I will try to ask these questions briefly. Please be kind as this is my first post to this forum, and I am fairly new to the industry. (as far as GT's are concerned).
7EA GE Gas Turbine (DLN1+) MKV GE Control System
GE Steam Turbine
Plant MW Control Enabled
Michigan (cold, 10 degrees F)
1. While in preselect load, and lowering MW manually, how can an Operator determine at what point the GT will enter into Lean-Lean Mode? How does an Operator go about determining this point to avoid it.
2. What affect does operation of the GT at preselect load vs. base load have on GT maintenance? I've read somewhere that every hour spent in preselect load is considered 4 hours at base load from a maintenance standpoint.
3. While operating in preselect load, the bleed heat valve is fluctuating between 20-40% open which is causing MW swings. Is there a way to minimize these swings to increase plant MW output stability?
I'm sure there will be more questions, but I will start with this. Thank you in advance for your responses. I appreciate the time you have taken out of your day to respond.
I don't have access to any drawings or manuals at this writing, but there are combustion mode reference temperatures (TTRF1, usually) that are used as switches for changing combustion modes--usually. You need to find out what the values of the Control Constants are that TTRF1 is compared against to determine when the unit will switch combustion modes. And, then the operators need to monitor TTRF1 with the knowledge of the various temperature setpoints and not let TTRF1 get too close the value when lowering load.
And, contrary to extremely popular--and false--belief, it's NOT NECESSARY to use Pre-Selected Load Control in order for the unit to maintain a load setpoint. Simply use the RAISE- and LOWER SPEED/LOAD buttons on the HMI to get the load to some value where TTRF1 is not too close to switching out of Premix Mode, and then just leave the unit alone.
I believe as long as the unit is Premix Mode, whether it's on Droop speed control or Pre-Selected Load Control or at Base Load, the factored fired hours are all the same. It's only when the unit is NOT in Premix Mode that the factored fired hours change. Search the World Wide Web for GER-3620R for more information on factored fired hours and operation in different modes, but I don't think you'll find anything specifically about operation in Pre-Selected Load Control mode affecting factored fired hours--only the combustion mode.
As for what might be happening with the IBH valve, are you sure the tail is wagging the dog, or is the dog wagging the tail?
What Process- and Diagnostic Alarms are present on the unit when the load swings are occurring? (Please list all of them, even the ones you might not deem relevant.)
Thank you for your response. I downloaded the manual that you recommended and have started reading it. A lot of it is over my head at this point, but I will get there.
As far as TTRF1 goes, I understand that this is what causes the operational mode change, but I'm not sure how far we can lower our MW output before we get there. When we run the GT in preselect load mode, and are lowering MW, when the IGVs reach about 57 degrees, the IBH valve opens 20% and modulates from there. As we continue to lower MW, the bleed heat opens more and the IGVs close more. If I understand this correctly, this is designed to control the exhaust temperature, and also prevent the compressor from stalling. At baseload, our firing temperature is about 2065F, and when we lower MW to a point where the IGVs reach 57 degrees (about 67MW), the firing temperature is just over 2000F. If I am reading this manual correctly, I believe the GT enters into Lean-Lean at TTRF1=1955F. So how close is too close? I guess it's a matter of comfort level.
As far as IBH is concerned, we don't have any alarms generated when the IBH valve opens and the IGVs reach 57 degrees. It just happens. It appears as if the IBH is modulating to maintain the MW at the preselect load setpoint, but I'm not sure what it is basing it's movements on.
Also, with regards to your comment about using the RAISE LOAD/SPEED button, what does that command do to the operating mode of the GT? If I am in Base Load and I hit lower load, would that not take the turbine out of baseload operating mode and place it in preselect load?
Thanks again for your response. I do appreciate it.
Yes; GER-3620R is pretty chock full of egg-head, ivory tower stuff--and it's a "one size fits all" document, meaning that it's written to cover most GE-design heavy duty gas turbines, from Frame 5s to Frame 7
FAs and maybe even the H-class stuff by now. It can take some time to sift out what you need from all of the other information, but it's there. Don't despair; just take it in small chunks.
There is a relatively recent thread where someone, MIKEVI, I think, or maybe it was glenmorangie (two of our esteemed responders here on control.com) actually listed (if I remember correctly) the names of the applicable Control Constants, and the deadbands for the switching.
As the unit is loading at TTRF1 is increasing, it will get to a certain TTRF1 and then the unit will transition from Primary to Lean-Lean. Then as the unit is loaded further, and TTRF1 increases further, it will transition from Lean-Lean to Premix. (I think it's pretty much the same for DLN-I+ and DLN, which I'm much more familiar with.)
As the machine is unloaded, TTRF1 will decrease, but it has to decrease to a value LESS THAN the temperature at which it transferred into Lean-Lean from Premix in order to transfer from Premix back to Lean-Lean (that's usually called the "deadband" and it's usually something like 50 deg F, if I recall correctly). The deadband is there to prevent the unit from toggling back and forth between Premix and Lean-Lean during the transitions (loading and unloading).
That's the information the operators need to know--and which can easily be added to a HMI display so it can't be forgotten, right next to the value of TTRF1.
Again, I'm not sure if the tail is wagging the dog, or the dog is wagging the tail about the IBH/MW swings. I need to get to some drawings and then I tell you some signals to trend.
Pre-Selected Load Control, if yours is a typical unit, requires the operator to enable it with a couple of key-strokes, and it also requires the operator to enter a load setpoint (reference) for the function to adjust load to try to match/maintain. If the unit is at Base Load and you click on LOWER SPEED/LOAD, it's going to take several, perhaps many, clicks for the unit to actually switch from CPR-biased exhaust temperature control to Droop speed control, but AS SOON AS YOU click on LOWER (or RAISE-) SPEED/LOAD, you have canceled automatic maintenance of CPR-biased exhaust temperature control. At that point the Mark V is not going to adjust the fuel to try to maximize power output.
Once the status field indicates the unit is no longer in Exhaust Temperature Control (and load has actually started to decrease), it is in Droop speed control mode--NOT Pre-Selected Load Control. (I'm presuming the operator is ONLY using the LOWER SPEED/LOAD button to reduce load--and HAS NOT enabled Pre-Selected Load Control.) Contrary to popular (and FALSE) belief, when the unit is in Droop Speed Control with Pre-Selected Load Control disabled (inactive) the load WILL NOT drift around. Nobody will ever try it, but that's the way it works. I guarantee it. If the operator clicks on LOWER SPEED/LOAD until the MW meter reads 78 MW, for example, and then does nothing, it will stay at 78 MW--probably more stably than if Pre-Selected Load Control was active. Until the operator clicks on RAISE- or LOWER SPEED LOAD or until the grid frequency changes the unit will remain at approximately 78 MW indefinitely (presuming no one turns on the evaporative cooler or the fogger, etc.).
Just about EVERYONE thinks that Pre-Selected Load Control MUST be used to change load and maintain a stable load setpoint--but I can tell you that Pre-Selected Load Control is a fairly recent invention and units were operated for decades without Pre-Selected Load Control and they operated stably and without any problems. Pre-Selected Load Control is a convenient way to change load automatically, but one the desired load is achieved, one should click on RAISE- or LOWER SPEED/LOAD to cancel Pre-Selected Load Control and the unit will maintain the desired load, probably even more stably than if Pre-Selected Load Control was active (that's because Pre-Selected Load Control is rarely tuned properly during commissioning).
I am asking about any Process- and Diagnostic Alarms active BEFORE, DURING and AFTER you are experiencing the IBH swings. The IBH should be pretty stable; I just can't recall at this writing what the name of the IBH control valve reference is, and how to tell you to monitor it and the various references which feed the output signal to the IBH control valve to try to determine what might be the cause. But, if you're using Pre-Selected Load Control, and it's not tuned very well, that could be part of the problem. Just click on RAISE- or LOWER SPEED/LOAD to cancel Pre-Selected Load Control, and just be patient and watch and see what happens.
And write back to let us know!
I have a Control Spec. for a recent Frame 7. It is a mechanical drive, but the TTRF temps should be roughly the same. Drop me an email to firstname.lastname@example.org and I'll send you a copy.
Just got back from a job overseas. Now my brain is out of work and out of jet lag, I've been reviewing your problem again. A question, what is Plant MW Control and what is it doing? If IBH is modulating of course it will change MW as it is trying change Inlet Air Temp. but I see you are operating at 10F(-12C), IBH can cut in around 50F(5C) dependent on your humidity. I see you are using modulating IGV, normal for DLN to get you into Premix a bit earlier. It is possible that as you come to closed IGV at 57 Deg., your IBH is trying to keep adjusting as your IGVs close and your inlet air flows keep changing. Can you try and just reduce a little bit more and see if the IBH position steadies? Can you give us some values of TTRF as you reduce load to see if we can see when you are coming out of Premix?
Some check points L26FXL2=1 when reducing load is minimum Premix Transfer to Lean/Lean. L26FXP1=1 is transfer to Premix from Lean/Lean.
See if you can see by raising and lowering load when these points change. Try not to operate in this region, push load up to get into stable Premix, push load down to get into stable Lean/Lean.
OK for now, good luck and keep asking, we can fix it !!
Plant MW control is set at ~124MW for our plant. We have 1 gas turbine and 1 steam turbine. What this does, is, if the GT is putting out, say 100MW, then the ST will put out 24MW. It is a setting for the ST that adjusts the MW output of the ST to make the total plant MW output equal to whatever the setpoint is.
Our operating mode has been changed recently, we are normally a baseload plant. Due to market conditions, we have been dispatched lower than baseload. So, our operators are unfamiliar with anything other than operating at base load. We are using Preselect load and plant MW control to lower the output of the plant, slowly down to the requested total output. I hope that makes sense. Basically, for example, if we had been dispatched down to say 100MW, we would set the GT preselect load setpoint equal to the baseload point, and then take it out of base load, and into preselect load. Then we slowly back the preselect load setpoint down, while simultaneously lowering the plant MW control setpoint. We lower these two setpoints in proportion to the machine ratings such that we don't affect the rest of the cogen plant too much (i.e. duct burner pressure, extraction pressures, etc.) At the time of these changes, we never leave premixed DLN mode. I think from your previous response, you have assumed that we did.
As we lower the MWs on the GT, at about ~58 degrees IGV angle, the IBH valve opens (instantly) to 20% and modulates from there. (i.e. the further we lower MW, the further open the IBH valve goes). As far as TTRF1 goes, we usually sit at about 2065F at baseload, and when reducing MWs I have seen as low as about 2005F. So, I don't think we are in any danger of running into lean-lean at this point. I was just curious, since the IGVs are in temperature control mode as soon as you go to preselect load, won't the IGV and IBH settings prevent (or try and prevent) a change in the firing temperature, or is it exhaust temperature. (Sorry, new to GTs here)....
Based on CSAs response, I am still a little fuzzy on what mode of operation the turbine would be in if it were not on base load, and using the RAISE/LOWER SPEED command button. I basically assumed that button was to lower the preselect load setpoint without having to type an actual number into the setpoint box on the MK-V. I may need further insight on that.
I'm not sure if this is common knowledge or not, but once our turbine enters lean-lean, we have to substantially reduce load to get TTRF1 back down around 1850ish? I think, to get the interlock to clear to allow us to bring the GT back into premixed mode.
Thanks for the help. I am off shift currently, but I will keep checking back and responding when possible.
If you and your operators are used to operating at Base Load or Temperature Control mode, then I understand they might not be familiar with what is happening at this point. And there is a lot of terminology to digest and understand. Basically the machine has to be in some sort of control mode or limit.
Base Load AKA Temperature control. In this mode the control system is admitting as much fuel as possible to generate power. This limit is based on the maximum calculated firing temperature for the turbine hardware installed in the machine. In your case your TTRF1 limit for your hardware is 2065 degf.
Typically any load less than base is referred to as part load although you may also see speed control. Looking at your screen for FSR(Fuel Stroke Reference) you should see bar graphs that represent different FSR values, one of them will always be in control.
TTRF1 is the calculated firing temperature at the 1st stage turbine nozzles. As fuel is reduced, and mwatt output falls the TTRF1 will decrease since less fuel means less heat. Because you are operating in combined cycle mode and the IGV's are in temperature control mode (For a DLN machine they HAVE to be in that mode) they will close to try and maintain a target temperature for the exhaust. This has to do with controlling the amount of airflow through the machine and into the combustors.
As discussed in my previous post at the point the IGV's get near 58 deg angle the IBH will pop open to a value around 20% open. As the IGV's close further the IBH will open more, its a fairly simple (well not really) inverse relationship. But the basic idea is to reduce the amount of air going through the compressor. All of this has nothing to do with preselected load mode. Its all about the machine load being reduced below a point that if the IGV's did not close and the IBH did not open that there would be too much air going through the turbine and the mixture in the combustion cans would get so lean the flame would be blown out.
For the operator they typically are worried about keeping the TTRF1 number above the transfer point(usually 1950 degf) that would allow the unit to "fall out" of premix. Basically to keep above this point is just keeping the load high enough. Most operators try to keep some margin, so moving load to keep TTRF1 above 1975 degf is usually pretty safe.
If the machine falls out of premix, or enters lean-lean mode then you are correct that you have to substantially reduce load (around 32mw depending on outside temperature)(usually TTRF1 less than 1900degf) to reset certain permissive before you can load the unit back up and get into premix. GE offers a software package called a High Load Premix Transfer (HLPT) that allows the unit to transfer back into premix at loads very near base load. This software package also has some hardware requirements depending on fuel type, valve arrangements etc.
Hope this helps. I could discuss this with you further since all this typing ............ But hopefully some of this discussion will help the next guy asking questions like yourself.
When you enable Pre-Selected Load Control the Mark V WILL NOT jump the fuel to immediately make the actual load equal to set point, so if the unit is operating at 85 MW at Base Load and you want to lower the load to 80 MW just change the Pre-Selected Load Control setpoint to 80 MW (from whatever it is currently is) and then enable Pre-Selected Load Control and the Mark V will start lowering the turbine speed reference until FSRN gets less than FSRT at which point load will start dropping below 85 MW and then stop at 80 MW. (If the Pre-Selected Load Control setpoint was 10 MW and the unit was at 85 MW at Base Load it will take a short time for the Pre-Selected Load Control command (which is different from the setpoint!) to ramp up to 80 MW so best to wait a couple of minutes for the setpoint to ramp to near the command value before enabling Pre-Selected Load Control or the load may drop below 80 MW and then increase back to 80 MW.)
The point is: changing the Pre-Selected Load Control setpoint to be equal to the current load at Base Load and then enabling Pre-Selected Load Control isn't doing anything, except wasting time and mouse clicks. If your operators are in to that (as many are) have at it. But it shows a fundamental lack of training and experience with the Mark V.
As for the mode of operation of the GT when it's between 0 MW and Base Load (with the generator breaker closed) that's called Droop speed control. (I'd absolutely pay money to hear your operators try to explain Droop speed control to you. And I'd also bet money everyone of them will have a VERY different answer, too.) It's not as difficult as their explanations make it sound, but it is fundamental to AC power generation and b with the exception of combined cycle steam turbines it's how EVERY generator prime mover governor operates when synchronized to a grid of any size. And yet you will be VERY hard-pressed to find anyone who can explain how it works or even what it does. It's so basic it's like opening a faucet or turning on a light--no one thinks about all that goes on in the background to make the water flow or the light illuminate.
But, I digress. Clicking on RAISE-or LOWER SPEEED/LOAD changes the turbine speed reference which changes TNR in the Mark V which changes FSR (Fuel Stroke Reference) which changes the fuel flow-rate which changes the load (MW). Clicking on RAISE or LOWER SPEED/LOAD changes load, yes, but it doesn't change a load setpoint it changes a speed setpoint. Isn't this fun?!?!?!
Now to REALLY blow your mind, Pre-Selected Load Control is a "modifier" to Droop speed control! (By now you think I'm joshing you, but I don't josh about Droop speed control, I assure you, as will most readers of the GE-design heavy duty gas turbine control threads here on control.com.)
Pre-Selected Load Control is a lazy man's way of operating a GE-design heavy duty gas turbine, and I predict it's not going to be long before NERC and other power generation regulatory bodies outlaw it's use in it's present form because in it's present form it causes GE-design heavy duty gas turbines to react completely opposite of how they should react to a grid frequency disturbance.
DO NOT allow the operators' careless attitude with respect to Process- and Diagnostic Alarms to infect and poison your response and attitude towards them! Please tell us ALL GT Process- and Diagnostic Alarms that are active when the unit is running when the IBH control valve and the MWs are swinging. There may be a condition that the operators are overlooking (read: ignoring) that we can point to to quickly help with this issue.
Also, is there anyone at the plant that knows how to trend signals? If I recall correctly the signal name for the IBH control valve actuator is CSRIH or CSRIBH or something like that. There should be a display for either IBH or IGV that shows the IBH control valve actuator output signal value. While the display update rate is not very fast it might show a fluctuating signal, or if the signal is stable it might indicate a problem with the IBH control valve actuator (sometimes called a positioner). What we need is an idea of whether the instability is being caused by the Mark V or some other component.
Finally, DO NOT believe people when they try to blame the Mark V for every problem (real or perceived). It's a good control system which is much maligned, poorly documented and not obsolete. Here at control.com we have solved MANY problems falsely attributed to the Mark* (IV, V, VI, and VIe). BUT, to do that we need actionable data. And we need to know what Process- and Diagnostic Alarms are active. Help us help you. It's worth the effort. Tell us what alarms are active and what the output signal to the IBH control valve actuator is doing and we can start troubleshooting the load instability that is occurring when you say the IBH control valve is opening as load the IGVs are closing below 58 DGA. Also it would be helpful to know what the Gas Control Valve is doing at this time, as well as the Splitter Valve and the Stop-Ratio Valve (signal names FSG, FSGT and FSGR).
You're new, but let's focus on the problem, and you can ask other questions as time goes on. Get us some data!
This may help you with your understanding of IBH and IGV interaction.
IBH is a VERY poor name for the function that it provides--as MIKEVI says, it's purpose is to reduce air flow through the compressor when the IGVs are closed below 57 DGA when the axial compressor is at rated speed. They open a degree or too early (when unloading), and stay open a few degrees later (when loading). This protects the compressor against stalling/surging.
The reference for IBH control valve position is a percentage of air flow through the axial compressor when the IGVs are at low angles at operating speed. This is all programmed into the Mark* via Control
Constants, so when the IBH is commanded to go to, say 20% stroke, that corresponds to a percentage of air flow for the given IGV angle.
To my way of thinking if the IBH is moving because of a problem with the output reference coming from the Mark V then there's a problem with some input to that reference calculation--either the IGVs are also moving (unstable), or there's some problem with the cards that are producing the output signal to the IBH control valve, which should be accompanied by a Diagnostic Alarm, or two.
If the output signal is stable but the IBH is moving, then there's a wiring problem between the Mark V and the IBH control valve actuator (positioner) or some problem with the actuator itself. Usually, the IBH actuator is powered by instrument air. Is the instrument air supply steady and at rated pressure? There is usually a quick exhaust valve on the air supply to the actuator, and sometimes it's been known to have problems--especially if the air is not dry. And sometimes there is a local air pressure regulator which fails (again, usually because of moisture in the air).
All of these problems with the IBH control valve actuator and air supply can seem to be caused by that danged Mark V, when in reality the Mark V is doing it's thing properly. It seems that the Mark V, with all that wiring, and all those LEDs and printed circuit cards, well, it just must be the root of all problmems, right? But, it's really a pretty darn good control system with lots of years left in it. And, it will usually tell you--usually with Diagnostic Alarms--if something's amiss with the hardware or software, occasionally with Process Alarms.
I know--the Mark V annunciates a LOT of alarms. And most of them are probably thought to be nuisance alarms. Usually operators just ignore alarms until the unit trips or shuts down, and then they don't know how to read the Alarm Log to determine what caused the trip or shutdown. They believe their only job is to silence alarms and click on targets to keep the unit running. Not true, but widely believed. Again, if you're new to power generation, don't pick up that bad attitude. EVERY alarm means something--and if it isn't a true condition then the technicians need to fix whatever's causing the alarm so it doesn't get annunciated. But, because the turbine control system is so poorly documented and most people don't get good training or because the Mark* is pretty darn reliable and doesn't fail very often that the training they do receive is forgotten over time and can't be recalled when needed. Actually, the reliability of the Mark* is a double-edged sword. And the complexity and lack of documentation is a problem for most sites.
I don't know if this will help your understanding and troubleshooting or not. Again, the more actionable data you can provide us (instead of anecdotal data) the better and more concise our replies can be. We're just throwing a lot of things out there right now--and concentrating on too many other things (Pre-Selected Load Control and Droop speed control for example).
Again get back with answers to our questions and some good, actionable data. And let us know how you are faring with the troubleshooting and resolution.
I don't have a huge amount of time to write as we are in an outage so I will just write some quick bullets to your questions.
For DLN1 and DLN1+ there are several modes of combustion that are all controlled by TTRF1 which is the calculated temperature of the gases as they reach the 1st stage turbine nozzle. This is a calculation that uses various turbine inputs to estimate mass flow through the machine and correlate temperature at the 1st stage nozzles to exhaust temperature.
Primary mode-from lightoff until TTRF1 reaches ~1600 degf. This should be the control constant(FXKTCS1)but without seeing your logic I can't be 100% sure of the signal name or exact value. These items should be in your control spec section 5.02.23 DLN Combustion ref Temp Comparators. Fuel is admitted and flame is occurring in the primary zone of the combustion liner. High Nox is produced during this mode of combustion, but the flame is very stable and controlled.
Lean Lean mode from 1600 degf to 1950 degf (FXKTS1) TTRF1 temperature. Fuel and flame exist in the primary and secondary areas of the combustion liner. This mode of combustion also creates large amounts of Nox and is typically not the desired mode to run in for long periods of time.
Premix transfer to premix steady state. I could write a huge article on this to properly describe the sequence. But basically fuel is transitioned from the primary nozzles to the transfer nozzle that is out in the secondary area of the combustion liner. Once fuel is reduced in the primary zone the flame will be blown out. Once the primary fuel valve is closed and flame is verified as out in the primary zone, then the reverse will occur where fuel is transferred back to the primary nozzles but with no flame in that area. Fuel is now being admitted in the primary zone and allowed to premix with the air, and then as it enters the secondary area of the combustion liner it will ignite and burn with fuel that is also being admitted from the secondary nozzle. Once certain conditions are met the unit will indicate it is operating in premix steady state. It should stay in this premix mode as long as load is kept high enough to keep the TTRF1 number above the transition point of 1950degf.
Extended lean-lean mode can be manually selected or may occur if there is an auto re-ignition of the primary zone at high load. This mode is not desirable as it produces large amounts of Nox emissions and also is a multiplier to fired hours that reduces your time between combustion hardware maintenance cycles.
Pre-selected load and base load have no effect on maintenance factors. The unit is designed to operate at base load as this is the most efficient operational area for the machine. Operating in extended lean-lean mode does count against you in maintenance time.
The purpose of inlet bleed heat is to allow more turndown of the machine, or allow a wider operating range. It was not originally intended as an inlet heater or anti-icing as some people might think. As the machine is unloaded and less fuel is admitted for combustion the mixture in the combustion liner becomes more lean. To control airflow the IGV's are modulated closed to reduce air into the machine. At around 58 deg angle the inlet bleed heat valve will open to further reduce mass flow through the machine and keep flame in the combustor from blowing out, and also to prevent a large rise in CO production. The inlet bleed heat reduces mass flow in 2 ways. One is raising the temperature of the inlet and making the compressor less efficient. Secondly by robbing compressor air and recycling it back into the inlet. As a byproduct it does heat the inlet and can help prevent inlet icing which may occur as the IGV's go further closed and cause a drop in pressure and temperature. But its real purpose is to allow the unit to operate in premix mode to a lower mwatt load than a unit without IBH which can usually only operate in premix down to ~80% load, vs a unit with IBH can operate down to ~60% in premix. The bleed heat should not cause the unit to "swing". If something like that is occurring then the control of the IBH should be investigated.
I hope this helps. I need to get back to our outage.