Hi.

First I want to thank control.com for a lot of practical informations about GE gas turbines.

I am a PHD student who is developing a mathematical model of Gas Turbine PG 9171e. I have some questions about the compressor in this machine.

1. Can someone give me information about design compressor efficiency and design compressor capacity? I need this values for scaling an existing map of compressor.

2. What is the maximum flow through inlet bleed heat valve (I mean percentage values)- Combustion Chamber is DLN-I

3. And the last question – I need a value of air mass flow rate in case when the Inlet Guide Vanes are full closed (ankle 57 DGA) and full open (84 DGA ).

Thank a lot for answers!
Best regards.
Marcin

 

Marcin,

We're happy you're getting a lot of good information from control.com.

However, the information you're asking for can be considered by GE to be proprietary information about axial compressor design, which GE wants to protect as intellectual property.

In all the years I worked for this OEM I never saw the information you asked for. I did see people trying to use instrumentation on the axial compressor of machines to try to ascertain this information (illegally).

It shouldn't be difficult to understand how a company who spends thousands of man-hours designing and refining their axial compressor--not to mention the rest of the unit design--would not want this information to be public. If you have a document that depicts per unit air flows, I would be very surprised--not shocked, because some governments are demanding technical information like this for a variety of reasons (some reasonable; some questionable)--so that's possibly why there are not firm values and only per unit values on the drawing. It still provides useful information--but not all the information which someone could use to reverse engineer or copy a design.

I may be wrong about whether or not GE considers this information proprietary--but only GE can answer that for sure. I suggest you try to contact the GE division in Belfort, France, to obtain more assistance with your project. You may find that they would welcome some help from a university student/doctoral candidate and might be willing to provide some information in exchange for some work. It's worth a try!

 

I have often wondered... Can't this info the poster seek, be had from the control constant of a new machine? A new machine is a scaled up version of prototype, right?

 

Marco Polo,

This information isn't in the Control Specification (which, among other things, contains the factory values of Control Constants).

 

Thanks all of you for answers.

I did not think that, these data are very confidential. I thought, that I can get these information from people from power stations, where these machines are working.

Is value of air mass flow rate measured?

If not, how air to fuel ratio is varying, while load is increasing (decreasing) in case, when combustion chamber is DLN – I?

In conventional combustion chamber air to fuel ratio is constant while load is changing?

PS. I found on control.com all constants for CPD-biased temperature control system and for Droop speed control system, so I thought I will get these informations.

Best Regards
Marcin

 

CSA,

I respect your expertise immensely, but I am not convinced.

> This information isn't in the Control Specification (which, among other
> things, contains the factory values of Control Constants).

 

This simply amazes most people--because they assume that the Great and Vaunted Speedtronic adjusts fuel flow-rate and air flow-rate to maintain a desired air/fuel mixture at all operating conditions. Most new automobiles and trucks do this today (check combustion stoichiometry)--and they don't have the Great and Vaunted Speedtronic control system. But the air and fuel flow-rates for automobiles and trucks are several orders of magnitude less than those of a GE-design Frame 9E heavy duty gas turbine.

No; there are regions of allowable stoichiometry and the combustion hardware (nozzles and liners) are designed to operate with the axial compressor and IGVs and the expected fuel characteristics--without special instrumentation to monitor or control air/fuel mixtures in the regions which reduce NOx. And GE spent hundreds of millions of US dollars developing DLN combustion systems, and to publish the information--even in documents labeled "Proprietary" would allow anyone to learn or reverse engineer their DLN system. And that would just mean more competition--and they don't like that. It lowers margins (profit margins).

And, DLN combustion systems require tuning to optimize emissions--and sometimes that tuning requires changes to fuel nozzles or liners to achieve desired results. And emissions on a GE-design Frame 9E heavy duty gas turbine are only lowest near rated power output--not at operating conditions.

Nope; the Speedtronic isn't that sophisticated. Think of it like an automobile with a carburetor--it will run fine if the carburetor is not tuned perfectly. It will run better with fuel injection, and even better with computer-controlled fuel injection. And even better if the computer controlling the fuel injection has a lot of instrumentation for measuring and controlling stoichiometry. But it doesn't need fuel injection or a computer or lots of instrumentation to run.

The DLN combustion system--including the axial compressor--is operated using parameters calculated using detailed design knowledge without making that knowledge (detailed design parameters) publicly available in a document or in the programming of the Speedtronic.

Also, in an industrial environment such elaborate instrumentation requires care and maintenance--and if you've seen some of the sites I've been to recently you'd know the units wouldn't be running at all if the instrumentation were necessary to run....

Lastly, gas turbine exhaust can be very "stratified" and measuring exhaust gases can be very difficult. One would also need to know fuel make-up on a near real-time basis. And even measuring axial compressor air flow-rate at such high air flows is very problematic and would require more sophisticated instrumentation than GE currently provides as a purchased option (the Performance Monitor package).

Personally, unless I had laboratory quality calibration instruments and an unlimited maintenance budget just for instrumentation and a couple of well-educated and highly trained technicians I wouldn't want to work at a gas turbine plant with a control system that measured and controlled stoichiometry. No way. The intangibles and the potential for error is just too great.

Let's not forget--TTRF1 (Turbine Firing Temperature Reference) is a <b>calculated</b> value. And Exhaust Temperature Control is an approximation for a constant firing temperature using axial compressor discharge and exhaust temperature instead of actual firing temperature--which isn't measured.

Marco Polo, can you please provide the names of the Control Constants that the original poster is asking for--or the specific section of the Control Specification that contains the information the original poster is looking for? Because if it's published in a Control Specification then the manufacturer isn't too worried about it leaking to competitors. And I would be happy to refer people's to the information. The original poster is asking for very detailed and specific information about axial compressor design, not operating parameters as normally contained in a Control Specification. Such detailed information isn't required for normal operation if the operating parameters are chosen with the knowledge of the details of the axial compressor design.

 

@CSA: I don't think GE consider some of the information to be proprietary, as air mass flow is given in data sheet of GE Frame 5 machine at my site. But I don't know if it is legal to post the values here.

 

Thank all of you for answers.

I read that TTRF1 (FIRING TEMPERATURE) can be also estimated (calculated) from “British standard – BS 3135:1989 (ISO 2314:1989) Gas turbine acceptance test.”

Now I am trying to model a DLN-1 combustion chamber. I found one mathematical model of this combustor. This model consists of mass and energy balances plus heat transfer equations. But these equations require a knowledge of geometry of combustor and combustion air to cooling air ratio. Off course I don't have these informations. What is more, this mathematical model assumes complete combustion process.
But few minutes ago, one very interesting idea crossed my mind:
I know, that while fuel is combusted in DLN I less air is required, so exhaust temperature from turbine (TTRM) is higher than in case when combustor is conventional.

And now, when I use a traditional model of combustion, which main assumption is that combustion process is complete (without energy equations – only stoichiometric calculations), the ttrf1 – I mean FIRING TEMPERATURE is maintained by Temperature Control System (knowing CPD and TTRX – Combined cycle). This control system is adjusted by constants: TTkn_I, TTKn_C and TTKn_S. So, for DLN I these constants TTkn_I, TTKn_C and TTKn_S are different than for conventional combustor (I am talking about the same gas turbines but with different types of combustors).

If this idea is correct, the conclusion is that I don’t need energy balance of combustion chamber. Why? Because FIRING TEMPERATURE is adjusted by Temperature control system, and it doesn’t matter which type of combustor is mounted in unit (DLN-I or conventional). This informations are included in these constants.

Off course I can calculate this temperature but it is not necessary.

Best Regards
Marcin

 

Marcin2706,

I didn't find a question in this latest post....

In either type of combustor the only time firing temperature is controlled is when the unit is operating with the IGVs full open and on CPD- or CPR-biased exhaust temperature control ("Base, Peak, or Peak Reserve). TTRX is continuously calculated--even when the machine is not running, but the only time it's used to indirectly control firing temperature is when the unit is at Base, Peak or Peak Reserve Load with the IGVs full open.

Calculated firing temperature is used to determine when to change combustion modes, and that's all. GE stopped saying TTRF (or TTRF1) is actual firing temperature a LONG time ago--because it's just an approximation (albeit a good one--better on some units than others) and it is only used to determine when to change combustion modes. It doesn't have to be exact; it only has to be GE (Good Enough).

It's not clear how good your model needs to be--or what it's to be used for. I can tell you that a LOT of money has been and will continue to be spent on models for training and reverse engineering. I can also tell you that some of the specific data you are asking for is definitely considered proprietary. Having said that, there are governmental and regulatory agencies that force companies to "publish" such information so I'm never surprised to learn or find bits and pieces of such data published in one form or another. But, again, companies that spend money to develop processes and the equipment to support those processes have a right to protect their property--intellectual and real. Just because there is a World Wide Web does not mean that any piece of information can--or should--be published. There are people who provide for their families based on the income they derive from developing and maintaining these processes and equipment and they have a right to keep doing so. While corporations may charge exorbitant amount for the processes and equipment, corporations are groups of people who have lives and families.

 

This mathematical model is developing as a support for diagnostic system in given power station. The main purpose is to develop a model which can simulate a behavior of gas turbine under different operation conditions. The model must be relatively not very complicated (on-line simulation) with good accuracy of calculations.
I don’t have a real map of compressor. To handle with this lack of information I found a compressor map of different compressor. Then, I want to calibrate this existing compressor map to get the real compressor map of PG 9171E. Of course this is not a very good solution but very popular. To solve this problem, I need some informations about nominal compressor map efficiency, nominal capacity of compressor and nominal pressure ratio. Pressure ratio is 12.6 and capacity is 407 kg/s – informations found in internet. But I am still trying to get the value of nominal efficiency of compressor. This is the first problem.

The second problem is DLN – I Combustion system. In DLN I system the exhaust temperature TTXM is higher than for unit with conventional combustion system. This statement I hope is true.

So, while unit is operating at base load the IGV are full open and FIRING TEMPERATURE (for PG 9171E – 1120 Celsius degrees) is maintained by Temperature Control System (knowing CPD and TTRX - Combined cycle). The first question is following – Is the firing temperature lower while unit is operating at part load than at Base load? At part load Firing temperature is calculated, but this value is only required to change the mode of combustion process in DLN I??. Is this true??

Second question – can I assume that combustion process in DLN I combustor is complete?

At Base load, the approximation of firing temperature is not really required because the control system is focusing only on two parameters – CPD and TTRX. ONLY THESE TWO PARAMETERS ARE IMPROTANT AT BASE LOAD???? (approximated value of firing temperature is only additional data)??

Marcin

 

Marcin2706,

> The second problem is DLN – I Combustion system. In DLN I system the exhaust temperature TTXM is higher than
> for unit with conventional combustion system. This statement I hope is true.

Actually, if you look at the Base Load exhaust temperature of a conventional combustor-equipped unit and that of a DLN-I combustor-equipped (both GE-design Frame 9E heavy duty gas turbines) I believe you will find the exhaust temperature of both machines is very nearly the same. The difference between conventional combustor-equipped and DLN-I combustor-equipped units is that the flame temperature in the combustors of the conventional combustor-equipped units is much higher and the "flame" temperature of the DLN-I combustor-equipped units is much lower--and this is how the NOx formation is reduced. The firing temperature (the temperature of the hot combustion gases leaving the first stage turbine nozzles is relatively the same for both machines--it's just that the hot combustion gases leaving the conventional combustor must be cooled and diluted with axial compressor discharge air before admission into the first stage turbine nozzles. DLN-I combustor hot gas discharge temperatures are much lower because there is relatively no diffusion flame so the hot gas temperatures are much lower to begin with. But the temperature of the hot gases entering and leaving the first stage turbine nozzles are roughly the same--otherwise the power outputs of the two machines equipped with different combustors would be very different, and they are nearly identical. So, the temperatures and pressures of the hot combustion gases admitted to and expanding through the turbine sections of both machines must be nearly identical to produce nearly the same power output. And, this also means that the exhaust temperatures of the two machines <b>at Base Load</b> will be nearly identical.

However, when the DLN-I combustor-equipped unit is operating at Part Load (loads less than Base Load) the IGVs are used to control air flow through the machine in order to maintain minimal air/fuel ratios (which are NOT coded/programmed into the Speedtronic turbine control panel!) so as to prevent combustion instability and loss of flame. If the air flow through the machine wasn't reduced as the fuel flow-rate was reduced while unloading the machine then the air/fuel mixture would become too lean and combustion would cease taking place. So, the IGVs are modulated to control air flow--which results in elevated exhaust temperatures at Part Load versus machines with conventional combustors that are operating in simple-cycle mode.

Machines with conventional combustors can operate in combined cycle mode using IGV Exhaust Temperature Control, which is essentially keeping the IGVs closed as long as possible during loading and unloading in order to maximize exhaust temperature to produce more steam when the gas turbine is exhausting into a waste heat recovery boiler (or heat recovery steam generator, an HRSG).

But the maximum exhaust temperature of any GE-design heavy duty gas turbine--regardless of combustor type--can never exceed the CPD- (or CPR-)biased exhaust temperature control reference, TTRX, for the given operating conditions. And, again, TTRX for both DLN-I combustor-equipped and conventional combustor-equipped machines is pretty similar. I don't believe you'll find too much difference in the Base Load exhaust temperature of machines with either type of combustor. The DLN-I machine might have a slightly lower exhaust temperature at Base Load, but if it did I would also expect it to have a slightly lower power output rating, also.

The axial compressor sections of both machines is pretty much the same; about the only changes are in the combustor types--coventional and DLN-I. Even the turbine sections are pretty similar, and so are the power output ratings. So, that tells me the firing temperatures (the temperatures of the hot combustion gases leaving the first stage turbine nozzles) and the exhaust temperatures are pretty much the same. Instead of excess axial compressor air flow (excess being over and above the amount required for stoichiometric combustion) being used for cooling and dilution of diffusion flame combustion (as is done in a unit with conventional combustors), in a DLN-I combustor the "excess" axial compressor air flow is directed into the primary combustion zone of the DLN-I combustor where it is pre-mixed with fuel resulting in a very lean air/fuel mixture which results in combustion that does emit an ultraviolet flame like conventional combustion does--and this is what reduces the NOx formation (since NOx formation is a function of temperature, if the combustion temperature is lower the NOx formation will be lower).

Yes; firing temperature will be lower at part load than it is at Base Load. There is a region where as the machine is near Base Load and the IGVs are being closed that firing temperature and exhaust temperature will be higher than for a conventional combustor-equipped machine operating in simple-cycle mode (not trying to maximize exhaust temperature at part load). But, there sill come a time when, as load is reduced further that fuel flow and air flow will definitely result in lower firing temperature and lower exhaust temperature. Less fuel; less firing- and exhaust temperature.

> At part load Firing temperature is calculated, but this value is only required to change
> the mode of combustion process in DLN I??. Is this true??

Yes, this is true.

> Second question – can I assume that combustion process in DLN I combustor is complete?

I'm not a combustion engineer; I can't answer that question.

> At Base load, the approximation of firing temperature is not really required because the control system is
> focusing only on two parameters – CPD and TTRX. ONLY THESE TWO PARAMETERS ARE IMPROTANT AT BASE LOAD???? (approximated
> value of firing temperature is only additional data)??

Conventional combustor-equipped machines do not calculate (approximate) firing temperature, and because firing temperature cannot be directly measured it is approximated using empirical data derived from decades of experience using exhaust temperature and CPD. For conventional combustor-equipped machines constant firing temperature is represented by the negatively-sloped line of the exhaust temperature control algorithm that defines Base Load. The firing temperature is assumed to be constant for various values of exhaust temperature and CPD. Firing temperature, TTRF1, is not calculated in conventional combustor-equipped units.

Units equipped with DLN-I combustors have the same negatively-sloped exhaust temperature control line that defines Base Load. The difference is that "firing temperature" (TTRF1) is calculated at all times during operation because the value of TTRF1 is compared to parameters that define when to change combustion modes in order to achieve premix operation which is the lowest NOx emissions mode. The turbine can't be started and operated in premix mode, so it has to transition from diffusion flame to premix mode, and the transitions from primary to lean-lean to premix are scheduled based on TTRF1. And when the unit is being shut down, the transitions from premix to lean-lean to primary modes are also scheduled as a function of TTRF1. TTRF1 is not used for any other purpose than to determine when to change combustion modes--so it's only calculated on machines with DLN-I combustors, which have to change combustion modes during operation.

 

Just a slight correction: When the IGV's are closed to maintain exhaust temperature, the steam production will decrease. The steam temperature will be maintained. As the IGV's close, the airflow and fuel flow are decreased and the exhaust flow will necessarily decrease, so the steam flow will also decrease (there is less energy being supplied to the HRSG). Maintaining the temperature reduces thermal shock to the steam turbine.

> Machines with conventional combustors can operate in combined cycle mode using IGV Exhaust Temperature Control, which
> is essentially keeping the IGVs closed as long as possible during loading and unloading in order to maximize exhaust
> temperature to produce more steam when the gas turbine is exhausting into a waste heat recovery boiler (or heat
> recovery steam generator, an HRSG).