Hello

I have a question. Inlet Guide vanes are used to regulate the air flow into the gas turbine, and introduce a swirl to control the mach number. But what exactly is the function of EGVs?

1) Its function is to eliminate the swirl in the air at the exit of compressor? or is there any other special function as well.

2) If it eliminates the swirl that means in case of high spread, any thermocouple showing high temperature (above alarm value) means that we have to check the particular combustion chamber directly behind that particular thermocouple showing high temp alarm. The concept of swirl will finish?

3) Currently I have a software which shows the number of chambers to be checked in case of high spread. It asks the frame of GT, the load at which the spread exists. After we put the data it gives the location of chambers against the Thermocouples by incorporating the swirl effect. Swirl is Maximum at zero load while minimum at Full load, why is it so?

4) In this case where we have EGVs installed, do we have to take into account the swirl effect? I mean should I trust that software.

Thanks
--
Izhar

 

Izhar,

The description below presumes your questions refer to a GE-design heavy duty gas turbine.

1) IGVs do not introduce exhaust swirl; they control the air flow into the axial compressor. When the air leaves the axial compressor it makes a 180 degree turn as it enters the can annular combustors where it makes another 180 degree turn as it enters the combustion liners where it is mixed with fuel for combustion and to cool and dilute the hot gases of combustion.

Swirl is introduced as the hot gases pass through the turbine section.

IGVs are variable (some are only open or closed, while others are modulated to control air flow more precisely in response to one of several references (through a min select gate)), and EGVs are not variable, but are fixed. EGVs do help reduce the swirl of the air leaving the compressor, but because of the path the air takes to get into and through the combustors and the turbine the EGVs can't help reduce swirl at the exhaust.

2) Exhaust swirl is the result of reduced gas flows through the turbine section that introduce a slight offset between the temperatures of individual combustors and the temperatures measured by the exhaust thermocouples. It surprises most people to learn that there is very little mixing of exhaust gases as they pass through the turbine section (which is rotating) and so combustors having hotter or colder temperatures entering the turbine can be traced from exhaust temperatures if the swirl angle is known.

3) Swirl is less at high load because the air flow is higher at higher load when modulated IGVs are more open. The pressure of the hot gases is also higher at higher loads when modulated IGVs are more open and this also results in less swirl as the hot gases pass through the turbine section.

4) No; EGVs do not affect exhaust swirl.

If you're having high exhaust temperature spreads, please describe the turbine (Frame size; type of combustor: conventional (diffusion flame) or DLN)) and the fuel being burned when the spread is high (sometimes on multi-fuel machines the spread is high on one fuel and not high when burning the other fuel). It is possible to help determine--approximately--which combustor(s) are experiencing problems without the the use of a swirl angle calculator <i>which is only a tool for <b>approximating</b> swirl angle,</i> not for precisely pinpointing combustors based on load and Frame type. Air flow through a GE-design heavy duty gas turbine can vary depending on how the IGVs are being controlled, if Inlet Bleed Heat is in service, cleanliness of the axial compressor, cleanliness of the inlet air filters, back-pressure of the exhaust, etc.

<b>And,</b> not all exhaust temperature spread problems are fuel nozzle-related--some are caused by cracked liners or improper installation of or failing transition piece side seals, or failed hula skirt seals--so just swapping fuel nozzles may not resolve the exhaust temperature spread.

Exhaust swirl is not caused by IGVs or EGVs; it's the result of air flow (which is also a function of air pressure) as hot gases pass through the turbine section. Higher air flows result in less swirl--as the calculator correctly shows; and lower air flows result in higher swirl. Seems counter-intuitive, but that's how it works in GE-design heavy duty gas turbines. Since modulated IGVs affect air flow, IGV position affects exhaust swirl angle. But EGVs are fixed and don't affect exhaust swirl since swirl is a function of gas flow through the turbine section--and the gas flowing through the turbine section has taken two reversals after it leaves the axial compressor before it enters the combustion liners and then the turbine section.

Hope this helps!

 

What to say, CSA has left nothing to add.

If I had initialized the answer I would have used, much more technical jargons. the answer truly shows CSA should be listened, loved and respected.

"you understand physics, if you could teach it to your grandmom"
Einstein

 

Thank you so much for such a wonderful reply.

Currently we are not facing high spread problem, but nevertheless like to know how to find the location of chamber which is creating the problem. We have a Frame 9E gas turbine, fuel is Gas, NON-DLN reverse flow type combustor. We are facing high Exhaust temperature issue(at a load of 111 MW). How can we find the swirl angle and location of combustion chambers in case of high spread issue at a particular load.

Looking forward for your reply.

Thanks

 

Hello There,

Swirl angle is the angle between measured representative exhaust gas temperatures(TTXM), at varying loads and the known combustion source location. However, Swirl angle is not a rigidly controlled parameter and may vary between units. So, it should only be treated as a tool.

Using map and graph for calculating swirl :

a) locate the cold region by looking at the exhaust temp. data.

b) Select the cold thermocouple and its corresponding location on the map.

c) From the clockwise location of the cold thermocouple back trace(clockwise on the map) the amount of swirl angle to identify the location of probable cause.

Troubleshooting using inferences

* Hot streak (signifies excess fuel not enough air) Probable solution/cause
- Inspect liners for plugged holes
- check fuel nozzle assembly

* Cold streak (signifies excess air or fuel deficiency)
- Inspect fuel nozzles, orifices may be giving trouble
- check valves (NRV)
- cross fire tube leaks
- transition piece seals and Hula skirt for proper installation and leak (this very symptom is usually visible after a major overhaul and henceforth plant operators needs to inspect initial temp. profile very carefully so that problem is treated on the spot.)

Allowable spread

Based on exhaust temp. and CPD on base load operation spread is usually between 40 to 55.

Having said that, the calculation still remains unanswered. But since I can not upload Excel sheet on the site. therefore I will suggest you that you should refer to www.ccj-online.com and search for a gentleman named J.C.Rawls and look for the paper which he presented on this very subject at 2013 frame VI Users Conference, there you will find an excel sheet explaining the concept of swirl and associated calculation. However, to grasp the calculation you need to patiently apply the method described there on a situation of your own. Rest assured, this helps a lot. But, if you still find it difficult then I am at your service 24*6.

Right now I am calculating (recalculating) the Critical speed for a Frame V turbine over which CSA has raised some doubts. There has to be some platform where we all can share our calculation, and results.

I you have any idea please do reply.

I hope this will be of some help to you

Regards,
fluidflow

 

Fluid flow

Indeed a very interesting website recommended by you. I have seen some of the presentation by J.C Rawls, great they are. I am also studying the excel sheet and the swirl angle presentation and will make my own excel sheet on frame 9E as well. I have one question:

1) Where from can I get the graph of load and swirl angle ? of course there must be a load vs swirl angle graph for Frame 9E.

2) Secondly, can you please share your study on Critical speed of Frame 6. I will try to give my inputs plus it will be a great learning opportunity. Please share your study.

Thanks again and special thanks to CSA, you are beauty of this forum.

B.R
Izhar

 

Dear Izhar,

I regret to respond after such a long pause.

I was busy with some mundane affairs. Nevertheless, I am glad to hear that you found my suggested site useful. For sharing my calculations regarding critical speeds, I am forwarding you my email Id, ([email protected]). kindly reply back on it so that I can email you my calculations.

Regards,
fluidflow

 

Thanks fluid flow. I have sent you an email.

 

> Thanks fluid flow. I have sent you an email.

Hello there

Just wanted to update about the problem I was facing. As I mentioned earlier we were facing problem with our GT (high exhaust temperature). Recently we carried out combustion inspection of our Gas Turbine and found out that the cross fire tube connecting chamber 03 and chamber 04 was almost melted (badly damaged). Transition piece 03 was also not okay, it was having heavy black spots unlike others. Floating seals of many of the TPs were found broken out.

We successfully changed all the parts of our Gas Turbine. What can be the major cause of high temperature at a specific nozzle? Most of the NRVs were passing (both Fuel and purge air), all were replaced with new one. After startup no high temperature issue persists.

Hope you to hear from you guys. Fluidflow pls check your email.

Thanks

--
Izhar

 

Izhar,

Thanks for the feedback!

We don't know what kind of combustion system is used on the turbine (conventional, diffusion flame; or DLN-I)--not that it makes a lot of difference, but it is helpful to know. If the unit has DLN-I combustors it could be that dynamic pressure oscillations in the combustion system have caused increased wear which has resulted in ill-fitting components and the type of damage you are reporting.

It's not clear if the machine is new, or has been in service for some time. So, if the unit has been in service for some time, and if the unit has gone through one or maintenance outages in the past, then the assembly practices used could be suspect. Or, if the parts used for re-assembly were non-OEM parts, they could be suspect (quality; dimensions).

These are the most usual reasons for the type of damage you reported. We don't know how long the combustion components had been in service before the damage was discovered, which could also have some impact. We don't know if the parts were re-installed after a previous outage and what the condition of the parts were when re-installed. There's just a lot we don't know.

We don't know how many emergency trips the turbine has experienced since the last maintenance outage, or on this set of combustion components. Trips can be very injurious in their own right.

But, re-assembly practices and part condition/quality are two of the most common problems. If the unit has DLN-I combustors, when was the last time the unit was tuned? Is the unit operated in Premix Steady State combustion mode during normal operation, or is it operated in Extended Lean-Lean and/or Lean-Lean mode during normal operation? These can also impact parts life; GE estimates one hour of Extended Lean-Lean operation is equivalent to ten (10) hours or Premix Steady-State operation. And, it's also known that long-term Lean-Lean operation can also result in premature combustion liner wear and even accelerated damage. (GE is still studying the effects of long-term operation of DLN-I combustion systems in Primary and Lean-Lean combustion modes.)

 

Hello There,

Based on my experience with GT, I will try to answer your questions. However I shall first start from very basics as, I feel dutybound to share as much as I can.

Part A covers basics regarding combustion and parts played by individual parts with reference to your turbine.

Part B Development Of a theory to tackle various problems

Part C Providing explanation at each point if required.

so lets start,
yours is GE MS9001E. Correct!

It was the worlds first gas turbine larger than 100MW. Yours' a NoN DLN m/c with reverse flow combustors. ONLY TWO COMBUSTION LINER DIA. ARE USED BY GE. 268mm FOR MS 3000, 5000, AND 6000 WHILE 358mm FOR MS7000 AND MS 9000.

The number of combustors is proportional to machine air flow divided by pressure ratio. Therefore you have 14 combustors (check it from your data sheet).

In your machine we have a reverse flow multiple combustion system which is a result of years of intensive development and field success. The firing temperature in your Turbine has to be around 1000-1100C. I believe we are dealing with Base load operation with peak load period. Kindly acknowledge the presented information and let us know that we are on the right path.

Kindly tell me what you do with the exhaust flue gases. One more thing Izhar, I am not an expert of the subject like CSA, and I believe experts shall give there valuable comments and guidance.

Regards
fluid flow

 

Thanks again CSA/Fluidflow. I acknowledge all the details of the turbine and yes we on the right path. Our is a 217 MW combined cycle powerplant, flue gases are used to generate the steam for steam turbine.

Our plant was commissioned in 2008. We had our HGPI in 2013 in March(we replaced all the CI parts). Then we has a rotor earth fault in the generator of Gas Turbine, our turbine was out for approximately 03 long months. Therefore this year the CI was carried out in August rather than in March. And the refurbishment of parts is done by non OEM.

Other than this all the details provided by fluidflow is correct.

Hope to have expert opinion of you guys.

Thanks

--
Izhar

 

nice point there..yeah it truly helps

 

E class Turbines

Ap = exp(0.018 &#916;Tf)
Ap is Peak firing severity factor and &#916;Tf is peak firing temp. order.

• Reduction in load does not always mean a reduction in firing temp. e.g. In HRSG operation, to maximize the exhaust temp., IGV is modulated for power plant steam and here firing temp. does not decrease until load is reduced below 80% of rated value.
• Another factor which affect combustion system is Acoustics dynamics (mentioned by CSA already). However for multi fuel combustion nozzle system like yours this dynamic pressure is greatly reduced.

UNDERSTANDING DYNAMIC PRESSURE

P(total) = P (static)+ P(dynamic) {directly measured using 5 point probe or pitot static tube)

Dynamic pressure is that component fluid pressure which represent kinetic energy. I am stressing this because in Combustion dynamics measurement of Dynamic pressure for gas turbines is indicative of “Humming”. I know that this dynamic pressure is continuously monitored by some probe of Benetally Neveada, but I do not know how they do it. Perhaps, CSA SHOULD HELP
.
Now if your nozzle is clogged then this will result in significant increase in Dynamic pressure and will distort the flame stabilization in concerned CC, which will result in poor performance.

COMBUSTION AND COMBUSTION CHAMBER
Internally a CC contains a cap and liner assembly. Compressed air at temp. about 300C leaves compressor and flow in opposite direction in the annular space between liner and CC. Liner is thus completely surrounded by relatively cooled compressed air, which continuously flows in liner via louvers. Thus casing of CC is only exposed to 300C and no special heat resistant material is required.

The liner however sees inside the flame (around 2000C) and thus requires special heat resistant material and abundant cooling air flowing through louvers.

Fuel is fed in CC via cap and liner assembly via vortex breaker. In reaction zone only a little amount of air is mixed with fuel to sustain flame otherwise abundant air can make the mixture very lean. The reaction zone is roughly 1.5*Dia. of Combustion length. After that in dilution zone hot combustion gas are then mixed with remaining air in dilution zone so that turbine metal can tolerate it.

Remember that Liner at rear end the transition piece slides into the HULA skirt and ensure expansion in axial direction.

FLAME STABLIZATION

It is mandatory to maintain a stationary flame within a high velocity air stream, accordingly the mixture has to be established between certain limits for flame stabilization. i.e. for flame to travel in speed higher than mixture speed, the flame blow down occurs which we see as post ignition trip.(ha ha ha). In order to overcome this difficulty a region of recirculation done stream the main burner or BLUFF BODY should be established. This region creates areas of local low stream velocity equal to flame speed ( reread it) which is used to hold the flame , hence sustained combustion.

In our case liner assembly caps have vortex generators and machined bluff body nozzle periphery add swirl to incoming air (Remember J.C.Rawls ppt) in setting up of radial and axial pressure fields which in turn influence flow fields. If swirl is very strong than this will increase axial pressure gradient very much which will result in an internal circulation zone(eddy formation) that will increase relative velocity of mixture relative to flame speed out of limits and flame will be swept.

And if axial gradient is too low then flame speed will become much higher than mixture velocity and too rich mixture will eventually lead to a trip. Now you can see that how mathematics is helping us in understanding Swirl and how swirl is a dynamic parameter and why it vary with changing load.

TRANSITION PIECE

Function is to provide a flow path for hot gases from combustion liner to nozzle first stage. At nozzle end the TP ends in floating seals. IT is only proper aerodynamic design of this fish tail that a radial temp. profile is generated such that temp. near bucket root is lower than bucket tip .This lowers thermal stress and prolong bucket life.

Thus most probable reasons for all your problems are(of course in my opinion only)
1. Dynamic pressure
2. Swirl and vortex generation (How flame propagates)
3. Firing temp.

Secondary Reasons
1. Creep deflection
2. Corrosion
3. Oxidation
4. Foreign particle ingestion

Also Transition piece problem which you mentioned can be credited to inefficient secondary cooling. TP has a thermal barrier coating and if owing to excessive tepm. Gradient and oxidation this elopes then black spots are observed. They are result of recrystallization and grain boundary distortion owing to thermal fatigue enhanced by oxidation.

In my opinion black spots which you mentioned over TP might also be formed if some compressed air is leaked in from floating seal bypassing CC. In that case spots will be oval with diffused boundaries but distinct appearance. As compressed air is far more cooler than dilution zone flue gas.

I hope this shall provide you some help in understanding the basic reasons of failure and how to troubleshoot them.

By the way, when such damage was taking place then, there should have been distortions in spread and you might have witnessed a high differential between thermocouple installed in Wheelspace. If this wheelspace temp. difference is higher than 85C during base load operation then OEM advice to shutdown the unit and look into the matter.

Moreover, if there were frequent trips and emergency shutdowns then this greatly reduces the Rotor equivalent life. This will become more clear to you if you refer to GE GER-3620L.1(10/10) by David Balevic at al.

Regards
fluidflow

 

Thanks for the feedbacks here is my response.

Background: Combustion Inspection of Gas Turbine was due and was carried out during Aug 01, 2014 to Aug 07, 2014. After outage machine exhaust temperature spread was on a little higher side ( Got increased from 15 deg C to 25 deg C). Also machine output was low by ~2MW. During this CI we had also installed new modified nozzle tips (for Gas startup).

Findings: The case was taken up with GE and MJBi but both of them ruled out any problem due to new modified nozzle tips. But we decided to replace fuel nozzles with old design and during inspection collars of all of the Combustion Liners were found damaged (All 14 were found damaged while 08 out of those were found missing, Also locking plates of 02 Fuel Nozzles were found damaged. It was decided to further open CI and HGPI parts in search of missing collars. Following are the findings.

1. It is suspected that all of the missing collars got carburized due to high temperature zone and only very small particles (less than 1 inch) were found

2. Metal of missing liner collars was found fused on Transition Pieces, 1st, 2nd and 3rd stage nozzles and buckets.

3. Coating was found worn out on 1sts stage nozzles and buckets during boroscopy.

4. 1st stage buckets leading edge tips were found chipped off (~5 mm, picture 5).

5. 1st stage nozzles trailing edge tips were also found chipped off (~6 mm).

6. Small indentations (~ 3 mm) were found on 3rd stage buckets (Picture 4) at same periphery on all buckets (metal was also found chipped off on leading edges of 13 buckets out of 92)

7. RCA was also conducted determine root causes, according to which following are the possible causes

• The new fuel nozzle tips caused the flame area to shift
towards combustion cans / uneven combustion dynamics which resulted in the failure / disengagement of liner collar due to carburization of holding perforated plate.

• Holding perforated plate of liner collars got disengaged due to some manufacturing / repair defect resulting in uneven combustion dynamics which caused the failure of CI parts.

Way Forward:

1. Transition pieces removal is in progress for complete inspection / boroscopy of 1st stage buckets and nozzles for any abnormality.

2. GE suggested to send their T/A for complete RCA of problem . Findings will be shared with them

3. We have 01 used set of all CI parts removed during outage in Aug 2014 which are in usable condition. These parts will be used until we receive new parts at site after repairs.

4. Detailed assessment of all Turbine parts to be done again with GE T/A.

5. PMI (Material Identification)of both old and new liner body, cap , collar and perforated plate to be carried out to rule out material / welding failure.

6. FPI of 3rd stage buckets to be carried out for detailed inspection of buckets.

7. Similar problems have been faced in Frame 9E turbines of neighboring industries. Their opinion to be sought on the failure.

Sorry for my late response as I was busy in the shutdown.

Thanks

 

Hello There,

Dear Izhar, terribly sorry to hear about such a loss!

However, thanks for such a prompt and detailed feedback. From your experience sharing forum is gaining a lot.

Regards,
fluid flow

 

Hello There,

Izhar, you did not mentioned about the blade tip clearances. Any increase in it may lead to power loss.

 

Hello fluid flow

Yes the turbine clearance may have increased. The turbine is operation with 1-2 MW loss. HGPI will be carried out soon.

THanks

--
Izhar

 

Hi fluid flow

Can you please email me some stuff (presentations/pdfs) regarding cooling and sealing air system of Gas Turbine(Frame 9E)

Thanks

--
Izhar, izharkhan101 [at] gmail [dot] com

 

Hello there,

I regret to reply after such a long pause. Right now I am about to join a new company.

Once I finish these mundane affairs, I will come back to forum.

Regards,
fluidflow