GE 9E/Non-DLN/Primary Fuel: GAS Low BTU/Start up and backup fuel: HSD, Control: Mark Iv

In CCPP

Plant was in commissioning phase, GT was operated only 300 hrs during these activities Combustion chamber 8 was burnt badly. Combustion chamber cover was melted near fuel nozzle area. fuel nozzle front flange and gas tip locking strips were melted. combustion liner cover cap and cowl was burnt. flow sleeve and chamber casing (can) were over heated.

After that fuel nozzle was pressure tested by GE but no internal leakage was found. design and material of parts were checked and found as per recommended.

These all parts were replaced, GT started again. after 5 days operation bore scope inspection was performed. Combustion line#8 was damage was started similar as previous.

Again investigation was done. this time little problem was observed at gas side. Condensate was coming with gas. It might be pass through coalscer filters. Gas composition was re checked, it was also varying from Design parameters.

It was recommended by GE to install knockout drums and increase gas temperature up to 70C, improve filtration.

All was done. lines were blow out, filters were replaced, additional heater installed, Knock out separator installed, C hangover from HSD to Gas was limit to 50%load after start up/during shutdown.

During each activity many check valves of water injection, fuel oil and purge air were found passing in reverse direction and as well as in forward direction before recommended pressure. Fuel oil mixing was found from water injection manifold, and fuel oil also in purge air manifold. These were also replace.

Moreover each possible relative system were checked and recommended action were taken.

And now after two week operation same failure happened at chamber-9. liner is worst damages well as fuel nozzle, combustion chamber cover and flow sleeve are also affected. Liner particles also travel forward to affecting transition piece.

In each accident parts of liner and cover also affect forward hot gas parts.

In each time all parameters were normal, our focus was at exhaust temperatures, in first case was absolutely normal. second time only once a single thermocouple was showing high temperatures but spread was not high even that point. third time also temperatures very normal.

Now my question is very simple
Why...?

It is very interesting...GE machine is suffering since one year, I would appreciate to look in detail.

I will provide as much detail as I can, Picture if anyone want to see I can mail

Thanks

 

It seems that there is something wrong with fuel system complete site study should be conducted to analyze issue

 

There are so many bells and alarms going off with this post, but, the originator has indicated he would provide information, so, let's see just how much he's willing to participate in understanding and possibly solving the issue(s).

I need to start off by saying that while low-BTU gas fuel operation is not common, it's also not unusual. There are many units around the world operating successfully with many different types of low-BTU gas fuels. The key is that the fuel supply needs to be relatively stable, meet expected characteristics and composition, and most importantly be free of entrained liquids and of sufficient temperature to prevent condensation of liquids as it passes through the various components (each with a pressure drop and associated temperature drop) all the way to the fuel nozzles.

Next, I'm not going to touch the check valve issue. It's most likely completely separate from the low-BTU gas fuel issue, and depending on the age of the unit (and the originator indicates the control system is a Mark IV, so that leads one to believe it's about 20 years old at least) there are known issues with check valves of that era.

First, the originator indicated the control system is a Mark IV. Please, Kashif, clarify that statement. When was the unit originally installed and put into service? What has the service been since the original installation: Base Load operation; part load operation; peaking operation?

Second, has the unit ever burned gas fuel in the past, either natural gas fuel or low-BTU gas fuel? If so, estimate the percentage of time operating on gas fuel versus liquid fuel. And, if the gas fuel operation was natural gas fuel, please indicate the percentage of operating time on natural gas fuel as well.

Third, was this unit ever intended to burn low-BTU gas fuel? If so, has it ever reliably burned low-BTU gas fuel? If so, when and for how long (days, weeks, months, years)?

Fourth, if this unit was not originally intended to burn low-BTU gas fuel and it is being converted to burn low-BTU gas fuel have you contracted with GE or some engineering/consulting firm to help with the conversion? Or are you attempting to commission operation on low-BTU gas fuel for the first time since the unit was originally installed and put in service?

Fifth, you wrote:

>Gas composition was re checked, it was also
>varying from Design parameters.

Please clarify that statement, and provide details of precisely how the fuel composition was varying from Design parameters.

Sixth, what is the dewpoint temperature of the low-BTU gas fuel, and what is the temperature of the low-BTU gas fuel just before it enters the Stop/Ratio Valve?

Seventh, you've indicated there is some heating of the low-BTU gas fuel. Is there any 'bypass' or recirculation of heated fuel prior to transferring to low-BTU gas fuel from distillate?

Eighth, the gas manifold is usually a continuous ring manifold with a single entry point, usually at or near the bottom of the manifold where the gas fuel from the discharge of the Gas Control Valve enters the manifold. Looking at the gas fuel manifold entry point, please tell us the numbers of the combustors closest to the entry point. (For a typical Frame 9E with an entry point at the bottom of the manifold, the two closest combustors would usually be 7 & 8 or 8 & 9, depending on the vintage of the machine and the locations of the combustors.)

Ninth, when you were told by "GE" to install a knockout drum who specified the design, sizing, and placement of the knockout drum in the gas fuel supply piping?

Tenth, has the knockout drum removed any liquids from the low-BTU gas fuel supply during operation?

Eleventh, going back to the low-BTU gas fuel temperature, the 70 deg C value you mentioned, was that 70 deg C of superheat or 70 deg C supply temperature?

Twelfth, how is the low-BTU gas fuel heated prior to be sent to the gas turbine? What is the discharge temperature of the fuel heater? Is the piping after the fuel heater insulated all the way to the Stop/Ratio Valve?

These questions are all related and relevant to the situation being described. You wrote that the fuel nozzles were inspected and found to be good, so that seems to eliminate the fuel nozzles as the source of the problem.

Based on past experiences with low-BTU gas fuel and even with natural gas fuels in some DLN combustion systems, having the proper gas fuel supply temperature to prevent any condensation of liquids as the fuel passes through the knockout drum, coalescing filter, strainer, SRV, GCV, and fuel nozzles is critical to preventing hot gas path part failures.

So, achieving and maintaining proper fuel supply temperatures for preventing the condensation of liquids at any point in the system is of paramount importance.

Lastly, presuming this is a 20 year-old unit and that it's been operating for at least some of that period on some fuel (gas and/or distillate) it's likely that hot gas path parts have been replaced at least once. Were the replacement parts (liners, transition pieces, etc.) purchased from the OEM or from third-party suppliers? Were they new- or refurbished parts?

Please provide answers to all of the questions.

 

First of all, I appreciate the way you responded to Mr. Kashif, its very convincing.

I would like to take on the issue, and reply point wise, anyhow before proceeding forward, please let me explain you. The under discussed GE 9E(PG 9171) unit is new.

Please refer to below response pointwise.

First: The unit was placed on foundation in May, 2008 and commissioning activities started in May, 2009. The unit is believed to be supplied new one. Total fired hours of unit at project site are 1478, and about 75% operation is on partial load. No peaking load operation.

Second: Out of above mentioned 1478 fired hours.
1365 = low Btu gas fired hours
113 = Oil (HSD) fired hrs.

Third: Unit has been designed as per low Btu gas being fed.

Fourth: Please refer to third.

Fifth: Low Btu gas specs are C6+ = 0.14~0.16
Samples of gas were taken at different stages of commissioning and extended analysis (C11+ & C14+)were also got carried out from internationally reputed labs. Although there is variation in C6+ values but its mostly falls under design value i.e. C6+ = 0.16 max.

Sixth: Dew point of gas is about 40`C and when added degree of superheat, it is 65`C min temperature recommended to be fed to GT, while we are maintaining 70`C when it enters SRV.

Seventh: Query is not as much clear to me, anyhow there is no recirculation of fuel gas, but there is vent line as per GE standard, and one additional vent has been installed later on, which only used prior to start up of GT, to release gas at GE Coalescing skid, for achieving recommended temperature for GT feed.

Eighth: Yes, gas manifold is continuous ring with a single entry point connecting between combustor no. 6&7.

Ninth: KODs & additional heaters were installed after the second failure of combustor no. 8 in March this year. Placement and sizing of knockout drum and additional heaters were done by another local company.

Tenth: Since, we have installed KODs we did not experience liquid etc coming along gas, which was experienced and in bulk qty at fuel gas station near GT before the installation of KODs, also we did not see any liquid collection in knock out drums (installed by us on downstream of fuel supplier tie in point, which is about 15 km away from project Site, another KOD is installed at Site). We believe gas supplier has improved their facility.
Previously, we experienced wet GT coalescing filters, but since we have installed KODs, filters are seen in very good condition and dry.

Eleventh: Yes, it is superheated gas, please also refer to sixth.

Twelfth: Water bath heaters (WBH) are installed to heat the gas, and temp at outlet is about 75`C. Piping downstream of additional heaters is about 50% and it is still not insulated, while pipeline downstream of actual heaters to GT is insulated, which is about 50% in length.

Condition of fuel nozzle swirl tip is always found good after the incident, we can assume that fuel nozzle should not be root cause of problem, but we should not eliminate this possibility 100%.

KODs are installed on the upstream of WBH, and please note that our GT is Non DLN.

Your last statement is quite strange for us, since we have purchased a new unit, and we believe it should not have run except factory testing. Anyhow concerning use of third party spares, we did not use any part supplied by third party. GE has always provided replacements of incidents.

In the end, I would like to draw your attention towards an issue, which has been left by Kashif. Deterioration of cowl of liner. I believe this phenomenon start initially, which propagates and leads to complete combustor failure, and it is quite unique. Cowl of liner is found melted every time, it indicates that designed profile of flame inside the liner is not achieved. Either flame is propagated very close to cowl and melts it, or there is some local fire outside the liner. Anyhow it is mystery to be investigated and must be found, in order to run the GE Frame 9E unit safely.


Thank you and best regards,

 

Thanks for all your views

First of all I will clarify a mistake. Control system is Mark-VI not Mark IV, That was a typing mistake. So you will already understand. Gas turbine is erected on site last year, First fire in October, 2009.

Till now operating hours (approximately) are like this:
Total: 1200, Gas: 1050, Liquied: 150
Start up: 65, Trip: 32

First incident was happened after 640 operating hours and second after 885 (843 on gas, 42 on liquied) operating hours and third time after 1200hrs
System is originally designed at Low BTU.normally for base load operation.

Turbine is started at Liquied and Change over to gas later (initially it was around 30MW and after 2nd incident it is 50% load. There is no bypass or recirculation line at Gas line, usually before starting Gas is vented through vent valve (FSV353 at Gas filtering/shutoff /vent valve skid) by forcing a signal from markvi. when temperature is around 75C then GT is started and after 60/70MW it is manually transferred to Gas.During load rejection Auto changeover on liquied.

At Gas manifold entery is from right bottom (looking toward flow direction).At right side entry point hose#6 and left is hose #7
Chamber #6, 7, 8, 9 are at lower elevation

Gas temprature (70 0r 75) is measured at Filtering skid.Gas module (VSR/VGC) are approx 10meter away from that point and piping is insulated.Supply pressure is 25bar and Intervalve presure (VSR/ VGC) is 20bar.

Dear Mr.CSA your remaining queries also will be answered soon.

 

Dear Mr. Kashif;

Thank you for the clarification regarding control MK-VI, otherwise it was creating unnecessary confusion.

And please there is small correction in your response i.e. present GT fired hours are 1478 not 1200.

Regards

 

Thank you for taking the time to respond to each and every point; this is most helpful.

So, this is a new unit, with a Mark VI (not a Mark IV), and it has over a thousand hours of operation, most of it on low BTU gas fuel. (A thousand hours in a year-and-a-quarter is not a lot of hours, and suggests a lot of "commissioning" activities are taking place.)

I'm not a chemist, and I'm not a combustion engineer so I don't know what effect the change in gas fuel composition can have on combustion and performance.

You say the gas fuel pipe is insulated from the water bath heater to the SRV. But, when gas fuel first starts flowing through this line (and we don't know how long that run of pipe is), it's likely that the pipe will be cool and it will take some flow of heated fuel to raise the pipe temperature so there is no fuel temperature drop in the pipe. That's why I asked about some kind of recirculation line, to be used to warm the pipe prior to admitting fuel to the turbine. Because if the pipe is not warmed prior to the flow gas through the SRV and to the turbine, then there is likely some cooling of the gas during at least the initial fuel flows.

Because most of the problems I've seen with "burned" hot gas path parts were eventually traced to the formation of hydrocarbon "chains" of liquid droplets in the combustor from the nozzle to the flame (diffusion flame) and when ignited these "chains" allowed the flame to propagate back to the nozzle the resulting high temperatures caused the damage. (The other problems were the result of deposits (entrained oil that carbonized because of the heat, and other particulates entrained in the fuel that were not removed by filtration) that form on the nozzle tips and eventually ignited.)

The liquid droplets were the result of condensation because of the temperature drops through the gas path, and the final temperature drop at the nozzle was sufficient to cause the condensation to form these droplets which would eventually "link" up to form a path for flame to make it's way back towards the nozzle, and in one case to cause nozzle portions to be liberated resulting in foreign object damage to the turbine nozzles and buckets.

The solution was to increase the amount of superheat to ensure there was no possibility of liquid condensation.

And the worst problems I've seen were on natural gas, but I've also heard of similar problems on low-BTU gas fuel. And, most low-BTU gas fuel is the result of some "by-product" or some process or some other "tail gas" (meaning, it's the leftover gas from some other process) and it's been rather difficult to ensure the gas is of consistent composition.

For the kind of damage you have described to be occurring, you are correct; the flame has to not be in the normal position. Hence, my belief that there might have been some kind of condensation of combustible liquids which allowed the flame to make it's way to an undesirable area and result in damage.

I've also seen damage to combustion liner cowls caused by leakage from the gas fuel tips, which are generally screwed onto the end-cap. If these loosen then gas fuel can exit around the base of the nozzle and enter the liner through the cowl, resulting in damage to the liner cowl and nozzle. There are usually locking plates which are staked to hold the nozzle tips in place to prevent fuel leaks.

Another similar incident was traced to a manufacturing problem which resulted in a very loose fit between the nozzle tip and the end cap which allowed gas fuel to leak out (it still surprises me to this day that the person assembling the nozzles didn't notice that the tips were so loose and just staked the locking rings even though the tips were not tight at all).

Now, why this keeps happening in the same area of the turbine (cans 8 & -9) is another mystery. And we don't know if you've moved the nozzles from 8 & 9 to other cans or not. And we don't know if the nozzles from those cans were replaced after the damage.

Without being able to see the entire installation, piping runs, lengths, nozzles, liners, heaters, coalescers, and the damaged components, it's very difficult to make any more informed analyses.

We also don't know what the source of the gas is and how consistent it is, and if there are any particulates which might be contributing to the problem. One problem was caused by oil from the gas compressor. They were using a lot of oil because of bad rings in the gas compressor, and that oil was eventually causing problems in the combustors! Once the fixed the compressor rings, the combustor damage problems went away.

And, because I'm not a chemist or combustion engineer I can't say for sure what the kinds of changes in composition might have on combustion efficiency or performance.

Kashif stated that the exhaust temperature profiles were very "flat" and didn't really indicate a combustion problem via exhaust temperature spreads as one might expect with such a problem. I would attribute that to the fact that the same amount of fuel was being combusted in each combustor, but that in the combustors which suffered the damage that combustion wasn't occurring in the desired location, as you have suggested, as well.

So, presuming the fuel is of consistent quality and temperature (and I'm not certain of that at all times) that leaves us with leaking fuel nozzles or some nozzle/end cap problem which is allowing fuel to enter the cowl and be burned there instead of inside the liner at a proper distance from the nozzle tip and cowl, or some kind of combustible materials (particulates, liquids) that are forming in the liner and allowing flame to exist in the nozzle tip and cowl area.

But, I'm out of ideas at this point and without a considerable amount of back and forth and some more education on my part about the effects of variations in fuel composition on combustion I'm afraid I don't have much more to offer.

Please write back to let us know how you progress with this!

Actually, there's one more thing that Kashif mentioned, the leaking check valve issue. I wonder if there might be a problem causing a small amount of liquid fuel to dribble into the combustor. The liquid fuel might be igniting or carbonizing and then igniting the low-BTU gas fuel closer to the nozzle and cowl. This is kind of a long shot, but very often these problems are not one single issue, but a combination of issues that combine in very unusual and intermittent circumstances and can be very misleading.

There have been reports of leaking liquid fuel check valves allowing pressure to flow back towards the liquid fuel stop valve and through the flow divider and push liquid fuel out of other nozzles, and cause carbonization at the liquid fuel nozzle tips. One wonders if it might be possible for something like this to be occurring where there is some low-BTU gas fuel characteristic which is causing a near condensation of some liquids and combined with carbonization of liquid fuel at the nozzle tip might ignite the liquids and cause the kind of problem you are describing.

It's a very long shot, and it might not be the exact mechanism in this case that's causing the problem but it might also lead to some other investigative and troubleshooting steps that might resolve the issue if not lead to the discovery of the exact cause.

If I think if anything else, I'll write back again. Best of luck!

 

I am again so thankful for your detailed response;

Concerning the queries asked / raised from your response, please refer to below;

There is one vent line installed just upstream of SRV, which now we use to vent the gas to achieve the required temperature for GT feed, and pipeline downstream of SRV & GCV to GT is very short in run, since the gas module is installed very close to GT compartment, additionally the purge air coming from CPD keeps the same hot during fuel oil operation. Therefore we can say there is no part of pipeline downstream of WBH to GT fuel nozzle, which is cool, and can be the cause of possible vapor / liquid formation inside the pipeline, unless the calculated the degree of gas superheat is correct.

Also, you are talking of deposits on fuel nozzle swirl tip i.e. entrained oil, if your assumption is towards lubricating oil of compressor, then there is no chance of lubricating oil entrained to GT combustors, since the gas is supplied at natural pressure, no means of artificial compression is used.

The gas of this project is not a by-product or permeate gas, but it's a dedicated gas for this project, coming directly from natural wells.

Your indication towards leakage from fuel nozzle is quite logical and understandable to me. In this regard all fuel nozzles have already been sent for leak test. Awaiting the results.

Regarding swap of nozzle from #8 to #9, yes it`s true we did it after Mar, 10 incident, but the replaced nozzle was new one, but nobody can assure that nozzle was perfect, but previously in Jan, 10 the damaged nozzle was got repaired/rebuilt due to long lead time for supply of new nozzle and was used and same combustor #8 again failed in Mar, 10, meanwhile there was free water and condensate etc observed in gas, therefore investigations revealed us that the cause of failure is low temperature of gas and condensate etc, while during incident of Jul, 10 there was no sign of liquid etc, and fuel gas temperature was as per requirement, therefore we call it mystery.

Regarding fuel gas temperature, last time it was consistent, while for quality of gas, I would say it`s a vast term, if you are talking about the liquid etc, then there was no liquid etc reported in the gas. As Kashif already stated that there was nothing as such reportable variation in the GT exhaust temperature profiles, it was almost normal, therefore it also supports the stance that gas quality was good.

Concerning dribbling of liquid fuel at fuel nozzle tip, it is not as such convincing to me, because when I recall the scheme of fluid flows at the swirl tip of fuel nozzle it`s like, liquid fuel is at centre (inside all), then its water injection jacket, atomizing air and at very outer its fuel gas. There are four fluid ring ports at swirl tip, and nothing is spare, either working fluid or purge is always flowing with adequate pressure. If fuel oil droplets by any reason reach to swirl tip along purge air, the droplets should burn like mix flow instead of falling down. Anyhow this could be my understanding only.

Regarding reverse flow of liquid fuel to stop valve and then to flow divider and ... , because of leaking liquid fuel NRVs. This is also not as such convincing, since there is purge air in this circuit, which should cause the above phenomenon, not the CPD which has not contact with NRVs , and purge air leakage from NRVs should not be dangerous , since very little quantity of purge will reverse through NRVs and major will reach the swirl tip and fuse into flame.

One thing very important that atomizing air compressor must be working and developing adequate pressure, otherwise it could be dangerous.

Thank you and best regards,

 

Zahid Hussain,

I will take exception to a couple of your statements. With regard to exhaust temperature being an indication of gas quality, I disagree. Exhaust temperatures will indicate whether or not more fuel is being burned in one area versus another (which will cause a higher temperature in that area) <b>OR</b> whether there is a leak of CPD into the hot gas path (which will cause a low temperature in that area versus another).

Because of the ring nature of the gas manifold, it's highly unlikely that some bad quality gas will make it into one combustor and not any others. Also, if the unit were operating at Base Load (on exhaust temperature control) and the gas quality was low then it's likely all the exhaust temperatures would be low, not just one or two or three, then the Speedtronic would just increase the gas flow to make the actual exhaust temperature equal to the exhaust temperature reference. (That's if the unit was operating at Base Load, on exhaust temperature control.)

On the topic of leaking Liq Fuel check valves flowing in the reverse direction, please review the Piping Schematic Diagrams (P&IDs) for the Liq Fuel- and Liq Fuel Purge systems. On every GE-design heavy duty gas turbine running on gas/distillate I have ever worked on when the unit is running on gas fuel Liq Fuel Purge Air is being continually supplied to the area downstream of the Liq Fuel check valves to purge the liquid fuel out of the nozzle so that it doesn't carbonize. (Also, there is CPD that the Purge Air has to overcome which is acting on the Liq Fuel check valve even if there were no Purge Air, which there should be.)

If any of the fourteen liquid fuel check valves leaks in the reverse direction, then Purge Air/CPD ("pressure") will flow backwards into the liquid fuel line, back to the Liq Fuel Flow Divider, and has been known to "push" liquid fuel in the other liquid fuel lines out through the nozzles, usually dribbling and causing carbonization of liq fuel on the nozzle tips.

If the reverse flow through the Liq Fuel check valve(s) gets high enough the Liq Fuel Stop Valve of Some Frame 9Es have even been known to be pushed open and air flow into the Liq Fuel supply piping. This is an extreme case, and would indicate the "failure" of two components (the Liq Fuel check valve(s) AND the Liq Fuel Stop Valve), but I have personally witnessed this at one site and I was personally surprised to have seen it.

This would be an extremely unlikely contributor to the problems at your site, but, on problems like this I have usually found that they are not the result of one simple cause. Rather, two or more conditions must exist in a "perfect storm" scenario for the damage to occur. Which is one reason why they are intermittent and difficult to reproduce or monitor as they are, in a sense, fleeting.

Again, I'm out of ideas here, but would greatly appreciate it if you could keep us up to date on the progress and, hopefully, the resolution of the issue.

 

Dear Sir / CSA

What I understand a bad quality gas is, an off spec gas and carrying undesired elements such as liquid/condensate or heavy hydro carbon beyond permissible / design limits as already communicated. I do not believe / understand that with only little change in BTU/unit mass or volume, which is majorly varied with methane content, gas could be called bad quality.


If bad quality gas as mentioned above is received. Studies / investigation have revealed to us that it will not distribute equally in all combustors but the lowest will be affected, and it can be observed at exhaust temperature, and the originator has already mentioned that in Mar, 10 there was one exhaust thermocouple showing abnormal values.

I totally understand and agree with your explanation for gas heating value and base load operation.

Concerning the leakage of CPD into HGP, as per my experience it is majorly caused by improper fitting of TPs with S1N, but it should not be an intermittent problem, ones it is detected then there is no remedy until proper fitting is ensure. And I believe cooling of HGP components from CPD should not be cause of such leakages.

On the issue of liquid fuel check values leakage and CPD , although there is different opinion, anyhow I respect your opinion.

If we come to the conclusion for this discussion for finding possible root cause (s) of problem faced at our site. I fully support, as you mentioned time and again from your experience that "contributor to problem at our site is two or more than two" and every possible cause should be given due concentration, to reach a permanent solution to the problem.

Thank you & best regards,

 

Dear Mr. Zahid Hussain,

Actually sir, Hours I mentioned just was estimated by me, Actual data I also collect later. And my above post (to clarify mistake) appeared very late. So, sorry for confusion

And about Fuel nozzle check valves, I think that Fuel oil NRV(VCK1-1-) will not cause trouble, because During Fuel gas operation there is purge air in that line. before Fuel oil NRV, line is fill with oil, Air can not travel toward Fuel oil circuit, may be minute quantity mix with purge air and go toward nozzle until fuel line is depressurize.

If fuel oil purge air NRV (VCK2-1-) is passing and GT is on fuel oil then oil will be passing back through NRV and will collect in tubes and manifold of purge air. if passing is in big quantity then it will drain through purge air control valve(VA19-1). (VA19-1 and VA33-1 are three way type valve, if the purge air is off then, it is open toward drain, i.e is called tallet leak off, any liquid coming from these point indicates purge air NRV are passing).

Then during change over from fuel oil to gas, remaining oil will travel in nozzle with purge air and will burn in nozzle, I think oil will might go in few nozzle. so in this condition there will be exhaust temperature momentary high in few zones. but in this case I did not see that Ex. temp are going high.

And in case of water injection, if water injection NRV(VCK8-1-) is passing and water injection is off then there is purge air that can leak toward water injection manifold. only in start up when purge air pressure is not enough, then there is chance that oil come back through fuel nozzle and come in water injection manifold as well in water injection purge air manifold (if water injection purge air NRV is also passing). and during water injection mixture of water and oil will in nozzle. I think this will also not cause any disturbance, And if oil is collected in water injection purge, when purge air is on then it will go in nozzle and. If water injection purge air NRV(VCK2-21-) is passing and water injection is on then water will collect in purge air manifold. In case leakage is high then it will drain through purge air control valve(VA33-1).

Moreover we can not neglect this issue. This is also happening frequently, one thing still is not clear that incident is caused by gas or fuel oil

 

Messrs. Hussain and Kashif,

Every GE-design heavy duty gas turbine configured for gas/distillate operation on conventional fuel nozzles (non-DLN combustors/nozzles) has had the following configuration (as best as can be drawn using simple text characters in this forum):
<pre>
--> (Direction of flow)
Liq Fuel ->-----|/|-----------> (fuel nozzle tip)
Liq Fuel |
Check Valve |
|
_
Liq Fuel / ^
Purge Check _ | (Direction
Valve | | of Purge
| Air Flow)
|
^
|
Purge
Air
From
VA19-1

</pre>
In the above "sketch" (which should look like the one on your unit, when the unit is running on gas fuel there should be no Liq Fuel flowing through the Liq Fuel Check Valve and Purge Air should be coming from VA19-1 (which should be in the "purge" position to allow air from the Purge Air Pre-cooler through VA19-1 and to the Purge Air manifold and to each of the Purge Air lines, through the Purge Air check valve. The purge air should enter the Liquid Fuel line downstream of the Liq. Fuel Check Valve, which means there would be "back-pressure" on the Liq Fuel Check Valve.

When the unit is running on Liq Fuel, there will be Liq Fuel flowing through the Liq Fuel Check Valve in the forward direction and putting back-pressure on the Purge Air Check Valve. VA19-1 should be in the "vent" configuration, allowing any liquid/air from the Purge Air manifold to drain out of the Tell-tale Leak-off line. If the Purge Air Check Valve(s) are failing and allowing Liq Fuel to flow backwards through the check valve(s), then there will be Liq Fuel flowing into the Purge Air Manifold, through VA19-1 and through the Tell-Tale Leak-off.

It's sometimes necessary for one to use two or more Piping Schematic (P&ID) drawings to properly link them all together (one drawing says this line comes from another drawing, and that line goes to another drawing, etc.). But, unless GE and/or its packagers are doing something very unusual, it's difficult to believe that they have changed their piping arrangements from a decades-old, tried-and-true, proven design (though with the emphasis on cost-reduction and simplification and with the young engineers with little or no field experience reviewing and re-designing some of these systems, just about anything is possible these days; GE used to be "If it ain't broke, don't fix it!" and now they are about, "Reduce the cost at just about any cost; break it to cost-reduce it if necessary!").

Now, if the Liq. Fuel Check valve is failing and allows the Purge Air to flow backwards through the Check Valve, then if there is sufficient flow the purge air, which is at a higher pressure than CPD *and* at a higher pressure than the cracking pressure of the Liq Fuel Check Valves, will push Liq Fuel back through the Flow Divider and out at least one of the other Liq Fuel Check Valves. This usually results in a "dribbling" of Liq Fuel out of the Liq Fuel nozzle tip, which usually results in carbonization of the fuel in the nozzle tip or on the face of the nozzle tip.

This has been documented to have happened on more than one GE-design Frame 9E heavy duty gas turbine. It is not a normal occurrence, but it has happened.

You say you are plagued with check valve failures. Have you tested the check valves in BOTH directions and for opening at the proper cracking pressures in the forward direction?

You both say you don't believe this could be happening, but you won't rule it out, yet you are both "discounting" it. I'm <b>NOT</b> saying this is happening at your site, only that it has happened on other sites. A LOT of check valve failures can be directly attributed to improper flushing of the lines prior to initial operation--especially for Liquid Fuel Check Valves and Water Injection Check Valves! Both of these systems are at least partially assembled in the field, and in the case of the Water Injection system a lot of the piping is installed in the field. I have personally seen cigarette butts, weld slag, metal filings, pieces of gasket material, rocks, dirt, washers, nuts, screws, even lathe turnings flushed out of field-installed piping. Even on the Liq Fuel tubing installed between the Accessory Base (where the High Pressure Liq Fuel Pump and Flow Divider are usually located) and the Turbine Base! It's simply amazing what comes out of these lines when they are <b>properly</b> flushed during commissioning (which doesn't happen very often these days, with rampant schedule delays and rampant budget and planning shortfalls and alleged "clean piping installation" practices being used by less-than-skilled labor under less-than-experienced supervision and politics driving a lot of commissioning and start-up practices).

And in closing, in what will be my last posting on this thread, you, Kashif, have basically posted two issues and have tried (rightly or wrongly) to associate them. It's not clear if they are related or not. From what we have been able to extract from the two of you, it appears that there is some combustible "material" which is igniting in or near the cowl of the combustion liner and that is causing the burning issue. Why it's only happening on one or two combustors, at the "bottom" of the machine, and near where the low-BTU gas fuel enters the gas fuel manifold, is not clear. What the combustible "material" is and where it comes from is not clear. The source of the check valve failures is not clear, though check valves have been a thorn in many dual fuel (gas/distillate) sites for many years. In some of the GE-design heavy duty gas turbines, they have gone to air-operated valves instead of check spring-loaded check valves, which adds many more tubing lines to the turbine compartment, adds more maintenance-requiring components (the new air-operated valves), adds additional solenoids (to control the air flow to the air-operated valves) and in some cases requires instrument air from a source other than axial compressor discharge extraction. Imagine adding fourteen more tubing lines from another manifold with another supply from another solenoid-operated valve, for (1) the Liquid Fuel "check" valves, (2) the Purge Air "check" valves, (3) the Water Injection "check" valves, and (4) the Water Injection Purge "check" valves (if so equipped). Yep; pretty crowded and complicated.

There has been a vendor which has posted to control.com in the past indicating they have worked with GE to develop and supply a check valve design that has alleviated some of the problems, if not all of them for some units. You can use the 'Search' feature to find contact information.

Best of luck to the two of you! Again, there are turbines running low-BTU gas fuel and distillate fuel successfully around the world, and more are being converted or installed each year as companies are trying to find cheaper sources of fuel and GE is trying to apply their machines to these previously unused (on combustion turbines) fuels. Patience and a logical, methodical approach is necessary, as well as an open mind and a willingness to hand-draw some systems and an application of some good, old-fashioned analysis and problem-solving techniques

 

Dear Kashif,

If you can remember, once unit was tripped on high exhaust temperature, but I believe with the sudden release of unwanted/uncontrolled FO, as described in your response will cause the unit to overspeed rather high exhaust temperature trip.

CSA to respond over it.

Regards

 

The configuration as drawn by you, for liquid fuel and purge air is fine and correct.

The cracking pressure and dribbling of liquid fuel out of the nozzle tip,as described by you is unclear and not understandable to me. What I understand from these two terms is as under;

Cracking Pressure: The pressure at which NRV starts opening in right direction.

Dribbling: The dropping of fuel oil droplets at the fuel nozzle tip.

Concerning the inspection of check valves, yes we have been checking in both direction, and the check valves have failed all the times, whenever inspected either in right or reverse direction, although the no. of failed valves was different each time.

The check valves have been replaced more than two times, and still the problem is persisting, I believe this rules out the possibility of improper flushing of pipelines / fittings etc.

Anyhow, as you already mentioned the reason of check valves failure is not clear to you, the same is case here.

Regards,

 

Dear CSA,

Check valves always checked in both direction, mostly if one was passing in reverse that was also passing in forward direction below cracking pressure.

And about flushing of system, Fuel oil system was flushed utpo fuel nozzle, more over if there was some undesired material (like debris or rust), It may be pass through once. but check valve passing is occurring again and again. And same about purge air check valves.

I am not saying that combustor damages are due to this problem. But that is thing which is frequently happening along with combustor damages

Yes, I also believe that this is happening due to some undesired burning, Flame is deviating from its original place or there is another addition flame outside the liner. why? That still to find
I appreciate all of your response in this matter. At one side vendor is trying find out causes of this failure, they are designer and are more experienced. Hopefully earlierr or later they will find out it.

But I would like to keep it up at this forum, all of our discussion may not give the solution but I am sure it will give an opportunity for discussion, experience sharing, and most of all how to understand and deal the things and issues, this is very much helpful to me as well as to other readers of this thread

And about exhaust temperatures, remember once unit was tripped at High exhaust temprautres. may be it was started after 1st incidents repairs. It was happened during fuel change over from HSD to Gas at FSNL. Later on this was declared not recommended. Now change over is possible after 50% load)

On 17 Feb, 2010, only once in start up GT trip on High exhaust Temperatures. Then inspection of few nozzles and chambers were carried out. No any abnormality was found at that time. After that it was not happened again. And moreover On 01 March, 2010 At that time during base load operation Gas Turbine exhaust temperatures at one thermocouple was high, temperature at TTXD-15 was 588C (alarm is at 585). Because Exhaust spread was within allowable spread, that’s why it was not noticed. After that during shutdown when borescopic inspection was performed comb. 8 was found damage. Keep in mind at that time liquid/condensate was coming with gas.

Dear Mr. CSA, I have uploaded few pictures and P&I diagrams. I hope that it will help you to understand the issue

http://www.4shared.com/photo/E6e2kCl9/1st__incident-1.html
http://www.4shared.com/photo/LnbhnV_T/1st__incident-2.html
http://www.4shared.com/photo/xYe7gSm8/1st__incident-3.html
http://www.4shared.com/photo/LkQCAffr/1st__incident-4.html
http://www.4shared.com/photo/cSEeIcpJ/1st__incident-5.html
http://www.4shared.com/photo/rsb7eVSR/2nd_incident-1.html
http://www.4shared.com/photo/OhJ0BhBX/2nd_incident-2.html
http://www.4shared.com/photo/be6SijPm/2nd_incident-3.html
http://www.4shared.com/photo/3mcKg_gS/2nd_incident-4.html
http://www.4shared.com/photo/DL_bsAoF/2nd_incident-5.html
http://www.4shared.com/photo/gD54SiSf/3rd_incident-1.html
http://www.4shared.com/photo/LEmgsDpp/3rd_incident-2.html
http://www.4shared.com/photo/TLUEWLLw/3rd_incident-3.html
http://www.4shared.com/photo/Zs-fNzFu/3rd_incident-4.html
http://www.4shared.com/photo/2g_wWPJw/3rd_incident-5.html
http://www.4shared.com/photo/XE8adjEl/Atomizing_Air.html
http://www.4shared.com/photo/j7ibkmo9/Comparison_End_cover_1st_and_2.html
http://www.4shared.com/photo/HVKchNES/comparison_liner_1st_and_2nd.html
http://www.4shared.com/photo/nM0yelKR/Comparison_Nozzle_1st_and_2nd.html
http://www.4shared.com/photo/qe4Z87Q5/comparisonflow_sleeve_1st_and_.html
http://www.4shared.com/photo/2cUI_Kwk/Drain_sump_system.html
http://www.4shared.com/photo/vGdSl-yD/Fuel_Gas_filtering_skid-Coalsc.html
http://www.4shared.com/photo/OV-q6icj/Fuel_gas_station.html
http://www.4shared.com/photo/0tWNi5Nm/Fuel_Oil_filtering.html
http://www.4shared.com/photo/JvlCZ_u3/Fuel_Oil_forwading_system.html
http://www.4shared.com/photo/bN_aZiEh/Fuel_oil_heater.html
http://www.4shared.com/photo/lC8t07I3/Gas_module.html
http://www.4shared.com/photo/Ve2T1v5x/Gas_Purge_air.html
http://www.4shared.com/photo/PtQIKIMJ/GT-Fuel_oil_supply.html
http://www.4shared.com/photo/43jjBkqH/Manifold_gas.html
http://www.4shared.com/photo/ZZedePUl/water_and_fuel_oil__purge.html

Thanks
Regards

 

I would like to discuss the rule of atomizing air in this type of failure.
what could be wrong with atomizing air, which can cause damage to machine

Thanks

 

Atomizing air have very important character in gas turbine operation.Any trouble in Atomizing air parameters can cause improper combustion.As per this type of trouble...Yes it could be.If supply of atomizing air is not proper to fuel nozzle,it will not properly atomize the liquid fuel.Result will be unburned fuel.This will disturb many things.

1-Exhaust temperatures
Unburned fuel will continuously burned until it have energy.so if it burn after the turbine or within turbine,it will change the exhaust temperatures..

2-Affects on hardware
in starting unburned fuel will can collects in lower elevations and can cause a separate spot firing.it will damage the part where it will happen

3-Heat rate
improper burning of fuel oil will change the mass flow, contents of exhaust,emissions

 

Dear all

I have deleted all above mentioned pictures, thanks for all participants to take interest in this thread. All of this discussion was very interesting to me.

Thanks all

 

Hello Kashif,

I happened upon this discussion this morning and wondered how your machine is performing in the past few months. Your last post was in fall of 2010. As a supplier of fuel system components which afford dual fuel gas turbine owners with reliable operation on the back up liquid fuel, I found the exchange quite interesting and wanted to weigh in on my experience with the back up liquid fuel system. I would suspect that the issues associated with the nozzle failures have been resolved. However, unless you have performed an upgrade on the check valves, problems with failures are probably still occurring.

Specifically, on the issue of liquid fuel check valve failures, solidification of liquid fuel or coking is the number one cause of failure. It is important that the impact of such leaks not be under estimated. Nozzle tips, fuel distributor valves and check valves can be exposed to temperatures which exceed 1000 degrees "F" when such leakage occurs. A key point to consider is that purge air, in order to be effective, has to be high enough to overcome "CPD". When two or more liquid fuel check valves fail to seal in the reverse flow direction, a migration of combustion gas from the higher pressure combustor to the lower one is initiated. The flow divider acts as the manifold in such cases. Check valve leaks cause the pressure at this point to become lower than CPD. When this happens, combustion gas travels upstram through the nozzle. The severity of the problems which occur in the aftermath of this malfunction is contingent upon how many check valves are leaking and the condition of the stop valves upstream. I have spoken with a customer who observed all of their distributor valves glowing cherry red when operating on gas fuel. Inspection of the fuel end covers after removal revealed that the poppet in the fuel distribution valves had completely melted away.

While you will need to gauge the impact of this phenomena on your nozzle issue, new technology has been developed to address the issue. Specifically, JASC Controls water cooled liquid fuel check valves were designed to prevent coke formation. The elimination of coke allows the valve to maintain an ANSI class 6 seal in the reverse flow direction. This prevents the migration of combustion gas and keeps the liquid fuel system from evacuating so that the ability to transfer from gas to liquid fuel is close to 100%.

A thermal relief valve is also available. Its function is to relieve fuel pressure due to thermal expansion into the fuel sump rather than into the fuel nozzle. This pressure from thermal expansion is high enough to exceed CPD. While stop valves are design to prevent pressure in the forward flow direction, they are not designed to see pressure in the reverse flow.

I hope that this information helps with your perspective on the potential impact of combustion gas which migrates upstream.

 

Hello Kashif,

This is to Khalif, wondering what was wrong, and how it was fixed, just curious and very interesting thread. If you have time please post your findings. Is GT running?

Thanks