Purge diaphragm valves

Good day All,
Please can anyone enlighten me on the functions and operational principle of the purge diaphragm valves in the frame 9e gas turbine DLN compartment.
There are two of them standing beside each other.
Thank you.
 
Sirtaurus12,

The first two places you should begin your search for information is by referring to the Gas Fuel and Fuel Purge P&IDs for the unit(s) at your site, and then the Gas Fuel and/or Fuel Purge section of the Operations & Maintenance Manuals provided with the equipment by the manufacturer or packager of the equipment.

I PRESUME you are referring to two air-operated plug valves (the air acts on large diaphragms to open the valves)) which are in the purge air line from the compressor discharge casing to the secondary or transfer gas fuel lines. (Not being able to see the P&IDs for your site, I don't know if the secondary fuel manifold/nozzles or the transfer fuel manifold/nozzles are being purged. GE Belfort seems to have eliminated either the secondary gas fuel purge valves or the transfer gas fuel purge valves; I think it's the transfer gas fuel purge valves they eliminated--but the P&IDs for the unit(s) at your site will tell you for certain which manifold/set of nozzles are being purged. In any case, the operation is identical; just the manifold/fuel nozzles being purged changes.) These valves are arranged "back to back" meaning one valve is specifically installed to block flow in one direction and the other valve is specifically installed to block flow in the other direction. There is usually another small solenoid valve on the area between the two valves, which serves to vent the pressure between the two valves when gas fuel is flowing in the secondary or transfer gas lines. Finally, there is usually a pressure switch, which monitors the pressure in the area between the two valves and the vent solenoid valve, to detect a possible leak of one or both valves. Another way these two large pneumatic valves are referred to is 'double block and bleed' valves. There are also usually limit switches (two) on each of the valve assemblies to indicate whether the valves are fully open or fully closed (or somewhere in between, in a way).

When there is no gas fuel flowing in the secondary or transfer fuel lines these two valves should both be open allowing purge air from the compressor discharge to flow into the secondary or transfer fuel lines to, one, purge any fuel out of the lines, and, two, to prevent the backflow of hot combustions gases through the secondary or transfer fuel nozzles into the respective manifold around the axial compressor casing and out of another fuel nozzle in the same system. The solenoid vent valve closed, and there is pressure (CPD pressure) in the area between the valves.

When gas fuel IS flowing in the secondary or transfer lines the two pneumatic valves are both closed, and the solenoid vent valve between them is open, and there should be NO pressure between the valves. Because they are back to back, CPD on top of the upstream plug valve is forcing the plug to be closed, along with the spring of the valve actuator. Similarly, when gas fuel is flowing in the secondary or transfer lines the downstream valve's plug has gas fuel pressure on top of the plug to force the valve to the closed position, along with the actuator spring. If the pressure switch detects any pressure between the valves--that indicates one or both of the valves is leaking and because gas fuel pressure is always higher than CPD that would mean that gas fuel could flow into the compressor discharge area. Which could be disastrous as you could imagine.

The two valves you mentioned are not technically diaphragm valves, they are plug valves and the operator mechanism uses air pressure acting on a large diaphragm to open the valves, and a spring on the valves serves to try to keep them closed (fail closed, if you like that terminology). Again, the back to back arrangement of the two valves serves to utilize the two pressures (CPD and gas fuel) to help keep the valves closed by the pressure acting on the top of the plug (the plug rises to open). Pressure on the bottom of the valve plug can sometimes cause the plug to rise--open--and that's not good.

Hope this helps! The appropriate section of the Operations & Maintenance Manual should sort of explain this, but probably not in such detail. By using the P&IDs, it should be clearer how the valves serve to block flow and prevent backflow.

If you don't have your own copies of all of the P&IDs for the unit(s), you should make copies (as large as possible), and make notes on them as you study them or use them for troubleshooting. For, example, you can put the pressure switch setting from the Device Summary (another REALLY IMPORTANT document to add to your collection--whether you are an operator, or a I&E or I&C technician, or a maintenance technician, or an operations supervisor, or even a plant manager). That way, when you look at the P&IDs in the future, you can quickly know when the pressure switch will change state. This is an EXCELLENT way to get familiar with the devices and instruments on each system of the unit. And, then, when you have the time, and the unit is not running, you can go out to the various compartments and locate the devices--that way, when there's a possible problem with one of them you will know exactly where it is!!! There really is NO BETTER WAY to get to know GE-design heavy duty gas turbines and auxiliaries. None. Training won't do it--it will only expose you to the drawings and documents and some of the theory and logic. If you want to learn this stuff, you have to do your "homework" and that means studying the drawings, and finding the devices out on the unit.

If you have questions, you can aske them here. But, avail yourself of the P&IDs, and a copy of the Device Summary, and you will be far ahead of most people. Studying them and locating the devices out on the unit will make you a respected employee (more respected!).
 
CSA,

I can't thank you enough for this detailed explanation. When I was typing the above subject matter, I prayed for you to be amongst the persons to bless with with explanations and you did, thank you so very much and I always wish you well.


But then I have another question;
Do the plug valves or pressure switch(63pg) have anything to do with the unit going into extended lean lean every now and then, especially when there are frequency fluctuations on the grid and the unit is set on pre-select and frequency control?
 
Sirtaurus12,

This is going to be difficult to answer, but I'm going to try.

If the problem were related to the purge valves and purge valve limit switches and the purge pressure switch there would be Process Alarms to that effect. So, unless you are seeing Process Alarms related to the purge valves, they wouldn't be causing the unit to transfer from Premix Steady-State to Extended Lean-Lean.

Frequency fluctuations on a DLN-I machine are NOT good. Not good at all. When frequency is fluctuating--especially if it is fluctuating very quickly--can drive a DLN-I machine to go crazy. If you are operating at Part Load with Pre-Selected Load Control AND "Primary Frequency Response" active AND the Pre-Selected Load Reference is close to the TTRF1 where the unit would transfer back out of Premix and back to Extended Lean-Lean, it's not going to be pretty (I'm referring to unit operation).

DLN-I is NOT a good machine to try to operate in Premix Steady State when the grid it is synchronized to is experiencing frequency fluctuations. It doesn't require much of a bobble in EITHER fuel flow OR air flow to result in a primary combustion zone re-ignition when operating in Premix Steady State--which will cause the Mark* to switch the combustion mode to Extended Lean-Lean. When the frequency is fluctuating, the speed of the generator and turbine--AND the axial compressor--are also fluctuating. So, that affects air flow (because the axial compressor speed is fluctuating). AND, because turbine speed is fluctuating that affects the P2 pressure reference, which means the SRV (Stop-Ratio Valve) is fluctuating to try to control the P2 pressure to maintain the P2 pressure reference--which is changing because the turbine-generator speed is fluctuating. It's just not pretty.

And, if the Pre-Selected Load Control reference is such that the unit is operating with a TTRF1 not too much above the temperature at which the unit would automatically transfer back to Lean-Lean from Premix Steady State, and the frequency (speed) is fluctuating--then the DLN logic is going to have a hard time trying to figure out what to do.

To complicate matters, if the TTRF1 is fluctuating (and it will be!) a lot the DLN logic may also be trying periodically to re-initiate a Lean-Lean to Premix Steady State combustion mode transfer. When THAT happens, the DLN logic "grabs" the governor control and puts a constant RAISE on during the attempted transfer. This, along with speed fluctuations, and Primary Frequency Response and Pre-Selected Load Control is just a recipe for complete confusion for the Mark*. It can get downright ugly, and lead to a trip.

When the frequency fluctuations start it's probably just best to transfer the unit manually to Extended Lean-Lean and just leave it in Extended Lean-Lean combustion mode. Yes; the NOx emissions will be high. Yes; the load will still fluctuate. But by manually selecting Extended Lean-Lean the DLN logic in the Mark* will most likely not be trying to transfer the unit back into Premix Steady State if the TTRF1 drops low enough to re-initiate the transfer to Premix Steady State.

I would also cancel/abort Pre-Selected Load Control (which should cancel Primary Frequency Response also). Now that sounds worse than it is--but Droop Speed Control is going to do exactly what Primary Frequency Response does, without any interference from Pre-Selected Load Control. IT'S IMPOSSIBLE TO MAKE YOUR TURBINE'S LOAD BE CONSTANT WHEN THE GRID FREQUENCY IS FLUCTUATING. IMPOSSIBLE!!! That's NOT even supposed to happen. For the grid to have a chance at stabilizing, it's necessary for ALL units to respond to frequency fluctuations by changing load. That's one of the functions of Droop Speed Control--to try to help stabilize a grid when there are frequency fluctuations.

If your unit load was stable while the grid frequency was unstable and changing, that would mean that your unit would be generating power at a stable frequency while the frequency of the grid was unstable. And, that just CAN'T happen. No matter how much ANYONE wants it to happen. Not the operators; not the operations supervisors; not the plant manager; not the President, the Prime Minister, the Emir, or the King or Queen. It's against the laws of physics as we know them today, and as they have been known for more than 100+ years. It's just not possible for one unit to be generating 50.09 Hz and another unit to be generating 49.7 Hz, and another unit to be generating 51.2 Hz, and another unit to be generating 49.5 Hz--and to have 50.0 Hz coming out of the socket on the wall, or feeding a large pump in a water treatment plant or a cement plant or a refinery. There's no frequency smoothing device anywhere in the world to do that--have a bunch of generators generating at very different frequencies and have a smooth frequency at the wall socket or the Motor Control Center for customers.

During frequency fluctuations, if they last very long (how long do they last, typically, at your location?), it's not going to be possible to get the unit back into Premix Steady State combustion mode anyway. If you need the power in the range where Premix Steady State would normally be in operation it's just best to leave the unit in Extended Lean-Lean until the frequency stabilizes. And, then unload the unit to get it to the point that it can be transferred back into Premix Steady State (or press that magic button to transfer without changing load--I hear there is such a thing these days!).

DON'T run a unit in Extended Lean-Lean for long periods of time (except in an emergency) when the grid frequency is stable. The OEM says one hour of operation in Extended Lean-Lean combustion mode is the same as TEN (10) hours of operation in Premix Steady State combustion mode. It's VERY hard on the combustion hardware. There's not much choice if you want to keep the unit running when the frequency is fluctuating, but Extended Lean-Lean is only meant to be a temporary mode of operation until the unit can be transferred back into Premix Steady State--or shut down if can't be (transferred back into Premix Steady State for some reason).

But, unless there are Process Alarms about the purge valves being out of position or the pressure between them being high when it should not be, it's NOT likely these valves are causing Extended Lean-Lean transfers, especially during frequency fluctuations.

I'm going to hit the max character limit--and this thread wasn't originally about Droop Speed Control.

Hope this helps!
 
CSA, can you or anyone please enlighten me on the working principle of the servo (MOOG) valve for SRV or IGV.
Since it has 6 wires coming out of it, I used to believe it has 3 coils inside it, but I happened to open up a faulty one and discovered there are 2 coils in it.
This got me confused the more.
Thank you.
 
Sirtaurus12,

This really should be a separate, unique thread....

GE used two-coil servos (for older turbine control systems--Mark I Speedtronic; Mark II Speedtronic) AND for SIMPLEX Mark IV (which was an upgrade product--never provided on a new unit), and for SIMPLEX Mark V (which was also an upgrade product). AND, many times if the unit had a TMR control system, and was being upgraded to a SIMPLEX control system, the third coil was just abandoned (to prevent having to replace the three-coil servo with a two-coil servo, thereby reducing cost).

It would be necessary to know what control systems have been used over the life of the unit(s) at the site where you are working to say much more, and to say for sure if the servos are appropriate for the control system presently in use on the unit(s) at your site.

If the present control systems are SIMPLEX or DUAL REDUNDANT, they can use two-coil servos, or three-coil servos (with one not in service). In this case, the amount of null bias current has to be changed if only two of three coils are used, but that's about all. (I believe if three-coil servos are used where two-coil servos would be adequate, the third coil should be shorted at the junction box closest to the servo to prevent nuisance and intermittent problems.

If your future questions are totally unrelated to an existing thread, please open a new thread.

Hope this helps!!!
 
My apologies for posting this topic on this thread, I desperately needed your attention on this.
Thank you very much for your usual and swift explanations.
 
Top