NFinch,

",... one [operator] mentioned they had reignited a flame-out [of a Frame 6B unit] in the past. ..." That's a pretty wide open statement. Did the re-ignition occur while the unit was at or near rated speed? How did they do that, exactly? Was it intentional? If so, it probably required a good bit of logic forcing and a super-size amount of bravado and luck. Did they accomplish it by clicking on the START button on the HMI (or twisting the START switch handle on the old control panel? What speed was it that the unit re-ignited at?

GE-design Frame 6B heavy duty gas turbines are pretty robust machines. They can take a licking and keep on ticking--usually. The fuel flow-rate of a Frame 6B at FSNL (Full Speed-No Load) is usually around 20% of rated, sometimes a little higher, sometimes a little lower. I would agree that one could calculate the electrical load based on fuel flow-rate and heat content of the fuel being burned.

Some heavy duty gas turbines are purged before ignition at a higher rotational speed because the purging time is shorter than it would be if the unit purged at or near "firing speed." When this happens, the speed is allowed to decrease after the purge time is complete before the ignitors are energized and the fuel is admitted to the combustors. This is to lessen the thermal stress on the combustors and hot gas path hardware versus having to put extra fuel into the machine when it's rotating faster--which would be required to establish flame and would result in a higher spike in internal and exhaust temperatures, which reduces parts life and reliability.

If the unit being discussed flamed-out while at rated speed and suddenly re-ignited (because of some ignition source in the combustor OTHER THAN an ignitor) that would mean that fuel was still flowing after the loss of flame (BAD!!!), and if the ignition occurred in the exhaust, THAT would be even worse and harder on the exhaust plenum, exhaust duct and stack, and the flame would have to travel in reverse through the turbine section and back into the combustor.

So, there's a LOT we don't know about this "incident" and even more important what the effect on the hot gas path and exhaust was.

I didn't quite understand the previous reference to OCGT (Open Cycle Gas Turbine???), but if it referred to a gas turbine that was not exhausting into an HRSG (Heat Recovery Steam Generator (a "boiler")) that's often referred to as Simple Cycle (as opposed to Combined Cycle--when the GT exhaust heat produces steam in an HRSG). Why do the units in your idea have to be re-started at synchronous speed? Do the units at your site(s) run that often, or are they necessary evils when solar production goes down in the evening until other generator can be brought on line?

I think this idea/concept is a short-term solution to a problem that is, or possibly should be, solved by storage. It could be pumped storage, or it could be compressed air storage, or it could be batteries.

There are many Combined Cycle power plants which have sea containers filled with batteries that are being used for energy storage and grid frequency control, right now today. Excess generation is used to "top up" the batteries, and when the batteries aren't being used for grid frequency control the output of the turbine-generators is simply reduced. I realize this isn't exactly the same as the problem you're trying to solve, but by storing and releasing the energy in response to solar cycles this seems to be a possible solution.

Does it make sense to try this idea/concept and invest money into research and a control system and possibly incur some unintended and possibly catastrophic damage to what you admit is old equipment (which might be somewhat "expendable")? (Peaking generation is pretty valuable right now, though.) Or to invest in storage capability, which can be used for many years, and is the way the world has to go with the increased reliance on renewal generation which has it's ups and downs--which can't be matched with load at times? (Remember the Texas scenario where wind generators were being paid to shut down a couple of years ago?)

It's an interesting, and innovative, idea/concept. It has risk, which storage technology also has but not as much. But, eventually, isn't storage going to be the way to solve the problem of excess generation and/or low demand/load?

I haven't mentioned this is not a very green solution to a real problem. About the same thing cold be accomplished by using large grids of reinforcing bars (rebar) welded together and submerged in brine. Applying electrical current to the rebar would result in boiling the water. But, this, too, is wasteful; the water usually evaporates into the atmosphere and has to be replaced with "new" water (but we are experiencing sea level rise for some strange reason.... but not climate change according to many non-scientists). (There could be a recuperative aspect to this, though.) This is a very low risk solution, and not really expensive to construct or operate or maintain.

 

NFinch,

I admit, I had to think about this long and hard, but with the statement, "... The control system could be adapted to periodically keep firing the ignitors as a protection against flame-out, otherwise the Reverse Power protection can protect against over-loading the generator. ..." you seem to be saying now that you would try to keep flame on while running the unit in reverse power.

GE-design heavy duty gas turbine control systems use reverse power detection to open the generator breaker under almost EVERY shutdown or trip condition (emergency shutdown). That's because the philosophy is to ensure the unit does not overspeed when the generator breaker is opened. (There is fuel, under pressure, trapped between the fuel stop valve and control valves and the fuel nozzles in each combustor which has to dissipate when the fuel stop valve is shut, either during a normal shutdown or a trip. If, when a trip condition is detected the fuel stop valve is shut at the same instant (there are very slight time differences) as the generator breaker is opened the fuel trapped between the stop/control valves and the fuel nozzles/combustors has to continue to flow into the combustors until the pressure of the fuel is less than the pressure in the combustors. (The fuel stop valves are prior to the fuel nozzles, sometimes by many feet, and the volume of the piping/tubing also factors into the situation.) So, what GE's control philosophy does is shut the fuel stop valve/control valves, allowing the fuel between the valves and the fuel nozzles/combustors to continue to flow into the combustors and burn while the amount/pressure is quickly decreasing. This means the power being produced by the turbine is decreasing VERY quickly, and going into reverse power--which the turbine control system and the reverse power relay detects and opens the breaker. This ensures the unit does not overspeed during a shutdown/trip because of the uncontrolled fuel flow of the fuel which continues to flow into the combustors even though the stop/control valves have closed. The grid prevents the overspeed, and as the amount of torque being produced decreases as the fuel flow-rate decreases the power output of the generator drops below zer0--while the unit is still at synchronous speed!--and the reverse power sensing opens the breaker.

It's probably possible to use another set of reverse power relays to detect a higher reverse power, and to use the turbine control system to modulate the fuel flow-rate to actually control the magnitude of reverse power. But--in my personal opinion--firing the ignitors to re-ignite the combustors on loss of flame at synchronous speed is very dangerous. On many GE-design heavy duty gas turbines without DLN (Dry Low NOx) combustors the ignitors (called "spark plugs") actually retract, pulling the tip of the ignitor (spark plug) out of the combustor to prevent the intense heat of the flame ball from damaging the ignitor. Those with DLN combustors use a different, high-energy ignitor which does not retract but is susceptible to damage from flame ball heat over time.

AND, this is an even less green solution--burning fossil fuel while using the gas turbine unit as a "brake" on the grid because of over-generation of solar power at times.

I don't know.... I'm sure someone will try something like this at some point, but the "crab pot" idea may be the simplest and least expensive to implement, except for the water thing. It doesn't have to be boiler quality (demineralized) water in the tanks being heated by the rebar; it can be brackish water from nearby lagoons, or even treated water from a nearby water treatment plant (since studies show most people won't drink even highly purified, treated water from a water treatment plant!). I don't know how much, or how complicated, a recuperation scheme would cost to build and operate and maintain, either.

Another thought would be to use the surplus power to drive a water treatment plant to provide water to communities or even farms. Clean water is going to be more expensive than fuel (gasoline or diesel) very soon, if it isn't already in some places. And, desalination plants or something similar could be used to produce clean water with surplus solar power (renewable power--including wind!).

Anyway, them's my thoughts--worth about 3.5 cents if anyone's interested. I'm a conservative guy when it comes to combustion turbines and I don't think this idea/concept is fully fleshed out, and likely has a lot of issues to overcome, including conservation issues.

GREAT DISCUSSION, though!!!

 

Hi CSA,

You've given me a lot to respond to! I've messaged the operator involved in the flame-out to try and find out further information. It was mentioned by another operator who could have confused it with some other long past event, I'll try to find out so we can address some of those points above.

OCGT - correct Open Cycle Gas Turbine or 'simple cycle'. OCGT is the more commonly used term in my neck of the woods.

So when I was talking about re-starting the units at synchronous speed, it was in relation to the operational profile I thought would best be used to manage the high rooftop PV generation. I had thought to start a GT normally for a morning generation peak, slowly ramp down to 'negative generation' during daytime high PV generation (running as a block load with IGV control), then as the sun sets in the afternoon you'd re-ignite combustion and ramp up to meet evening demand.

You're definitely correct that storage projects are the long term solution, but in our system I can't see them catching up for at least 3-5y. Most new storage projects coming online are also 1h - 2h systems, this operation would of course have the flexibility of continual load operation. A battery would become 'charged' after some time and could no longer be a load. This to me a really a possible short term solution to extend the life of existing assets and generally it shouldn't be too capital intensive to trial. The cost to implement this operation I envisage would be mostly modelling, testing, control/prot changes and regulatory such as connection agreement. The timeline for implementation would also be short as the assets already physically exist.

From my perspective there is actually an environmental benefit to this however I have not got the tools at my disposal to verify it with system modelling. My thought process is that even though you are turning say X MW of Solar power into heat, because you are doing so with a synchronised generator, you are also providing inertia/fault current/voltage control/freq control to the power system, thus enabling the utilization of my low-cost asynchronous generation. The net effect being that X + Y MW of PV generation can be online, displacing other higher cost thermal generation assets. Every day of the week my preference is for maximising renewable generation, I just see this as a possible method to further reduce reliance on thermal generation and a possible method to avoid building more 'energy transition bridging' gas turbine generation.

Ideally from a fuel use perspective I would prefer for this operation to have no flame. What I'm unsure about is a) the suitability of controlling on IGVs, and b) whether you can do a smooth load transition from FSNL to zero combustion or if there would be a large jump during flame out.

With a digital protection relay it would not be difficult to have two reverse power settings. Firstly a normal setting that detects any amount of reverse power flow and trips the GCB upon detection. I envisage this would then be inhibited by a digital input from the control system when in 'negative generation' mode and the device switched to a secondary reverse power setting much lower (or completely removed). It seems to me this isn't the critical challenge, more what to do when your flame is out and you want to run-up again. I had thought re-ignition and ramping back up to standard generation would be the way to go, but you're suggesting the spark plugs may not be able to handle this?

Practically speaking the surplus power would definitely be best used running things like desal plants, CO2 capture algae farms, or even large carbon-capture plants. I guess all these take time and cost to properly implement, and this whole concept is more just a niche opportunity that could form part of the broader changes to energy management.

Cheers
Nick

 

NFinch,

Again, personally, I wouldn't try using the generator to spin the turbine without flame--but, also, again, I am very conservative when it comes to gas turbines and auxiliary equipment.

If the generator breaker isn't opened when the fuel is shut off, the speed isn't going to change.... I would imagine the change in current direction and magnitude would be big and rather quick. It would be very important to keep the excitation running, because motoring the generator and losing excitation would cause very quick heating of the generator rotor and stator windings (due to the disruption of magnetic fields as the reactive current changes).

And, if the unit this is being tried on used a reduction gear between the turbine and the generator it would be a good idea to have someone review the gearbox set-up to make sure it is capable of reverse operation (the generator driving the turbine instead of the turbine driving the generator).

Personally, if possible I would try running the unit with the generator breaker closed at zero load, and then start reducing the fuel flow slowly to try to find a load that would keep the flame on while operating in reverse power. That's going to be the hard part, because it's very likely the unit will lose flame more than once when trying this. And, it may be that there is not much margin between zero load and flame-out. I would NOT try re-igniting the flame while at rated speed--just too risky (in my personal opinion). And, the design of the ignitors (spark plugs) is not known.

Anyway, it's an interesting idea/concept. Finding someone who can manipulate the control system and make any necessary control logic and protection changes may also be difficult.

Keep us informed! And best of luck!