I have been wondering about this SRV concept in gas turbines, AS i know it has been controlling the interstage pressure P2 by the pressure control regulator (NOT POSITION CONTROL) but controller getting the feedbacks of LVDT, SPEED and P2 Pressure.

My question here is: Why do you need a SRV Just before GCV and After Shut-off Valve VS4? for EMERGENCY SHUT-OFF we got VS4 and for Pressure control we Got Pressure control valves just after GAS FILTERING skid which can control P2.

Could anyone explain me what is the purpose of this SRV other than P2 Control?

Could anyone elaborate?

 

The exact name of SRV is Gas Stop/Speed Ratio valve. It is most important during startup when the gas turbine ignited up to FSNL. The stroke of SRV is directly proportional to TNH. That is the calculated gain. Normally SRV and GCV are considered as a package together with the Turbine. The valve u mentioned just after filtering skid was probably under another engineering company, so not under GE control, I think. Moreover GE has a library of developed Gas control Sequencing block (BBL) which of course with SRV and GCV included. Many other things are site specific.

During a normal shutdown/emergency trip, both GCV and SRV will be shut within 0.5 sec by the force of the springs. However the control system will continuously check the position feedback of SRV, not GCV to make sure it is tightly shut. IF not, it will sound an alarm and prevent you from start up.

 

In addition to acting as the gas fuel stop valve (before the incorporation of VS4 in recent years), the SRV serves to drop the pressure upstream of the GCV during starting and acceleration, and also to maintain a constant pressure upstream of the GCV during operation at 100% speed.

If there were only one control valve in the gas fuel system and it had to drop the pressure from supply pressure to less than approximately 0.6 barg during firing, that valve would not be operating in a controllable region, or it would have to be a very expensive equal-percentage trim valve (which the GCV is not). So, the SRV serves to act as a pressure reducing valve upstream of the GCV during firing and acceleration, and to maintain a constant pressure during 100% speed operation, which also causes the flow through the GCV to be nearly proportional to stroke (which is also desirable).

By maintaining a regulated pressure upstream of the GCV, it also serves to isolate the GCV from most supply pressure fluctuations and therefore helps to maintain stable load during most supply pressure fluctuations.

And, before the advent of the F-class machines (which required gas fuel supply pressures in excess of 20 barg) it also served as the gas fuel stop valve.

So, it's a very important valve for all gas fuel operating conditions, as P2 pressure is very important at all gas fuel operating conditions.

You can find the formula for the P2 pressure reference in other posts on control.com.

 

Dear CSA and Kevin Thanks for your kind reply..

Dear CSA, I have come up with one question over here which struck to my mind when i was going through this SRV stuff.

I have been stroke checking since long time going through AUTO CALIBRATE function and ENABLing MANUAL and then i would give some position (Normally from 0, 10, 20......100, 90, 80...0, -10). i found all the time that my SRV was standing at the position which i have asked it to stand and I have gone through this SRV stuff and i have understood that it is a PRESSURE CONTROL with position feedback. In fact i have verified my IO.CFG that it is a Type 77 regulator (Pressure control with PRESSURE and POSITION feedback) which means first pressure input and Maximum of 2 assigned LVDTs.

For your Information I have never changed before the Regulator type to 43 (Position Control with Position Feedback) before starting stroke checking

Please tell me How come this happens?

We got Three P2 Pressure Transmitter (FPG2), according to this type regulator it takes First assigned PT for FPG2, Right?

What about other 2 redundant TX? Will it become a first assigned PT when the assigned PT fails?

Can I see all the PT values on LOGIC FORCING display with three different Tag name? please tell the tag name if so

Thanks

 

Have you tried putting the P2 pressure transmitter signal in the Logic Forcing Display to see what happens? (It won't work as you want, but I'm curious why you asked before trying if?)

Have you RTFM? (Read The Fine Manual) I believe there is some good info in Sect. 7 about regulators, and I believe it shows how, when AutoCalibrate is being used that the regulator automatically changes from a Type 77 to aType 43 regulator. Which would explain why you've never had to inform me that you had to change the regulator type to stroke the valve. (Lucky me!)

Some of the Mark V inputs are used by each processor directly, without voting or high- or median- or low-selecting them. The P2 inputs fall into that category of inputs. If you want to see all three values of the transmitter inputs have you tried using the Prevote Data Display? If not, you'll be pleasantly surprised.

We are happy to answer questions, but we don't do doubts here. As long as you have questions, we will try to provide answers. Doubts are not good things to express here.

 

> You can find the formula for the P2 pressure reference in other posts on control.com. <

We run Non-ignition test after CI and HGPI whenever SRV/GCV calibrated. The formula for P2 pressure is:

(.08*RPM)-25= calculated value of P2

Then you compare actual P2 pressure. The absolute difference in both P2 value should be 1.2
Units used are psi.
Try this with (8% of 3600 RPM-25), when unit is running you will get P2 value.
Thanks

 

Dear CSA

YES i have tried that FPG2 in LOGIC FORCING display many time in my life with this MKV it would read just like all other ANALOG signals Say for Instance HQP, this would also read like 18.2 bar,18.3 bar in the LOGIC FORCING display and that's why I specifically mentioned LOGIC FORCING instead of PREVOTE data display, because i have seen that PREVOTE data display reading just before asking you for my verification, i know that i can see there all the THREE transmitter values with voted value.
Please answer me, how that value is voted?

My question comes from the root, because as per IO configuration these transmitters all are connected to the R -TBQB-23....34 with SINGLE VARIABLE NAME PFG2 and through ribbon cables it is fanned out to other 2 processors.

how do you represent THREE HARD WIRED transmitters in the TB as single variable?(Really i don't know about this stuff because some time ago while calibration we found that 1 TX was bad and other two was good , and we have never been noticed any alarm by operator too and our turbine was never tipped or disturbed)

Yes section 7 of APPLICATION MANUAL given about regulator but it is not having anything(As for me) about changing over during AUTO CALIBRATION, even i have found some of the threads telling to change it to regulator type 43 and then back to original regulator 77 after checking your stroke.

Please tell me the exact location of that document regarding the change over of regulator when you select MANUAL CONTROL.

 

The P2 pressure reference formula is (and always has been):

FPRG = (TNH * FPKGNG ) + FPKGNO

where FPRG = Fuel Pressure Reference-Gas (in psig)
TNH = Turbine speed, in percent
FPKGNG = P2 Gain (in psig/percent speed)
FPKGNO = P2 Offset (in psig)

The P2 pressure is different for different machines, sometimes even at the same site with the same fuel. But it's never the formula

Now, at your site, FPKGNG might be equal to 0.80, and FPKGNO might be equal to -25, but that's <b>NOT</b> true for every machine everywhere in the world.

But let's just take a look at these numbers for a minute. At 100% speed, this would mean a P2 pressure reference of 55 psig which wouldn't be true for any GE-design heavy duty gas turbine at 100% speed.

 

4_20ma,

I'm not really sure I like the tone of your responses; in fact, I don't like them at all.

If you've tried putting FPG2 in the Logic Forcing display many times in your life, why wouldn't you tell us you have and provide the results of your attempts?

In your first response, you said you had a question (thankfully, it wasn't a doubt), but you actually had several.

You don't even refer to the same device with consistent terms. Is it a PT or a TX?

I answered your question: Each processor uses its own value of FPG2 in determining the amount of servo current to be applied to its coil of the SRV actuator.

I don't know how the single value that's displayed on an operator interface as FPG2 is determined. I've never given it any serious thought, but I presume it's the median value. You can confirm that by looking at your newly discovered display.

I'm also going to say in the incident you are describing there were Diagnostic Alarms indicating that one of the P2 pressure transducers was out of limits. You either weren't paying attention to them, or you're not telling us about them. But, they were there. And there was a serious difference in the three SRV servo currents as a result, that went away when the pressure transmitter was repaired/replaced.

It's been a long time since I've had a read of Section 7 of GEH-6195. In early revisions I recall seeing the drawing of the regulators showing a logic signal that bypassed the type 7x regulator and converted it to a type 4x regulator. This would happen when AutoCalibrate was enabled.

Best of luck in your future queries! I hope someone can help you.

 

Didn't know about the change to 43 during autocal, thanks for the new knowledge.

 

Thanks CSA, this formula refer to RPM, and in M6 we are using TNH. If we use TNH in the formula then control constants might have different values. Correct me if I am wrong.

 

Clearly explained by CSA, keep it up your good work.

G.Rajesh

 

Hi CSA,
I am really sorry if something wrong with me, I have never meant to hurt anybody instead i just expressed my view and observation.

I just wondered why we don't have three different tag to name those three PT's, From the basic observation and understanding only i have expressed here.

As a maintenance person i won't be available in the <I> all the time so there could have been alarm in the night shift or some other holidays, i don't know really really but we found it when we normally schedule to calibrate in a shutdown.

Really sorry for not using the same terminology, it's my mistake.

It's your kindness that brings solutions and knowledge about Gas turbines to the millions of people through out the world, but if you say altogether that you won't answer that's really hurting me.

Here we may have highly experienced people, beginners and less experienced engineers and mostly immature to express things. In general people are not giving training and support in organizations then people like you can only solve the problem for millions of people who are looking something in the internet.

I am very interested to learn about GAS TURBINE and i am not getting enough opportunity to explore, because our management keep rotating people to different areas, some time BOP, STG, water treatment etc...

 

4_20ma,

After re-reading your recent posts, I'm still not convinced that there isn't some "anger" there, but I'm going to give you the benefit of the doubt on this one. You can be "upset" that things aren't always as you would expect them to be, but direct your disappointment elsewhere. I'm only the messenger; don't shoot the messenger because you don't like the message.

There are a couple of really important things you, and most everyone, needs to remember about GE-design heavy duty gas turbines. The first is that every site does <b>NOT</b> have the same auxiliaries. Yes, turbines all suck and squeeze and burn and blow, but they all do it with different auxiliaries and configurations and fuels and fuel supply systems and cooldown schemes and starting means, etc., etc., etc.

Your site has a pressure reducing station upstream of the SRV; some do not. And the GE philosophy has always been to keep control of critical parameters (like P2 pressure) and not use off-base or Customer-provided devices to do such things. It would also require expensive control valves to do the P2 pressure control off-base, so GE usually just required a relatively constant gas fuel supply pressure, which many pressure regulating valves can easily handle, and kept the P2 pressure control "in-house" by using the SRV for that purpose.

The second thing that people need to remember is that GE Speedtronic turbine control systems don't always act or behave as expected. And, there are exceptions to every "rule" when it comes to GE Speedtronic control systems. For example, can you find TNH or TNH1 in the IO.ASG file for the turbine at your site?

Can you find FD_INTENS_n in the IO.ASG at your site?

These are both pretty important, critical inputs to the Mark V, and they are not present in most IO.ASG files. And yet, the turbines still run and are protected just fine. There are reasons why these signals aren't in every IO.ASG file (and they are in some very early Mark V IO.ASG files, but not in most Mark V IO.ASG files) but they are absolutely necessary for control and protection. One would expect to find them in IO.ASG, but they're not always there.

You have assumed that every input <b>must</b> have a unique signal name. There is no standard or law or regulation that dictates that every input must have a unique signal name. GE has a way of doing things that don't always follow expected norms or standards. And, they're not alone in this; most Companies have unique ways of doing things that some people like and others dislike.

Each processor in a TMR Mark V has several inputs that are unique to each individual processor. P2 pressure is one; CPD is another; primary shaft speed pick-ups is another; MW transducers may be another (depending on the configuration; I'm referring specifically to the MAI16 inputs). Each processor uses its own input value in the execution of the CSP, and for most of the list above these are not voted values. So, if the P2 pressure transmitter input to <S> fails low, <S> will try to increase the P2 pressure by putting out more negative servo current to its servo coil, and there will be lots of Diagnostic Alarms to alert conscious operators and technicians to the condition.

Not all voting is done in the Mark V; servo outputs are voted at the servo-valve.

So, try to remember that every site is unique and every turbine is also unique. Yes they all have a starting means, but not the same starting means. Every site has a hydraulic supply system, but not the same hydraulic supply system. Most every site which burns gas fuel has a Stop/Ratio Valve, but not every SRV is the same (some are plug-type valves; some are rotary-type valves; some are right-angle plug valves; etc.).

And, try to remember that not every input is voted in the same way. GE turbine control systems have many idiosyncrasies, and while they may not always be understandable, one needs to at least know that they exist and are not always obvious.

For the values listed above, one can only see the individual values of each input to each processor in the Prevote Data Display.

Now, your next question is likely to be about LVDT feedback and how to see both LVDT feedback signals for each device, and that's another confusing story in itself. And another example of how things aren't done consistently or as expected.

I'm sure if you have experience with other control systems made by other manufacturers that you have encountered similar idiosyncrasies that confounded you, perhaps even upset you. That's life. Get used to it. And when you're dealing with GE Speedtronic turbine control systems, get used to it and expect it. If you're lucky enough to be at the site when the turbine control system gets upgraded to the next version of Speedtronic, you will find that many things have changed from your current knowledge and experience. SRV control is one of them.

Again, we can answer questions, and if we don't answer them to your satisfaction you have the responsibility to tell us specifically how we failed to answer your question so that we can try to explain it better. But to get angry and demand an answer to your question when one was provided that you apparently didn't understand and didn't tell us why you didn't understand it is not polite nor is it courteous.

I've spent a good deal of my life explaining things to people that didn't always like the explanation. And I've been the brunt of much dissatisfaction, and I'm pretty tired of it. You can like it or dislike it, but to express your dislike or anger to me is a waste of your time, and an even bigger waste of mine. I've heard it before, and I can't do a darned thing about it. It is what it is. And it's not going to change because you don't like it. You might seek a second opinion, but chances are you won't like it any more than the first (unless you get some differing information from the second person you sought out, and then you're going to have to decide which is correct...).

This stuff isn't easy, nor is it difficult, it just is. I try very hard to provide as much "supporting" information to make it more understandable or more palatable, but sometimes it just is what it is. It's been said here on control.com many times before: Engineering is a series of compromises. Not all of them are satisfactory to everyone.

If you need more information, be specific about what it is that you don't understand. And expect to get some answers that, while true, may not be to your liking.

 

sardar9,

In another post you have said that your site is using Mark VIe, and now you say M6 (which I interpret to be Mark VI, not Mark VIe).

Let's take another look at your numbers, during light-off, for example. The Frame 7E turbines at your site should be firing at about 10% speed so that would be 360 RPM (presuming the torque adjuster limit switches are set correctly).

(8% * 360) - 25 = 3.8 psig. That don't seem likely for natural gas fuel. (Now you're going to tell us that there are multiple fuels at your site and this is one P2 formula of two. But still the unit's ain't quite right. Please have a look and give us the correct units for all gas fuels at your site, presuming there is more than one gas fuel.)

I haven't seen TNH in any other engineering units than percent of rated speed. I haven't ever seen FPRG expressed as anything other than a pressure that is a function of TNH.

So, you say your site uses TNH, then what are the values of FPKGNG and FPKGNO, including engineering units, at your site?

Ask the I&C technicians at your site to show you the Network Topology drawings, referred to as MLI (Model List Item) 4108. (Of course, otised and I know that should have been A108, but someone (in their infinite, dyslexic wisdom) interpreted it as 4108 in recent years.)

 

CSA - some comments:

1. TNH is always speed in %. TNH_RPM is usually used for speed in rpm.

2. I agree that 3.8 psig is low for P2 pressure at firing speed. P2 has to be greater than compressor discharge pressure or you just aren't going to get the fuel to flow into the combustors.

3. The change from A108 to 4108 was intentional. I don't know the exact reason, but there may have been some change in the database software for the Model List system that needed to get rid of the alphabetic leading character. There used to be a lot of MLI's with a letter for the first character and I think most of them have disappeared. I don't remember if A108 was used on Mark V, but all Mark VI and Mark VIe use 4108 for the network topology drawing. Mark IV used A108, but its network was CSF (control system freeway, an Arcnet system), so they may have wanted to distinguish between Arcnet and Ethernet communication networks.

 

hi...

ours is a 9fa turbine...as per your comment, i tried for p2 calculation, but i am not getting the value which is showing in HMI...

as per manual:
gain constant = 4.01115074418605
offset constant = 21.1150744186046

At 3000RPM, value showing in HMI is 27.02 BAR but as per the calculation i am getting around 6.02 BAR. Where i am making mistake?

 

Any magnetic flow meter with appropriate lining (probably either PTFE or PVDF or something similar) and with appropriate materials of construction for the electrodes will work fine. There are about 45 manufacturers of 1 inch diameter magnetic flow meters.

Walt Boyes
Editor in Chief
Control and ControlGlobal.com
555 W. Pierce Rd Suite 301
Itasca, IL 60143

[email protected]
www.controlglobal.com

 

The gain and offset are based on pressure units in PSI. So, for 3000 RPM (100% speed), I get 422.23 PSIG, which is 29.11 barg.
The HMI is set to display units in bar, but the Mark VI (or VIe) does its calculation in PSI. The HMI converts the PSI to bar, but the internal calculation remains in English units.

 

The formula using speed in RPM is <b>INCORRECT</b>. And, what is the <b>actual</b> turbine speed at your site?

And you are probably not observing the engineering units, either, in addition to probably missing the negative sign in front of the offset in the manual. And, what's in the manual <b><i>should be</b></i> what's actually running in the machine, but you should use the values from the Control Constants Display that are actually running in the machine (it's not so easy to miss the negative sign in front of the offset when doing this).

FPRG = (100 * 4.01115 psig/%TNH) + (-21.115 psig) = 380.035 psig

380.035 barg = 26.20249 barg

And, when you say "...value showing in HMI is 27.02 BAR..." are you looking at the actual P2 pressure (usually signal name FPG2) in bara or barg, or are you looking at the P2 pressure reference, FPRG, in bara or barg?