All, I have a question regarding an Open Cycle, Frame 9E.03 machine running on Natural Gas. Assuming all the auxiliaries are in service and the GT is ready to start, what is the time take from barring to Synchronous Speed?, i.e. you press 'GO' roughly how long until the circuit breaker closes?
I am used to Siemens 260MW GT's and these took around 7 Mins irrespective of how hot they were, so am expecting similar values.

 

What is the starting means for the 9E.03? Is it an induction electric motor, or ???

I've only worked with GE-design Frame 9E machines which used induction electric motors as the starting means, through a torque converter (to vary speed). These machines used a torque assist from the induction electric motor starting means from zero speed/barring gear operation to 60% of machine speed (1800 RPM), and the remainder of the acceleration to rated speed just prior to synchronization (3000 RPM). All of these machines must have some kind of purge of the entire combustion section and exhaust--and many of these machines exhausted into an HRSG (boiler). The purge is generally two to three times the total volume of combustion section and exhaust and this occurs before admitting fuel and trying to ignite it with ignitors (sometimes called spark plugs) to be sure any combustible gases are purged out of the combustion and exhaust section of the machine.. The purge time can be anywhere from 60 seconds (for machines with no HRSG and a short exhaust stack) to as much as seven or eight minutes because the purge occurs at about 20% of rated speed with the IGVs at the minimum (closed) position--both of which mean the air flow through the machine is not very large, hence the "long" times, especially when an HRSG is involved. These machines would go from zero speed/barring gear operation to rated speed in anywhere from about 15 minutes to 25 minutes, depending on several factors including the require purge times. The acceleration rate is conservative because the thermal stress on hot gas path components can cause damage to them if the acceleration rate is too high (because fuel flow is high).

In the last decade or so, there has been a large movement to do everything possible to reduce start-up time as much as possible. Some of the methods involve placing oxygen sensors in the exhaust section to monitor for high oxygen content, signifying low combustible gas levels to reduce the purge time. (Personally, I have participated in several tests of the placement of oxygen sensors and can say with confidence that there is a LOT of stratification in the exhaust section of a GE-design heavy duty gas turbine and that there are areas that have residual combustible gas levels that aren't the same in other areas (which have higher oxygen levels. So, unless there are many (more than three) oxygen sensors in the exhaust section to monitor for oxygen levels it's really not a great method for detecting combustible gas levels.)

Other methods include lots of calculations to determine how much air is flowing after fuel flow is shut off and the machine coasts down to zero speed, how much air flows through the machine while it's on barring gear (called Cooldown), and what the natural draft caused by high temperatures in the exhaust section (HRSG and exhaust stack) which draws air through the axial compressor, turbine and exhaust sections. These methods have considerably reduced, and in some cases, eliminated the need for a purge sequence significantly reducing start-up time to rated speed.

I have heard there is a desire to use the generator as a motor during machine starting (using a static frequency converter to vary speed of the motor during starting). Other machines that use static starters like this have also significantly reduce start-up times to rated speed by using a torque assist from the motor (generator) all the way to 95% speed to reduce the thermal stress of using fuel flow to accelerate from 60% to 100% speed. I don't know if there are any of those 9E machines out there yet, but I do know there is some "push" to make this starting means change for many reasons, one of which is to reduce starting times to rated speed.

That's about all I can say without knowing a lot more about the machine you are referring to. I would make an educated guess that the Siemens machines you are referring to use static starters to make the generator a motor during starting, and probably have one or more of the methods listed above to achieve those starting times. (If you can confirm that, it would great.)

 

Thanks for the reply, much appreciated. The 9E machine I am looking at has an electric motor and torque converter, there will be no HRSG, so hopefully limited (if any) purge times. You are correct that the Siemens machines I have experience with have a Static Frequency Converter to use the Generator as a Motor on startup, this is used until around 37Hz where the combustion takes over to 50Hz (ignition starts at around 6Hz), the whole process from barring, at 2Hz, to Sync at 50Hz takes about 7 mins, there is no purge time, even with the connected HRSG, as it is assumed that the natural air flow after shutdown, and before sync when the air inlet damper and stack dampers are open is sufficient (no HRSG's have gone bang yet !!!). Assuming the above, limited Purge, electric motor to start via torque converter, no HRSG, so maximum acceleration, what could I expect? from Barring speed to Sync speed?, my aim is to reduce this time as short as possible, as I would like to use the GT to respond quickly, and want to get an idea of how many MW's I could generate within X mins.

 

NOTE: The is NOT referring to a GE-design heavy duty Frame 9E with DLN combustors!!! And basically refers to gas and/or liquid fuels.

GE has a (strongly) recommended acceleration rate limit; I forget the name of the Control Constant, but it's associated with the TNHA and TNHAR signals for acceleration while starting. This is done to try to limit the exhaust temperature spikes (which means the turbine inlet temperature spikes!) to protect the turbine nozzles and buckets (GE terms for nozzles and turbine "blades").

GE sell a Fast Start option which pushes the limit of the start-up acceleration rate--but for 9E machines there's really only so much which can be done. I know of a couple of simple-cycle 9E machines that can go from barring (Cooldown in GE-speak) to FSNL (GE-speak for synchronous speed) in about 9-10 minutes or so. BUT, that's only possible with a normal fired shutdown and about 20-24 hours on Cooldown ("barring").

Many GE machines have a "Fast Load START" function which REALLY pushes the limits of metal stress. Usually, that means they can go from sync to Base Load in about 4-6 minutes--but one can hear the hot gas path parts metal molecules SCREAMING when that's done--and GE have a multiplication factor that is to be applied to every Fast Load START that is performed (something like a factor of ten starts AND another factor for fired time, or something like that; download a copy of GER-3620, I think it is--and be prepared to do a LOT of skimming to find what you're looking for because it's a one-size-fits-all (really, a one-size-fits-none) document which can get VERY confusing at many, many points. (I'm NOT referring to a Fast START here--as referenced above--which is an accelerated start from zero speed/Cooldown ("barring") to FSNL (sync speed), as opposed to a Fast Load START, which is a highly accelerated loading from breaker closure (synchronization) to Base Load. If the machine has DLN combustors and Fast Load START is selected it's pretty likely that on at least one of the combustion mode transfers there will be a problem which requires unloading and loading again to get through successfully, especially if the machine is cold and the ambient temperature is low or very low or even if the ambient temperatures is high or very high (because air flows can change greatly with "extreme" temperatures, especially on a cold machine).

Finally, with Fast Load STARTing don't expect Water Injection (if the machine uses it) to start and stay running the first attempt.

Good luck; Frame 9E's are workhorse machines when they are operated "normally" (recommended start-up acceleration rate and recommended loading/unloading rates). EVERYBODY wants to bid into the market these days for the best revenue/rates for MWs, and some machines are just simply better at this than others. My experience is that GE B/E-class machines aren't the best for this kind of service/application, but your mileage may--and probably will--vary.

 

Fantastic response, understood, and thanks for the document reference, i'll try not to break the machine !!!

 

There's one more thing I forgot to mention about exhaust temperatures. The maximum allowable exhaust temperature for all operations (starting, acceleration, operation) is usually around 1100 deg F (595 deg C or so). This limit is defined by the materials used in the gas turbine exhaust (diffuser mostly) to protect the exhaust (diffuser) from being damaged by excessive heat. This limit is quite often reached during acceleration when the IGVs are still "closed" and speed is not at rated and the compressor bleed valves are still open. (Reaching this speed usually causes the acceleration rate to slow, sometimes considerably.)

I strongly suggest you work with GE Belfort, France, to understand what can and can't be done with regards to protecting machine internals as well as the exhaust. I'm very confident they have already developed some options for speedier times to FSNL.

GE Belfort people, besides being seriously prone to needlessly and excessively complicating very simple and tried and proven control schemes, are usually pretty approachable about these kinds of modifications/questions. It just takes finding the right person to help get the conversation started (salespeople these days are pretty clueless about what they're selling and don't always make the best introductions or know the right people). I would suggest working with one of the power systems service managers as they usually have reasonably good technical skills and also probably have some contacts in engineering and maybe even the controls group.

Best of luck. There are a LOT of variables and concrete stops in an application like this. (That's why having a variable speed motor drive (using the generator as the motor during starting) can be so helpful, because the motor can handle the acceleration without too much assist from the fuel system--which tends to make things heat up when the last 40% of acceleration to FSNL has to be entirely with fuel and without any torque assist from the starting means.)