in parallel, if EITHER L4DE
>>>OR<< L33CSE are Logic "0" >>OR<< L3SMTX is a logic "1" AND
>L14HA is a logic "0" then L3SMT will be a logic "1" and the
>turbine will trip.

Spot on. Thanks for the correction.
 
SANJAY MALLIK,

>But the starting device trip happens at around 2200 rpm
>before self sustaining speed achieve by the turbine.What we
>have observe is before turbine self sustaining speed achieve
>the jaw clutch get disengaged and thus turbine tripped.Now
>is that diesel torque problem or some thing else.

The jaw clutch clutch normally disengages when the speed of the turbine side of the jaw clutch gets higher than the speed of the torque converter side of the jaw clutch. (The speed of the diesel is usually NOT the same as the output shaft of the torque converter.) This occurs when the torque being provided by the starting means is insufficient to keep the torque converter half of the jaw clutch spinning faster than the turbine half of the jaw clutch. So, the jaw clutch halves begin to open and the springs of the mechanism fully open them. As the jaw clutch halves slide open the limit switch arm changes position to tell the Mark IV the clutch is disengaged and to slow the diesel (during a normal START). If the jaw clutch opens BEFORE it should (before L14HA changes to logic "1") then the START is aborted and the unit is tripped. (Or if L4DE drops out before L14HA picks up, or if L3SMTX picks up during the START).

You haven't told us if the Mark IV has Rung Display or not. If it has Rung Display you need to put rung L3SMT on the display, put TNH or TNH_RPM at the bottom of the Display, switch to the Main Display and initiate a START, switch back to the Rung Display and observe when L4DE or L33CSE or L3SMTX changes state.

If the Mark IV doesn't have Rung Display you need to put the above signals plus L14HA plus L3SMT on the Logic Forcing Display, switch to the Main Display and initiate a START, then switch back to the Logic Forcing Display and observe which signal changes state just before L3SMT changes state.

You should also remove the cover over the jaw clutch and station someone with a radio to observe when the jaw clutch opens AND listen for when the diesel slows down. This person can also have a hand on the torque converter casing to get an idea of how hot the torque converter casing is getting (it should not get so hot that he has to remove his hand from the casing).

Before starting the turbine the jaw clutch limit switch should be inspected to be sure it is firmly being held in place and that the arm of the switch is also firmly attached to the switch mechanism. With the jaw clutch halves closed it should not be possible to manually jiggle the switch or the switch arm and cause L33CSE to change state on the Mark IV. Then, with COOLDOWN selected OFF (the ratchet OFF - meaning the Aux L.O. Pump is not running, you need to CAREFULLY pry the two halves of the jaw clutch apart while observing the limit switch. There are springs which are trying to keep the jaw clutch halves open, but the angle of the jaw clutch teeth should be keeping them closed. You should be able to CAREFULLY rotate the torque converter half of the jaw clutch to disengage the jaw clutch teeth. As the teeth disengage, the jaw clutch halves will VERY QUICKLY open (because of this springs!) so the person(s) doing this need to be very careful!!!

As the jaw clutch opens the limit switch should change state (go to logic "0"). Check the condition of the jaw clutch teeth. They should be smooth and not have signs of being worn, flattened or broken teeth. If the jaw clutch teeth are damaged or broken the jaw clutch halves need to be replaced.

You need to tell us also if there is a solenoid-operated valve, 20TU-1, on the starting means. If so, that solenoid or the valve may be failing or failed. Check the Starting Means P&ID to see if the solenoid-operated valve is being used on your turbine. If it is, with COOLDOWN OFF and no pumps or fans running in the Accessory Compartment, you need to station someone at the solenoid valve and have someone force/toggle the logic signal of the solenoid to "1" and "0" and "1" and "0" several times slowly. The solenoid should click faintly when the logic changes state. It will help to have a hand on the solenoid feeling for the clicking. If the solenoid valve doesn’t change state or close fully during starting then the torque being transmitted may be insufficient to help accelerate the turbine.

If it's been a long time since the starting diesel has been maintained, it's possible that one of the solenoids is failed or failing or the fuel rack needs adjustment. If the diesel was very recently worked on, it's possible the fuel rack was not adjusted properly and needs readjustment.

Finally, you need to tell us, in chronological order, >>ALL<< of the Alarms that are active and annunciated during the start attempt and after the trip. ALL of the Alarms - including the Diagnostic Alarms!!! Even if you think the alarm is not related to the problem, include ALL Alarms on the list.

SANJAY MALLIK, If you choose not to provide the above information - ALL OF IT - there is nothing more we can do to help you. You or your Company will need to hire a knowledgeable person to come to site to assist with the problem. We can probably help- but you must provide meaningful information. Repeating the same thing ("It has been observed that between 2200 and 2300 RPM when the diesel is at 1900 RPM the turbine trips...") is not meaningful after the first or second time. We are not there. We cannot see or hear what you can. You have to be an active participant in gathering and providing useful information in response to our requests. Use the information we have provided to get the information we have requested and provide it to us and we can probably help. But continuing to tell us the same thing repeatedly when we have asked for more information isn't helping get the unit started.

Get the information or get someone to site who can help you. It's as simple as that. You wrote asking for help. Sometimes there is a simple answer; sometimes there isn't. We asked for information in an effort to help. We are getting paid exactly the same amount you paid to post your request for help: Nothing. Just because you paid nothing doesn't mean if you want help you don't have to do anything to get a resolution to the problem. Get us the information requested or go pay someone a lot of money for what is probably a relatively simple problem that you are unable to diagnose or get information about.
 
SANJAY MALLIK,

I want to be perfectly clear. IF the jaw clutch is in fact disengaging before L14HA goes to logic &#8220;1&#8221; the reason is most likely either:

1) there is insufficient torque being developed by the diesel,

OR

2) there is insufficient torque being transmitted by the torque converter,

OR

3) if the unit has a “torque converter spoiler” solenoid-operated valve (20TU-1) the spoiler valve solenoid or valve is not working properly,

OR

4) there is insufficient oil getting to the torque converter (the foot valve is not working properly).

There are multiple reasons why the diesel may not be developing the proper torque - some have already been listed in responses above. There are multiple reasons why the torque converter may not be transmitting the proper torque; some have already been listed above.

All of the above presumes the jaw clutch is opening prematurely. We STILL DON&#8217;T KNOW THAT. The jaw clutch limit switch could be loose, failing, have loose wire terminations, or be improperly adjusted. That&#8217;s why someone with a radio needs to be stationed at the jaw clutch to immediately notify the Control Room when the jaw clutch opens.

The jaw clutch teeth may be damaged and that could cause the clutch to prematurely disengage (not likely if there is sufficient torque being transmitted, but it&#8217;s possible.)

The diesel fuel filter may be dirty (choked) and restricting fuel flow.

It&#8217;s even possible that L4DE is going to a logic &#8220;0&#8221; while the diesel is running because of low diesel L.O. pressure or a failing diesel L.O. pressure switch or a loose wire in the circuit or something like that, which is shutting the diesel down prematurely and causing the jaw clutch to open prematurely. BUT WE DON&#8217;T KNOW THAT BECAUSE YOU HAVEN&#8217;T PROVIDED THE REQUESTED INFORMATION.

The diesel may be in need of refurbishment. If white smoke is exiting the diesel exhaust stack then one or more cylinders is not developing the proper compression to fully ignite the fuel. The fuel injectors may be worn and not properly atomizing the fuel.

It&#8217;s even possible the Mark IV has been improperly configured to accelerate the turbine faster than it should; you complained at one time about excessive exhaust temperature. BUT, if that were happening there would likely be higher than normal exhaust temperatures. So, while this is possible, it’s not likely based on the information provided.

Often, a problem like this is the result of more than one cause. Sometimes, you have to start replacing components one at a time with new components that are properly calibrated and configured. And, sometimes you get lucky and replace the right device, and sometimes not. You may even have to replace the torque converter, or have the diesel rebuilt after replacing the fuel injectors. Or have the diesel fuel rack solenoids (20DV-1 and-2, I believe) replaced or re-adjusted.

The basic causes of L3SMT picking up have been listed. Now you have to work through all the possible reasons why those listed reasons might occur. If you want our help, please provide the requested information - ALL of it. The Mark IV is likely not the cause of the problem (but it could be—but there would likely be Alarms to indicate a problem). And the Mark IV is not the easiest control system to troubleshoot with, but it’s certainly not the worst, either.

I want to help you solve this problem - but you gotta give us more, better information.
 
SANJAY MALLIK,

So, this may be part of your issue. MANY people credit the Mark* for doing more than it actually does. If you think the Mark* opens the jaw clutch when the turbine is self-sustaining you would be thinking wrong.

As described before, when the speed of the turbine side of the jaw clutch over-runs the speed of the starting means side of the jaw clutch the teeth disengage and the springs which are always trying to open the jaw clutch pull the jaw halves apart. This is when the turbine speed is considered to be "self-sustaining." (It may actually be self-sustaining slightly before this point, but it's technically when the speed is considered to be self-sustaining--because no assist is required from the starting means.)

When the unit is starting, if the jaw clutch halves are NOT engaged the hydraulic ratchet is used to engage the halves and the torque from the starting means works to keep the teeth engaged as the unit breaks away from zero speed and accelerates. When the unit breaks away from zero speed the hydraulic ratchet is de-energized and the hydraulic rams which closed the jaw clutch halves are de-pressurized, and, again, it's strictly the torque from the starting means and the shape of the jaw clutch teeth that keeps the jaw clutch halves engaged.

Until the speed of the turbine side of the jaw clutch is higher than the speed of the starting means side of the jaw clutch--at which point the jaw clutch teeth disengage and the springs which are always trying to open the jaw clutch halves pull them completely apart.

Everyone wants the Mark* to be the cause of every problem (many believe it is the root of all evil). But, in this case, it is most likely not. The Mark* doesn't control everything on the turbine (or its auxiliaries) and it doesn't know when the turbine speed is self-sustaining and doesn't require any torque assist from the starting means. It presumes the turbine speed isn't self-sustaining until it is above TNK14HA1 (the pick-up setpoint for L14HA) and that is why it trips the turbine if it thinks the jaw clutch has disengaged before it reaches L14HA.

The Mark* doesn't know how much torque is being transmitted--it presumes the starting means is working at full efficiency and the torque converter is working properly. The Mark* sends signals to a diesel starting means to change speed and the diesel fuel rack mechanism senses load and increases fuel to try to maintain the speed defined by the position of the fuel rack. BUT, the Mark* doesn't control the speed of the diesel--it doesn't monitor the diesel speed and change the fuel rack position to maintain a speed setpoiont. (Yes, it knows what the diesel speed is, but that's for a basic indication of whether the diesel has started and should have proper L.O. pressure--it's NOT for controlling diesel speed.)

The adjustment of the fuel rack positions, as set by the two diesel "speed" solenoids (20DV-1 and -2 if I recall correctly). The diesels used by GE and its packagers are not exactly alike, nor are the torque converters. The choice of diesel and torque converter defines the setpoints of the diesel fuel rack. It's not the same for all frame 5s nor all GE-design heavy duty gas turbines with diesel starters; they are similar but not always the same.

Some GE-design heavy duty gas turbines have a torque converter "spoiler" solenoid-operated valve that helps to reduce speed after purging which aids in establishing flame. So, if your unit has such a solenoid and it's not working properly that could be part of the problem, if not the problem. It could be a loose terminal in the circuit powering the solenoid; it could be a bad electro-mechanical relay in the Mark* that isn't closing and/or opening when it should; it could be a bad solenoid; it could be the valve the solenoid is operating is bad or is sticking (not opening or closing fully).

So, this should help understand what the Mark* does and doesn't do. It gets (wrongly) blamed for MANY problems that are completely out of its control. It just doesn't have all the I/O to do all the things people think (or wish) it needs to do all the things people think it does. It relies on many things in the field to be working properly. Newer turbines with Mark* controls have LOTS more I/O and do LOTS more things--but that also complicates things much more.

Frame 5s are the workhorses of the GE-design heavy duty gas turbine fleet around the world. They just take a licking and keep on ticking, and they are relatively simple and while they aren't very efficient they are usually extremely reliable. I've seen Frame 5s opened up for maintenance that had half the length of the second stage turbine buckets missing and yet the unit was still making near rated performance (output). They were overbuilt for their time, and the control systems of that period (Mark IV vintage) just weren't that sophisticated--and didn't really need to be.

I hope this helps. I hope it helps to convince you that you need to give us the information we have requested--and in the process of getting that information you will likely discover what the problem is and how to solve it. The Mark* by tripping the turbine before it's reaching 14HA is trying to protect the turbine. If it thought the jaw clutch opened at 30% speed, that would be much too low for the unit to be self-sustaining. 45% is close to self-sustaining, but close still isn't enough to protect the turbine.

I think we've covered just about everything that can be the problem, and now we've covered how the unit works and how the Mark* works and what it's capable of and what it isn't capable of. If the turbine at your site was provided by GE (and not one of its packagers) the Operations & Service Manuals provided with the unit have brief but basic descriptions of each system and most of the components of each system. The rest of the understanding of how things work together to make the unit produce electricity come from getting out and looking at systems and components, and looking at things when they are taken apart and put back together. Even things like the angle, shape and condition of jaw clutch teeth are important. As are things like the bracket that hold the jaw clutch limit switch in place. And many of these things are NOT controlled by the Mark*--but they have to be maintained and adjusted and kept in working order so the Mark* can do what is is capable of doing.

Looking forward to hearing back from you!!!
 
Top