from the electrics department...
Gas Turbine Basics Book?
Engineering and workplace issues. topic
Posted by thomson on 22 February, 2008 - 11:18 am
I would like to know how to understand the working of a gas turbine. Is there any book which can be used as a guide?

Kindly help

Posted by CSA on 23 February, 2008 - 9:57 am
I have yet to see a combustion turbine reference for the uninitiated or even the novice.

Most times when I get this question from people, what they're really asking is, "Where do learn how the turbine at my job works?" And as the discussion progresses, they want details of the start-up sequence, synchronizing, loading, unloading, shutdown, and cooldown. And this requires discussing auxiliaries (fuel control valves, hydraulic systems, fuel forwarding/handling systems, generators and electricity, and a whole lot more).

Have you looked at the offerings on and There are many books out there. Another source is ASME's International Gas Turbine Institute. They have a self-paced combustion turbine training course which isn't free but it's not expensive, either (I guess that's a relative term, isn't it?). As I recall, the course is more geared towards aero-derivative gas turbines, but most of the fundamentals are the same ("suck, squeeze, burn, blow" as we say "in the biz"!).

There are lots of books on the first, and little to nothing on the second (or you wouldn't be asking that, right?). Actually, if you want to know how the equipment at your plant works, it takes a lot of study and you need to learn how to read the sequencing- or logic diagrams for your unit. Most turbine packagers provide instruction manuals with system descriptions, and with knowledge of how each system works and the components of each system, you then start "reading" the sequencing- or logic diagrams to determine the sequence of events for that particular turbine. It really takes many years to learn how a particular unit works; and this is hard because at most sites people have responsibility for a lot of other equipment in addition to the gas turbine and the only time they really get to "study" the gas turbine manuals and diagrams is when something's perceived not to be working correctly on the gas turbine.

So, my question to you is: Do you want to know about the physical principles of gas turbines, or, do you want to know how the turbine and the device it drives at your job works including the auxiliaries and systems needed to operate the gas turbine and driven device?

I'd be very curious to know exactly what it is you're trying to learn, so please write back and also tell us a little about your situation (student, technician, operator, etc.) and what kind of turbine you have to work on/with. Look forward to hearing back from you!

Posted by thompson on 24 February, 2008 - 11:30 am
Dear CSA,
Thanks a lot for the detailed reply. I will try and reply on the questions you have asked me
I am working in the operations department of a power plant and we have 3 MS6000 GE industrial gas turbines, which uses the speedtronic Mark V for turbine controls.

I want to understand how this turbine works specifically and very well know that i must understand Gas turbine basics before i venture into understanding the machine. Pls correct me on this. I am a graduate in electrical engineering.

So what i wanted to know is that for a novice like me where does one start?? People like you who have so much of understanding of these machines will be the right people to ask this question (thats what i felt)

Sir, where does one start? Is there a path which one can follow which can make the learning process fun???

Help!!!!!! LOL.

Posted by CTTech on 26 February, 2008 - 12:10 am
Hopefully I can encourage CSA to provide a more eloquent explanation.

The 4 stroke Brayton cycle is simple and has many references to read dating back over 50 years.

The key to understanding a GE turbine is event driven. Speed pickups control nearly all events. Study 14HR, 14HM, 14HA, and 14HS and finally COMPLETE SEQUENCE events. Startup is the key, I have never witnessed a failed Stop. Events can occur during a stop but let us not "muddy up the water".

The protective circuit "The 4s" (4 ansi reference) is even more important. STUDY the protection circuit the MOST. Since you have a MKV, major protection is located in <P> core. Dig up the schematics from your application manual and study them closely.

Once one has a positive start, then one can deal with DLN or other processes.

Although I am a technogeek, I find it most enjoyable to KNOW exactly what is to occur NEXT after each speed pickup point. Some do not share in my excitement.

Posted by CSA on 28 February, 2008 - 12:06 am
CTTech has some good starting points. I had written a response, but now have reconsidered, as I was headed down a different path.

Yes; a Speedtronic turbine control panel can be easily "fooled" into thinking there's a running unit connected to it by using a frequency generator to simulate turbine speed. When the panel detects speed from it's speed pick-up inputs, it wants to do all manner of things. Most major "events" (pumps starting/stopping; many, but not all, solenoid-operated valves being energized/de-energized; etc.) occur at particular "speed levels" and those are primarily L14HR, L14HM, L14HA, L14HS (the "L" signifies they are logic signals in the CSP, Control Sequence Program). There are comparators in a BBL (Big Block Language Block) which compare the speed, in percent of rated turbine speed, to various setpoints, which are a part of a group of operating parameters and setpoints called Control Constants.

It's important to know that when L14HR is a logic "1" that the Speedtronic believes the turbine shaft is "at Rest" (that's where the "R" comes from). The "H" comes from "high-pressure" shaft, and on a Frame 6 the axial compressor and turbine are on one shaft, commonly referred to as the turbine shaft, but formally known as the HP (high-pressure) shaft. L14HR is a logic "1" when the HP shaft (the turbine shaft) is "at rest." L14HM is logic "1" when the turbine shaft is at or above Minimum firing speed. L14HA is a logic "1" when the turbine shaft is at or above "Accelerating speed" (an intermediate speed during acceleration that is used as a trigger, or "event," for some actions to be taken by the Speedtronic.

L14HS is a logic "1" when the unit is at or above "Synchronous" speed. "Synchronous speed" in this particular instance usually means 95% speed, with the implication that if the unit made it to 95% speed during starting that it will make to "real" synchronous speed. "Real" synchronous speed is determined by the number of RPMs that the turbine must spin at in order to make the generator rotor frequency equal to grid frequency.

On Frame 6 generator-drive applications, there is a Load, or "Reduction," Gear between the turbine (and axial compressor) and generator. The axial compressor and turbine for Frame 6s are designed to operate at approximately 5100 RPM, the exact speed for each turbine (and compressor) is determined by the gear ratio in the Reduction Gear. Some gears convert 5094 RPM to synchronous generator frequency (speed); some convert 5100 RPM to generator synch speed; some convert 5134 RPM to generator synch speed. I've even been on sites where multiple units were purchased over many years, and there were at least three Load Gear ratios. All of the generators ran at 3000 RPM (for a two-pole synchronous generator on a 50 Hz system), but the turbines ran at very slightly different speeds around 5100 RPM, depending on the Load Gear supplied with the units. When the generator is at synchronous speed, the turbine (and axial compressor) are also said to be at synch speed.

Speed is usually expressed as percent of rated--where rated is turbine (and axial compressor) speed when the generator shaft is at rated speed (which is directly proportional to frequency). So, if the Load Gear ratio on a unit requires 5134 RPM to produce 3000 RPM (again, for a 50 Hz system), then 100% speed will equal 5134 RPM on the turbine shaft. (Generator speed is directly proportional to turbine speed, through the Load Gear.) The Mark V signal name for percent of rated speed is TNH, where "T" means Turbine, "N" is the variable name most commonly used in matematical formulae for speed, and "H" stands for High-pressure shaft (the turbine and axial compressor shaft).

There are a few drawings and documents you need to amass to begin this learning experience, and, hopefully someone at your site already has them available (though it's been my sad experience that getting people to share their drawings and documents can be very challenging, if not impossible...). You need to get a current copy of the CSP (Control Sequence Program) printout, and that really consists of two documents: CSP.PRN and CSP_XREF.PRN. The second document is the cross-reference to the first, and is kind of like an index or table of contents to the first.

Next, you need a copy of the Schematic Piping Diagrams, as GE calls them; everybody else in the world calls them P&IDs: Piping & Instrumentation Diagrams. The GE Piping Schematics don't typically have as much information as most P&IDs (like pipe diameters, etc.) and don't use symbols that everyone is familiar with, but they are critical to understanding the various systems and the control devices, instrumentation, and components which the Speedtronic turbine control panel interfaces with, and many which are not directly or even indirectly controlled by the Speedtronic. If your unit was packaged by GE (which doesn't include GE Energy Products-Europe), you should be able to find these drawings in Vol. III of the unit Instruction Manuals, and in that tab of that volume you will also probably find a "Piping Symbols Drawing" which can be used to understand the symbols used on the Piping Schematic diagrams/drawings. If your units were packaged by someone other than GE (and this includes GE Energy Products-Europe), you'll have to search the Instruction Manuals provided by the packager.

In fact, I always recommend that people who are trying to understand how their turbines operate spend lots of time going through the Instruction Manuals provided by the unit packager. Lots of time, means many, many hours. Start at the beginning, and turn each and every page by hand, one at a time. Make notes of what you find, and make photocopies of what you deem really important, as you go through the manual(s). There's lots of really good information, and many "surprises." On your first pass through the manuals, you're really just trying to get familiar with what's in the manual, and make a few notes so you can back later and investigate some more.

You should plan on doing this same thing tens of times over the next few years, as you discover new functions and need more information. Unit Instruction Manuals are never complete, and quite often lack lots of specific information about various pieces of equipment manufactured by others and provided by the packager. But, with the Internet, you can quickly find lots of good information about specific pieces of equipment as many manufacturers have put their instructions and information on their websites.

One of the things that's most helpful in learning to read the CSP is to understand GE's nomenclature. We'll go over that next. Suffice it to say that it is largely based on ANSI device numbers (as interpreted by GE). So, it's very good to learn the device numbers commonly encountered on GE machines. One of those ANSI device numbers is "3", which is the number for indication of a "complete sequence." As CTTech mentioned, L3, the complete sequence logic indication in the CSP is very important in order to be able to automatically synchronize the unit (connect the generator to the grid to operate in parallel with other generators to supply a load).

Now we have a few questions for you; take your time to get all the answers, but do provide us with all the answers!

Does your plant have HRSGs (Heat Recovery Steam Generators, or, boilers) on the gas turbine exhaust? Are there "bypass stacks" on the exhaust (to bypass the HRSG and send the gas turbine exhaust directly to atmosphere)?

What fuel(s) do the gas turbines run on? What kind of combustor do the turbines have: conventional or DLN-I (Dry Low NOx, generation I)? Do the units have any kind of Water- or Steam Injection for emissions control or power augmentation?

What kind of starting means do the turbines have: an induction electric motor or a diesel engine or ??? Can you tell us anything about the torque converter and the clutch between the starting means and the Accessory Gear Box?

What is the frequency of the AC grid at your site? Does the plant ever have to run while separated from the "grid"?

Are the turbines packaged by GE, or are they GE-design gas turbines packaged by someone else (Alsthom, EGT, BHEL, John Brown, etc.)? (It makes a difference, because if the unit was packaged by GE we can point you to specific Instruction Manual sections; if it's not a GE-packaged unit, it's much more difficult because we don't know much about what information is included in your instruction manuals.) Are the turbines provided by GE EPE (Energy Products Europe) (their Instruction Manuals are hopelessly complicated and inconsistent in their make-up and layout; so I'm hoping they're not GE EPE units).

Does the Mark V have <I> or HMI operator interfaces?

Does the site have any trends of start-ups and shutdowns showing speed, exhaust temperature, FSR (Fuel Stroke Reference), IGV angle, etc.?

Posted by transporter84 on 1 March, 2008 - 12:14 pm
dear CSA :
thank u for ur valuble informations .. but could u help us to find the a soft copy of the GE manuals for gas turbines frame 9, specially the parts which deal with speedtronic and control.

actually , i work in the instrumentaion and control maintence in a power station. ( and i am fresh graduated)

thanks alot

Posted by maverick on 2 March, 2008 - 6:53 pm
For very basic function of GAS turbine

Posted by CSA on 2 March, 2008 - 6:56 pm
GE has begun converting many of the Instruction Manuals for units they packaged to electronic files; they usually fill two CDs, one for unit information, and one for Spare & Renewal Parts. The problem with these CDs is that many of them are made from scanned images, and therefore are not searchable which makes them all but useless for people who can only use a web browser-like interface to find information (like people who think this forum is text-message based, which it's not). So, for those who can't open a book and page through it and find information, they can't use the soft copy either.

GE has also "converted" some manuals for some of the GE-design units which were packaged by companies they have purchased (John Brown, EGT, Alsthom, etc.).

Many of these CDs are available via the Internet (but they are still scanned images for the most part and so are not very text-message user-friendly, er, uh, can't be easily searched). I believe they can be accessed by going to and registering to use the site (providing some information, site name, unit serial numbers, etc.).

Sit down with the paper-based manuals; page through them slowly; make notes about information you see there so that you can go back and read it and/or make copies of it for your files (or scan it into your PC). I've been to many sites which have multiple copies of the manuals, but they are in almost new condition, which means that no one uses them. They're not novels, so they are very boring to read. They don't have indexes, and they don't have good Tables of Contents, and the information which is sought isn't always found where one would think it would be found, and sometimes it's not even in the manual but there *IS* a lot of very valuable and useful information in the manuals. I've been to sites where the manuals are in some supervisor's office and are basically not available to anyone to look through and use; this is a sad comment about those supervisors who don't anyone to know more than they do, and usually they don't even know what's in the manuals and what's not.

I don't know what the situation is with the manuals at your site; hopefully, they're available for everyone to refer to and use.

Actually, I'm surprised with the proliferation of pilfered media on the Internet that someone hasn't put a copy of the Instruction Manuals somewhere on the World Wide Web. Of course, it would'nt be for your site, but it might have enough relevant information to help you or others if you can determine what's applicable and what's not. Perhaps someone here will provide a link to such information.

And, if you would, could you please tell me why a soft copy is better than a hard copy? Is it because you could put a copy on your personal PC and read it when you're not working? Is it because it's not as heavy and bulky as the paper manuals?

Posted by thompson on 3 March, 2008 - 11:03 am
Dear CSA ,
Thank you so much for your detailed response.It feels good to get this guidance from you and helps build up the confidence.
I will try and reply to the questions you have asked me

1)Does your plant have HRSGs (Heat Recovery Steam Generators, or, boilers) on the gas turbine exhaust?

Yes we have HRSG'S on the exhaust of the GT ,the steam is then taken to drive steam turbines.Also there is a bypass stack to send the steam directly to the atmosphere bypassing the HRSG .

2)What fuel(s) do the gas turbines run on? What kind of combustor do the turbines have: conventional or DLN-I (Dry Low NOx, generation I)? Do the units have any kind of Water- or Steam Injection for emissions control or power augmentation?

The units run on Naptha fuel. There is a provision for steam injection for NOX control.

3)What kind of starting means do the turbines have: an induction electric motor or a diesel engine or ??? Can you tell us anything about the torque converter and the clutch between the starting means and the Accessory Gear Box?

The GT has a Diesel engine for starting. There is a troque convertor and a clutch, have read about that in the manual.

4)What is the frequency of the AC grid at your site? Does the plant ever have to run while separated from the "grid"?

The frequency is 50 HZ. Yes there are instances when the plant runs disconnected from the grid, supplying only private customers.

The GT is GE design packed by BHEL.

THE MarkV uses triple redundancy concept thats what i understand and there is a HMI screen from which the GT can be monitored and parameters changed

I should be able to locate the startup trends i think at least hope so!!!

Hope i have answered the questions you have asked me.

Hoping to hear from you

Posted by thompson on 14 March, 2008 - 1:28 am
Dear CSA,

What is the diffference between Simplex and TMR in MARKV?

In the control sequence program does it make sense to start from the top of the program or by taking the speed pick up levels when understanding the program?

Posted by PNV Subbarao on 16 March, 2008 - 4:06 pm
Dear sir, I have more querries on gas turbines.

1. IS the cumbustor and compressor intergral pert of the gas turbine.

2. What is the gas inlet temeperature and pressure?

3. What is the compresssor air pressure?

Thanking you

PNV Subbarao

Posted by CSA on 17 March, 2008 - 11:55 pm
One thing to know about commercially-produced gas turbines: every manufacturer's is slightly different, and the different turbines built by each manufacturer can vary greatly. So, not all turbines operate at the same temperatures or pressures.

1. Most commercially-produced gas turbines utilize a compressor to increase the amount of air that is required to oxidize the fuel that is admitted to the combustor(s) (some gas turbines have a single, "silo", combustor; others have multiple combustors, some have a single annular combustor). Increased air means more fuel can be burnt. Most commercially-produced gas turbines drive the compressor with a portion of the power produced by the turbine. The compressor output is directed into the combustor(s) where the fuel is admitted and burned. The resultant hot gases are directed into the turbine section, where they are converted into torque, a portion of which usually drives the compressor, the remainder used to power some driven device, such as a generator or a compressor or a pump. So, the combustor and the compressor are integral parts of most commercially-produced gas turbines.

2. Are you referring to the gas fuel that's being burned in the combustor(s) or the combustion gas "produced" in the combustor(s) by burning the fuel? If natural gas is used in the combustor, the temperature can range anywhere from near-ambient to as much as 350-400 deg F. The key is that most turbine manufacturers want the gas fuel to have approximately 50 deg F of superheat.

If some other gas fuel is used, the temperature of the fuel is a function of the process or method of production.

If you are referring to the temperature of the combustion gases produced in the combustor(s) that is directed to the turbine, they can be as high as nearly 2300 deg F for some turbines.

The pressure of the gas fuel at the inlet to the fuel control valves needs to be higher than the pressure in the combustor(s) (in order for fuel to flow into the combustor). Usually, gas fuel supply pressures are around 250-350 psig, sometimes higher, in a few occasions lower, but still never less than combustor pressure.

The pressure in the combustor(s) can be as high as nearly 200 psig (depending on the turbine design), and that would be the pressure of the hot combustion gases entering the turbine.

3. The compressor discharge pressure is slightly higher than the pressure in the combustor--not by much, but it must be slightly higher in order for air to flow into the combustor.

Posted by CSA on 16 March, 2008 - 9:28 pm
In the simplest terms of GE-speak (every big company has their own interpretations and meanings of words and terms and terminology; GE is no different from any other manufacturer in this regard), SIMPLEX means a control system which does not have redundant control processors participating in some kind of "voting" scheme or "hot back-up" scheme. If there is some problem with a determination made by the non-redundant controller or control processor then that problem will adversely affect unit control or operation. Of, if there is a problem with some component of that controller or control processor, then the unit will have to be shut down to replace the component. This type of control system generally has a lower availability and reliability than a similar control system with redundant controllers or control processors, like a TMR control system.

In GE-speak, a TMR (Triple Modular, Redundant) control system, means a control system which utilizes three identical controllers or control processors all executing the same logic or sequencing for the control and protection of the unit. In the case of the Mark V Speedtronic control system, each of the three control processors makes its own determination of how much fuel should be flowing or whether or not a solenoid should be energized or de-energized using the same logic or sequencing as the other two control processors. There are different methods for "voting" these determinations, but, for the processes and I/O deemed "critical" those determinations are made using redundant inputs and outputs (I/O), and, in the case of electro-hydraulic servo-valve outputs (electro-hydraulic servo-valves are used to control the positions of fuel valves and IGVs (Inlet Guide Vanes) and similar critical components) the voting is actually done "in" the servo-valve itself. In the case of TMR, voting generally means two out of three processors must agree that a fan should be running before it is started, or a solenoid should be energized before it is energized. In the case of the servo-valve outputs, the individual current outputs of each control processor are "summed" in the servo-valve to determine the final position of the fuel control valve or IGVs. In other words, voting can occur in software, control system hardware, and control system devices (such as servo-valves).

The idea behind multiple redundant control processors is that they improve unit (turbine and driven device) availability and reliability in a couple of ways. First, by using multiple control processors to make determinations about fuel flow and control and protection and then "voting" those determinations in some way (as described above) unit operation should be unaffected by a problem with an incorrect determination made by a single control processor because it is "out-voted" or, in a sense, "over-ridden," by the determinations made by the other control processors. Second, should there be a problem with a component (such as a printed circuit card) of single control processor, that control processor can be taken out of service to resolve the problem (to replace the printed circuit card) without having to shut the unit down.

If a person had a watch on his or her wrist, the time that the watch is displaying is deemed to be the correct time, even it it's slightly, or even wildly, incorrect. If the watch starts to run slowly or jerkily, then the person relying on that watch will not be as punctual as when the watch is running correctly.

If a person has two identical watches on his or her wrist, when both watches are running perfectly and for a brief period after both watches are set to exactly the same time they will both indicate the same time. However, it's likely that both watches will not always indicate exactly the same time, and on occasion one watch may fail (start to run slowly) or even fail completely. In the case where one watch starts to run slowly, it can be difficult to determine which watch is correct and which one is incorrect, so if one is interested in exact time having two watches is not the best defense against a failing or failed watch.

If a person has three identical watches on his or her wrist, even as the watches begin to normally drift due to their own mechanical differences it's pretty likely that as one watch begins to fail or fails completely, the other two watches will indicate nearly the same time and by having three watches instead of two the punctuality of the individual is even more improved and he or she is more reliable. This is the idea behind triple modular redundancy for control systems: using some kind of two out of three scheme as appropriate for control and protection.

The Mark IV and Mark V versions of the Speedtronic turbine control system use three redundant (identical) control processors, called <R>, <S>, and <T>, which are collectively referred to as <Q>. (The use of the 'less than' and 'greater than' symbols to enclose a designation means that it's a physical piece of equipment or is associated with a particular physical piece of equipment.) There is a fourth processor in these two control panels called <C>, for 'Communicator.' The inputs and outputs (I/O) which are deemed critical to the operation of a running turbine are generally connected to <R>, <S>, and <T>. [NOTE: GE does *not* generally include starting means I/O in the definition of critical I/O.] All of the critical control and protection functions (with the exception of emergency electrical overspeed protection) are accomplished by <R>, <S>, and <T> (<Q>). In the case of the Mark V, the <P> core (<P> stands for 'Protective') has three independent control processors (<X>, <Y>, and <Z>, collectively referred to as <Z>) which monitor shaft speed pick-ups to protect against overspeed conditions; the TMR Mark IV does not have this processor. (The words 'core' and 'processor' are used interchangeably in GE-speak and documentation.)

In the Mark IV and Mark V, the <C> processor was equipped with limited I/O capability and because it did not generally participate in the voting scheme of the three redundant control processors (<Q>) the I/O which was connected to <C> was deemed non-critical. Now, to every turbine owner in the world, *ALL* I/O is critical and so this distinction is not valid in their opinion. But, the design of the control system and the philosophies used in the configuration and application of the control system made it possible for some I/O to be connected to the <C> processor, thereby reducing the cost of the control system (and cost is *VERY* important to every turbine owner).

The I/O associated with <C> is "visible" to <Q> and the logic and sequencing in <Q> uses many of the inputs to <C> and controls many of the outputs of <C>. <C> is capable of having its own separate and unique logic and sequencing, and in many cases does, but no critical protection and protection functions are accomplished using <C>'s logic and sequencing capabilites (in a TMR control system).

Many Mark IV and Mark V SIMPLEX control systems have a <R> control processor and a <C> control processor. And, in many SIMPLEX applications, there is some back-up control and protection accomplished using logic and sequencing in <C> and using <C>s I/O--but that's generally only for SIMPLEX applications.

Having said all of the above, there are "violations" of the intentions of TMR and the application of I/O to TMR (and SIMPLEX) Speedtronic turbine control systems. Some of these "viloations" occur because of misunderstandings of unpublished control philosophies and practices (some of these control system philosophies and practices are considered proprietary, and almost none of even the non-proprietary ones are published). Some occur because of downright ignorance. Some occur because of Customer demands (that's right; there are some Customer representatives who insist that they know better than GE how to control and protect GE-design turbines--and these sites are usually among the most vocal when nuisance trips and operational problems occur!). Some salespeople allow some Customers to dictate equipment or practices which conflict with unpublished control philosophies and practices. Some field engineers (technical advisers) also are unfamiliar with GE control philosophies and practices and unwittingly or unknowingly violate them, or are forced to violate them by Customer's representatives who insist they know better than GE how to control and protect GE-design turbines.

That's one reason why it's so difficult to troubleshoot some turbine control and operation problems remotely. Also, in case anyone hasn't noticed, every turbine operator and technician and owner and supervisor believes the configuration of their turbine is just like every other similar turbine in the world. To those with Frame 6s, the configuration of their Frame 6 and their Frame 6's control system is the same as everyone else's, and, more importantly, their Frame 6 and their Frame 6's control system is configured correctly. Unfortunately, this is *NOT* the case. Nearly every Frame 6, while functionally identical, is different. There may have been some different combustion hardware used in the construction of the unit, or at some time after the unit was started-up, that requires slightly different sequencing or control set points than another Frame 6 sitting 15 meters away, or even 15 km away. Different packagers of GE-design Frame 6s have different compartment enclosures or auxiliaries that require different sequencing that a GE-packaged Frame 6 sitting 15 meters or 15 km away.

I have been to multiple sites, each with multiple Frame x's (substitute 6 or 7 or 5 or 9 for "x") all installed and started-up (commissioned) at the same time, which have very different logic and sequencing and wiring and operational setpoints. And to these owners and operators and technicians all the units, when they are running and producing power, appear to be identical because functionally they are
identical: they produce torque at relatively the same rates. However,
when troubles begin to appear and they start comparing one unit to another, that's when the "fun" begins, and the questions and blame and accusations really start to fly. And that's when it's really hard to explain how these differences could occur.

So, that's probably a little more than you wanted to know about SIMPLEX and TMR. The above is the general meaning. In TMR control panels, sometimes the I/O connected to <C> is referred to as "SIMPLEX" I/O. In Mark VI Speedtronic turbine control panels there is no <C> processor but there are some inputs and outputs which are connected to single control cards in TMR racks referred to as "SIMPLEX." So, the context of the conversation is very important when referring to SIMPLEX and TMR.

As far as "where to start" in the CSP (Control Sequence Program), that's very difficult. Most people don't start "from the top", but they start where the function they are interested in is first determined or referenced or "written to" (in the case of an output) and then work around the CSP as necessary.

With respect to speed levels, what I think CTTech was trying to say was, it's very important to understand each speed level, when it's actuated and when it's de-actuated, and what occurs at each speed level (for example, if a solenoid is energized at L14HM or L14HA or L14HS, or when a fan is started (at L14HA or L14HS); etc.).

Have you been able to obtain a paper copy if the CSP for your site, usually called the CSP Printout? That should include two documents: the CSP Cross-reference (useful for determining where to look in the CSP to find where signals are written to or where they are used as permissives to other signals or functions), and the CSP (the printout of the relay ladder diagram and BBLs (Big Block Language blocks, or algorithms). Your own personal copy of the CSP can be used to write notes in as you begin to work around the document; it's very important to have one to refer to, with one's own notes and comments, in order to develop a good understanding of the operation, control and protection of the unit one is working on.

Let us know if you have your own copy of the CSP, can photocopy someone's copy on your site, or need instructions on how to produce one (beware--it takes as much as 400-500 sheets of paper, single-sided, to print a CSP!). To print a CSP, a laser printer is almost a necessity so if you don't have one or access to one it's pretty useless to try to print one using a dot matrix printer or an inkjet-style printer. Also, trying to jump around an electronic file of a CSP is very difficult for most people, and it's not really possible to make notes in an electronic file version. A paper copy is best.

Posted by Vibhav Mehta on 24 May, 2008 - 2:19 am
Thats a very clear and elaborate reply however there are few questions:

(i) What is the difference between TMR and "2 out of 3" control system? Is TMR a GE terminology, because many other vendors like Triconex also use this word?

(ii) As you explained there are many signals which are simplex in nature and are wired to only one processor, does it violate TMR?

Posted by CSA on 28 May, 2008 - 9:28 pm
i) I have no idea if TMR is a GE term or not. Can you think of a way for any control system with three processors to perform redundant control without doing two-out-of-three voting, regardless of what it's called? GE does have a penchant for TLAs (Three-Letter Acronyms); 2oo3 ain't very pretty, and TOOT is four letters.

ii) Not every input or output is critical to the process, so to reduce cost the non-critical ones are not connected to redundant processors. (It's just that everybody's definition of non-critical ain't always the same.)

Posted by Anonymous on 7 June, 2008 - 12:35 am
Dear CSA,

You have provided very good basics how to learn Gas turbines starting and learning.

Can you please write step-wise starting sequence of GE Frmae 9 GT for me?

Best Regards,

Posted by CSA on 13 June, 2008 - 8:03 pm

A detailed description of the starting sequence of any gas turbine would require the Speedtronic elementary or CSP or Toolbox file or application code for the particular turbine. There are just too many variables, even in the Frame 9 (presumed you are talking about a 9E, and not a 9F, 9FA, or 9FA+e) line. The auxiliaries also varied somewhat depending on the packager of the GE-design unit.

I will tell you that you can start looking at the logic beginning at the logic signal L1X and L1Z, and this presumes the unit has either a Mark IV, Mark V, or Mark VI unit with typical logic signal names as provided with the Speedtronic turbine control panels. When L1Z goes to logic "0" after a START is initiated, the auxiliaries will start. Auxiliaries include the Aux. L.O. Pump, which must develop pressure to overcome the low-low L.O. pressure trip logic signal L63QT. The Master Protective Logic, L4, cannot be picked up until the L.O. permissive has been satisfied (look at signal L4S). L.O. is required for bearing lubrication, Torque Converter fluid supply, and Hydraulic Pump suction supply (the hydraulic pumps usually supplied with GE-design heavy duty gas turbines cannot draw oil into their suction and must be pressure-fed from the discharge of the Aux./Main L.O. Pump discharge).

The starting means logic signal is usually L4CR, and you can follow that logic to see when it is picked up (set to a logic "1" to start/run the Starting Means).

The Torque Converter usually has two solenoids: 20TU-1 and 20TU-2. L20TU1X usually is the signal name of the logic that drives 20TU-1, which is the solenoid, that, when energized, will allow torque from the starting means electric motor to be transferred to the shaft of the turbine-generator. There is also a Torque Adjustor Drive Motor on most Frame 9E Torque Converters that is used to control the Torque Converter Guide Vane position, which controls the torque being transmitted through the Torque Converter, and it has Raise and Lower signals usually driven by logic signals L4TMR and L4TML, respectively. Usually, the Torque Converter is set to maximum torque for purging, then dropped back to minimum torque for firing, then pulsed back to somewhere around maximum for acceleration once flame has been established.

Purging takes place as the unit accelerates above approximately 10% speed (L14HM); that should be when L2TV starts timing out, which is the purge timer. K2TV defines the purge time, and T2TV is the "accumulate register" where the incrementing time is stored. When T2TV equals K2TV, L2TV will be a logic "1" and the purge time is complete and the unit should begin to decelerate down to firing speed.

The Torque Converter provides most of the torque for acceleration up to about 50- or 60% speed (usually L14HA or L14HC), at which point 20TU-1 is usually de-energized and the starting means electric motor then usually runs unloaded for a few minutes to cool the motor. The unit continues to accelerate based on the amount of fuel being admitted (which should be gradually increasing).

At approximately 78-82% speed, the IGVs should begin to open to miminum modulating position (approximately 54-57 DGA). Once the unit reaches 95% speed (L14HS), the Aux. L.O. Pump and the Aux. Hyd. Pump are shut down and the Exhaust Frame Blowers are started, usually one after the other.

That's the basics. The type of fuel will determine when fuel-related devices are energized or de-energized.

But, start with L1X and L1Z. To get a START, one must have a 'Ready to Start' indication. That comes from L3STCK and L3RS; those are the Start-check Permissives.

You need to tell us what kind of unit you have and what kind of control system and and what kind of combustor (conventional or DLN) and what kind of fuel and we might be able to help with more details or answer other questions. But, a detailed procedure can only be developed from the control system drawings/programming of the unit(s) at your site. We can try to help with questions but that's about the best we can do.

Posted by Ramesh on 23 February, 2008 - 9:56 am
Contact NPTI they have published a book on Gas Turbine. NPTI contact details available at

Posted by a_w_p_d on 27 March, 2008 - 2:25 am
I have, which I could lend to you depending on where you are, a set of vids that explain all about GT/ST/HRSG/CCGT/Rankin Cycle/Braton Cycle and so on...

Leave your email address and we can see if we can work something out.

Posted by bhavesh on 16 June, 2008 - 5:33 pm
my email address is

Posted by Myvon on 29 June, 2008 - 4:11 pm
It will be well serve if you can just show where to get these GT/ST/HRSG/CCGT/Rankin Cycle/Braton Cycle and so on vids for self-paced learning have, which I could lend to you. my email is

Posted by TurboYankee on 29 June, 2008 - 7:02 pm
Working for a utility that has 7EA's and 7FA's, I have found that GE is a good resource for training and technical information regarding their combustion turbines (and steam turbines, for that matter). The site documentation tends to be pretty thorough--much more so then other turbine manufacturers I've encountered--and their on site training is usually high quality, too. (The term high also applies to the price.)

We have an LTSA (long term service agreement) with GE, and they have made available on line interactive training for its LTSA customers called PATAT (for Pro Active Trip Avoidance Training) at no extra cost, which is useful for teaching some of the fundamentals of CT, ST and HRSG operation and maintenance. It covers subjects such as exhaust thermocouple trip logic, the TC's themselves, compressor bleed valves, lean blow out, 3 element drum level control, and so on.

We're also going to try out GE Energy's online 48 hour CT fundamentals course, along with their on line generator and steam turbine fundamentals. The cost for this package is about $700 if taken over 6 months and about $500 if taken over 3 months. Considering the cost of on site training for one week, this is a real bargain, and it's probably pretty good. We'll find out and let you folks know. They also offer more advanced on line training courses.

I haven't seen anything on line regarding Mark V training, though. My company had a seven day class on site years ago on Mark V w/I stations which was before my time and from what I've seen of the student manual it was pretty thorough and it has proven useful to me as a self-study guide.

No matter what training is made available, however, you can never really know your machines without the site specific documentaion. We have two two-CT-on-1-ST CC units, or four 7FA's total. Each CT pair uses a shared LCI for starting via the EX2000 exiters. These units are not only unique compared to 7FA's on other sites, there are differences between units on the same site even though they were built at roughly the same time.

Part of that is due to control wiring inconsistencies and outright errors, and the rest is due to equipment differences which only GE can explain. For example, the LVDT's on unit 1's GCVs are different than unit 2's. So when we bought spares, we based the purchase on unit 1.

The CSP's are probably identical but after years of operation we're finding problems with the CSP and Mark V configuration, such as missing inputs that "should have been there." The erection engineer wasn't very thorough, as it turns out.

To really know your CT's is going to take a long term commitment and a lot study. I think it's worth it, though, as people with that kind of knowledge are pretty scarce, making them valuable to many organizations.

Your use of this site is subject to the terms and conditions set forth under Legal Notices and the Privacy Policy. Please read those terms and conditions carefully. Subject to the rights expressly reserved to others under Legal Notices, the content of this site and the compilation thereof is © 1999-2014 Nerds in Control, LLC. All rights reserved.

Users of this site are benefiting from open source technologies, including PHP, MySQL and Apache. Be happy.

The average income of the modern teenager is about 2 a.m.
Advertise here
our advertisers
Help keep our servers running...
Patronize our advertisers!
Visit our Post Archive