4_20ma,

You say the unit at your site has an adjustable torque converter (Torque Adjuster). If that is true, then I would say the Torque Adjuster limit switches are *NOT* adjusted properly. I'm making this statement with the following caveat: It is a typical GE-design Frame 9E heavy duty gas turbine with no unusual auxiliaries, fuels, or sequencing.

The Torque Adjustor should be set to "purge" speed position to cause the unit to accelerate to *approximately* 20% speed during starting (the actual speed is irrelevant; it's supposed to be 150-160% of rated starting motor current) for the purge period. Once the purge period is complete, the unit should decelerate to 9.5% speed (actually the speed will dip slightly below 9.5%) and the torque adjuster should be set to "firing" speed, which will slowly accelerate the unit up through 10% speed at which time the fuel stop valve is opened, the fuel is set to firing value, and the spark plugs are energized.

When flame is detected, the fuel is cut back to warm-up value and after a short time delay the Torque Adjuster will be pulsed to "accelerate" position to begin accelerating the turbine-generator shaft. After the warm-up period is complete, the fuel will be increased to assist with accelerating the turbine-generator shaft. At approximately 60% speed the torque converter will be "de-pressurized" and the Speedtronic will use fuel control to complete acceleration to FSNL.

The above describes the typical starting sequence for many GE-design heavy duty Frame 9E gas turbines with adjustable torque converters (Torque Adjusters). There are some Torque Adjusters which have three settings (firing; purge; accelerate) and some which have only two (firing; accelerate). (The ones with three settings will have a limit switch input to the Speedtronic, usually 33TM-5, driving L33TM5P; those with only two positions will not typically have this input.) But, the basic operation is still similar to what's described above: purge at a high speed, decelerate to below minimum firing speed drop-out level; slowly accelerate back through minimum firing speed at which time fuel will be admitted and spark plugs will be energized; once flame is detected fuel will be cut back and shortly after that the torque adjuster will be pulsed to "maximum"; when the warm-up time is complete the fuel will be increased to assist with acceleration; at 60% TNH the torque converter will be de-pressurized and the Speedtronic will use fuel to complete acceleration to FSNL.

Adjusting the Torque Adjuster limit switches can be a daunting task the first time it's done; but if one has patience and makes notes, it's much easier to perform later. One has to understand how the Torque Adjuster mechanism works, which limit switches serve which purpose, how to change limit switch settings, and how to manually manipulate the mechanism. One very important thing to note is that one should <b>NOT</b> close the torque converter guide vanes with 20TU-1 energized; this can damage the mechanism, so if adjustments are being made with the Starting Motor running one needs to force 20TU-1 to logic "0" while closing the torque converter guide vanes.

One needs to have someone stationed at the operator interface who is familiar with forcing logic, and then someone needs to be at the Torque Adjuster mechanism. Someone also needs to monitor the current drawn by the Starting Motor. So, it really takes three people. It's loud and noisy and hot in the Accessory Compartment, and there is a lot of management pressure to complete the activity quickly.

No; I know of no procedure for adjusting the limit switches properly. And without being able to be on site to review the configuration and demonstrate how to adjust the limit switches it would be futile to try to write one and post it here on control.com.

Lastly, the difference between FSKSU_FI and FSKSU_WU is somewhat large. This may be due to someone increasing the firing FSR because the speed at which you say firing is occurring is higher than it typically should be, and that usually requires more fuel because of the higher air flows at the higher speed. Firing at a speed higher than approximately 10-12% increases the thermal stresses on the hot gas path parts.

I would also estimate that there may be some combustors which lose flame (as evidenced by high exhaust temperature spreads) when fuel is cut back to warm-up. That's just my guess based on past experience with the differentials similar to the one you describe.

This observation about high exhaust temperature spreads during warm-up and acceleration is just that: a personal observation, made without any data, just personal experience. It doesn't mean the unit isn't working properly, it's just an observation. But, any start attempt that results in flame being established is not a proper start, either.

Hope this helps!

 

Dear CSA
This is fantastic, The beauty of this forum is YOU and this gas turbine complexities.Here we go...

As you told we have got very less flame something less than 100 always some times it goes even below 50 counts and we have replaced that particular sensor multiple time and we checked that loop also but still is persisting.

Coming to torque adjuster ,Rightly said ,Thats really a big thing to adjust because our machine is running with those control constsnts for long time, so no one is really bothers about firing at high speed, GE might have left it like that i suppose or GE engineer has to come and do it for the first time.

I am going to get some more knowledge regarding this one because i got some questions.

With one Limit switch How do you manipulate those there operation (purging,firing &acceleration)?

What exactly does that limit switch do?As per our CSP, If limit switch logic becomes 1 and 20TU-1 De energized with 5 sec delay it actuates automatic shutdown(L94AX1), and it generates an alarm(Torque converter drain valve trouble)

I really don't know what is behind the same firing FSR for both liquid and gas fuel with two different valves(LFBV&GCV) ,can you please explain more about this?is there anything with the valve sizing?

I am eagerly waiting for our reply

Thank you so much

 

4_20ma,

Again, just because a turbine starts successfully doesn't mean it's starting correctly, or that the start is the best for the machine.

I suggest you find the Control Specification drawing that was provided with the Speedtronic panel. It's usually in the Service Manuals provided with the unit. There will be a section in there that talks about how the starting sequence should occur, and it may have specific information about how the limit switch<b>es</b> should be set.

Another good place for basic information about limit switches is the Starting Means P&ID (Piping Schematic).

That is 'switches', plural, because there will be either two or three involved in the starting sequence, depending on the vintage of the machine. There are also several other limit switches in the Torque Adjustor, but they do NOT require any adjustment as they are they to protect the mechanism from overtravel and from overtorqueing.

One switch will be adjusted to set the 'firing' speed setting; one switch will be adjusted to set the 'purge' torque setting (if so equipped, and this switch is usually identified as 33TM-5), and another will be adjusted to set the 'acceleration' or 'maximum' torque setting.

To see how the limit switches interact with the "control system" (which includes the Torque Adjustor Drive Mechanism MCC starter *AND* the Speedtronic) panel, you need to have the Torque Adjustor Drive Mechanism MCC starter elementary/schematic.

The Torque Adustor mechanisms are usually German-made, either by Siemens (yes, they put Siemens equipment on GE machines--for a while!) or Auma. They are very good mechanisms, and can be set for repeatable and reliable operation.

I have seen Torque Adjustor Drive Mechanism manufacturer's instructions, with lots of diagrams and drawings, in the Service Manuals. Unfortunately, they are usually in German; but fortunately, there are enough pictures and diagrams that one who doesn't read or speak German can usually understand what's being presented.

Again, it takes several people to properly adjust the limit switches, and they all need to understand what's happening and what they're part in each step of the adjustment procedure is. Because the electric Starting Motor is usually a high-voltage motor, the number of starts is limited, BUT with proper planning and understanding, it's not necessary to start the motor more than once. One just needs to use 20TU-1 (forcing it to "0" and "1" as appropriate) to keep from having to have multiple starts on the motor.

It's best to have someone have a look at the Torque Adjustor limit switches, with the cover off, with the unit at rest (not on Cooldown), and manually operate the mechanism to observe the limit switches and how they operate, and to identify the ones which need to be adjusted. It's also a good idea to document the as-found condition (where each switch actuates with respect to at least one end of travel; for example, switch 33TM-7 changes state at 1-1/2 turns from full closed; 33TM-3 changes state at 9 turns from full closed; and 33TM-8 changes state at 11-1/2 turns from full closed).

If one can get familiar with the limit switches and how they operate before starting the unit, that should make the process go a lot easier. I would suggest scheduling this two or three times before actually attempting to adjust the limit switches, to make drawings, take photographs/videos, return and make sure that one's understanding is correct, and make some notes for when the adjustment is actually done with the unit running.

Note that this will all likely need to be done with the Torque Adjustor Drive Mechanism MCC starter in the OFF position, or with the main breaker of the Torque Adjustor Drive Mechanism MCC starter in the open/OFF position. Otherwise, the Speedtronic will be trying to return the mechanism to a "known" position in preparation for the next unit START.

With respect to the flame detector intensity, it's not clear if you are referring to loaded, full speed operation, or if you are referring to the period during warm-up and acceleration. Many things can affect flame detector sensitivity, including cooling water flow-rate (if the detectors have the cooling water coils around the body of the flame detector), and the angle/position of the tube to which the detector is attached, as well as the inside of the tube to which the detector is attached. If the tube is not aimed properly or is bent, that can adversely affect flame intensity. If the inside of the tube to which the detector is attached is rusty or dirty or somehow obstructed that can also adversely affect flame intensity. Any of these conditions listed will usually adversely affect flame detector intensity at all speeds, not just at low speed or during firing/acceleration.

I've never seen the alarm you have described, and without being to see all of the CSP it's not possible to make any comment about that.

 

Dear CSA

Sorry for belated replay, after going through your last reply. as you said i opened starting means P&ID and i found clear picture about this mechanism (In my last post i said about one limit switch that's not 33TM-5. Instead it's 33TC-1 (Torque converter unloading valve limit switch) that's my mistake, i misunderstood)

Yes i found 2 limit switches 33TM-5 and 33TM-6 which is embedded with AUMA torque adjuster motor which again connected to "torque adjuster" control MCC not into MARK V (But 33TC-1 is connected with MARK V with the logic which i have said in my last replay)
As per device list

33TM-5=Torque adjuster limit switch -HIGH torque limit
33TM-6=Torque adjuster limit switch -HIGH torque limit

Upon studying the MCC elementary drawing i found that "torque adjuster" motor is a two directional motor (Cock W&Counter Clock Wise Rotation I suppose)

I went on to read Torque converter control BBL block(TMGVCV1)I found something interesting.

the control constants are
TMKGV_1=50 DGA (Turbine vent Torque converter guide vale angle)

TMKGV_2=15 DGA (Turbine Firing Torque converter guide vale angle)

TMKGV_3=43 DGA (Turbine Turning gear Torque converter guide vale angle)

TMKGV_4=46 DGA (Turbine Maximum Torque converter guide vale angle)

I have seen from my old real time trends that this maximum angle went to 67 DGA and it has been happening for all the start ups i have witnessed but here constant(TMKGV_4)is 46 DGA.How this is happening?

TNK14TM1=10.00 % (Minimum firing speed)
TNK14TM2=18.00 % (Maximum firing speed)

What is the difference between the 9TNK14HM1 & TNK14HM2) and (TNK14TM1 & TNK14TM2)?

From the above angle constants Raise (L60TMR) or Lower (L60TMR) command signal directed to "Torque Adjuster" MCC.

Some of the above things are confusing me greatly i am looking for your valuable replay to get cleared.

 

4_20ma,

It would seem that the chaps at GE Energy Products-Europe in Belfort, France--never satisfied to leave well enough alone--have succeeded once again in over-complicating what should be a relatively simple operation.

I'm afraid that without being able to see the application code for the unit at your site and without being able to see the Control Specification document that I won't be able to offer you much in the way of help or support.

It would seem that there is some kind of position feedback device from the AUMA torque adjustor mechanism, and that the Speedtronic will attempt to put the adjustor mechanism (torque converter "guide vanes" (nozzles)) in certain positions at certain times.

It would also seem that there is some logic for a low-speed start (the normal 10% speed) and some logic for a high-speed start (the 18% speed setting). How that selection or determination is made is unclear to me without being able to see the application code. I might even suspect that it is operator-selectable from some HMI display, but that's just a guess.

This is about all the commentary I can offer, again, being confounded by the Belfort contingent in their overwhelming desire to improve on the proven. Have you perused the Operations and Maintenance Manuals provided with the turbine-generator to see if there are any details about this starting scheme in the Manuals? In manuals provided by GE-USA, there used to be a System Description "chapter"/"tab" for every system that had a Piping Schematic (P&ID). Sorry; I'm not familiar with what GE EPE provides these days, but I can say that some of what they passed off as manuals a few years back were multiple volumes (14 or more for one unit!) of pretty poor information; hopefully things have gotten better in recent years.

Wish I could be more help, but, you seem well on your way to understanding the various drawings and documentation. Keep up the effort!

 

Dear CSA,

Sorry to keep you waiting. Our outage begins today. So i took the opportunity to test a few parameters prior to starting up the GT on distillate. First thing first, we blanked off the check valve after the atomizing air booster and installed a 3 bars pressure gauge before the check valve and turned on the AA booster motor manually through MCC. Pressure gauge reading was at 0.1 bar and motor amp for 3 phases is 8amp. We changed the cable terminal so that the pump would rotate on the opposite direction to check. Pressure gauge reading -0.1 bar and 9amp.

Second thing we did was to open up the FA-3 and FA -4 plug to drain off water in the porous stone filer. FA-4 contained alot of water(water flowing out for 30 seconds)but no water found from FA-3.

Our spark plug is installed in can#13 and can#14. We opened up the spark plugs and forced the logic to energize them. The sparks were similar. Both with same spark intensity. After checking all that, we proceed with the test.

Firing occurred at 10% speed. TNH dropped to 9.5 % and then went up to about 10% and firing was started. TNK14HM1 and TNK14HM2 are 10% and 9.5% respectively. During firing, FQLM1 was 0.46kg/s while FQROUT was 2.88%. FAL was -17.99. Our FSKSU_FI_LIQ during firing was 24%. After 1 minute, the process alarm failure to ignite was annunciated. Diagnostic alarm annunciated was IOMA Power supply out of limits P15.

The lack of air pressure of the AA booster pump is very worrying. Can we just increase the firing speed or decrease the FSR during firing when starting up on distillate?

We are gonna proceed to test GT2 AA booster pump to check the motor amp as well.

 

RY,

Thanks for the feedback.

So the spark plugs seem to be working fine before firing on liquid fuel; it would have been interesting to know how wet the plugs were after the firing attempt. If there's too much fuel, or it's not atomized properly the plugs can be "flooded" and not work at all.

Doesn't sound like the AA Booster Compressor is working very well, for starters. I still maintain that when run in the reverse direction the current drawn by the motor will be very high--presuming that the compressor is mechanically sound and working correctly. If I understand correctly, what you did by installing blanking off the discharge of the AA Booster Compressor is that you effectively prevented any air flow out of the compressor, and you still didn't see any pressure. Again, I think those roots-type blowers are as close to positive displacement compressors as you can get, so that shouldn't have happened.

(It's interesting to note, having done some research, that the company that manufactured the AA Booster Compressor is a part of Dresser, and GE now owns that part of Dresser. The assimilations just keep continuing.)

The results from your test of another unit's AA Booster Compressor and motor will be very interesting.

The unit will accelerate to purge speed during starting, then decelerate to just below minimum firing speed then accelerate slowly back up to minimum firing speed at which time firing should begin. That seems to have occurred during your starting attempt.

You should not change the firing speed to try to work around a problem with the AA Booster Compressor. And it would be interesting to know what the liquid fuel firing FSR value is on the other unit at the site.

I'm going to have to dig out an old Frame 9E CSP to see how to FQROUT and FQLM1 are related. There is another signal, FQR, I believe, that should be used somewhere in the sequencing. There needs to be a flow-rate reference that's compared to the flow-rate feedback from the liq fuel flow divider, and then the servo output current is adjusted to make the feedback equal to the reference.

You haven't said if the LFBV (Liquid Fuel Bypass Valve) has LVDTs or not. If so, did you record the position of the LFBV during the firing attempt, and if so, what was it?

I'm a little concerned about the value of servo current, because it's more negative than expected. If the actual liquid fuel flow-rate (the feedback) was equal to the flow-rate reference then the servo current should have been approximately -2.67% per processor, +/- 1.33%. But you reported -17.99%, and that's pretty excessive, and that could indicate a couple of things. That either the actual liq fuel flow-rate is much less than required, or the polarity of the currents being applied to the servo coils is not correct for all three coils, or there is some problem with the calibration of the LFBV LVDTs, or some scaling parameter is amiss.

You say your outage begins today. Will you be removing the fuel nozzles? If so, please report how the liquid fuel passages and tips of the nozzles appear on first inspection after removal. Was there any evidence of coking of the nozzle tips?

Also, will you be testing and/or replacing the liquid fuel- and purge air check valves? If they are tested, can you report the conditions of the check valves, specifically if they were passing in the opposite direction and what the cracking pressures of each was found to be?

Water in the lines is kind of problematic. There are supposed to be continuous blow-downs on each of the filters and in the low points of the AA and Purge Air piping systems. These continuous blow-downs are many times nothing more than small lengths of galvanized pipe with a very small diameter hole drilled in the side of the pipe, with a pipe cap on the end. When the unit is running air, and moisture, should blow out of these continuous blow-downs to prevent the collection of moisture in the lines and filters. They can be clogged over time with rust and dirt, and some people even believe (mistakenly) that the air coming out of the blow-down holes is a leak that must be plugged! Not so; that opening is there to prevent the collection of moisture and to provide a path for it to exit the piping. It must be kept open and clear at all times, and should not be plugged or otherwise shut off. It would be nice if someone would invent something like a steam trap that would open briefly when moisture was present, but no one has done so yet!

Water can cause problems for the AA Booster Compressor.... I believe that's one reason it's usually mounted above the Accessory Gear Box, to be in high location in the piping, similar to the Main AA Compressor which is normally a high point in the AA piping system.

So, please tell us if the LFBV has LVDTs, and if so, were you able to record the position of the LFBV during the firing attempt?

Also, was the FQLM1 value stable or was it oscillating during firing?

Lastly, you mention you have a second unit at site. Have you tried to start it on liquid fuel? If so, what were the results?

Thanks for the feedback; keep us updated on what you find and how this progresses, please.

 

CSA,

Thanks for the feedback. Today, we blanked off GT2 check valve (after the discharge of AA booster pump). Blanking off was done by inserting a customized rubber seal to cover the flange entirely. We turned on the AA booster; motor amp: 14amp & pressure: 0.1 bar, rotation CW. Subsequently, we changed the polarity (and thus the rotation of pump to CCW), the motor amp: 14amp & pressure -0.3 bar). Clearly, the GT2 AA booster motor was working harder than the GT1 AA booster.

Before the test was conducted, when the check valve was opened, water was flowing out of opening. A lot of water. Water was flowing out for 30 seconds continuously. Water wash wasn't done. We suspect that the water was due to condensation. Porous stone FA 4 plug was opened to drain the water as well the day before. The continuous blowoff drain is choked i believe. We can barely see the hole. Our AA temp has always been at 75 Celsius although the correct setpoint is 90-105 Celsius. Can condensation be the reason why there is water in the lube oil in the first place?

Anyway, back to the original topic (start up on diesel), what we are gonna do next is blank off GT 1 AA booster pump check valve again and tap in external air. We are gonna turn on the AA booster and tap in the external air while monitoring the pressure and amp reading. Our AA booster pump is rotary lobe. The pressure that we got was only 0.1 bar. Insufficient pressure compared to rated capacity of 0.7 bar.

The LFBV has LVDT. It was half closed when firing started. Our FSR during firing according to the sequence is 24 %. It was increased from 18.8% according to the manual due to recommendation from GE long long time ago.

Fuel nozzles wont be dismantled because this is an unplanned outage due to Tx problem. So far, we do not have any high temp spread problem, so little suspicion on the coking in the fuel nozzles. The VA 13 (fuel gas purge valve) was opened. The condition was really bad. The valve was badly corroded. This is the sign of water wash water damaging the valves. We have plans to replace valves with stainless steel ones.

<<<< (Also, will you be testing and/or replacing the liquid fuel- and purge air check valves? If they are tested, can you report the conditions of the check valves, specifically if they were passing in the opposite direction and what the cracking pressures of each was found to be?)>>>> How to we check the liquid fuel and purge air check valves? How do we determine the cracking pressure?

FQML1 value was oscillating but at around the same value. We will start GT2 on distillate after outage. As of now, my only concerns are: too much of fuel, insufficient AA pressure, and water in the AA line, purge air line dirty.

 

I keep going back to the fact that you say you can transfer from gas fuel to liquid fuel while running and everything works fine, though you haven't told us what the exhaust temperature spreads are when running on liquid fuel. The exhaust temperature spreads when running on gas fuel and liquid fuel will almost never be the same, as the fuel flows through different passages and orifices. So, if the liquid fuel passages are coked, then that can cause a different (higher) spread than on gas fuel.

Good information, but I question some of your conclusions.

>We turned on the AA booster;
>motor amp: 14amp & pressure: 0.1 bar,
>rotation CW. Subsequently, we changed
>the polarity (and thus the rotation of
>pump to CCW), the motor amp: 14amp &
>pressure -0.3 bar). Clearly, the GT2 AA
>booster motor was working harder than
>the GT1 AA booster.

Electrical current draw is a direct indication of load, and the current drawn by both motors in either direction of rotation is the same (unless there is a typo in the data above).

>Our AA temp has always been at
>75 Celsius although the correct setpoint
>is 90-105 Celsius. Can condensation be
>the reason why there is water in the
>lube oil in the first place?

The "AA temp" I believe you are referring to is the Main AA Compressor Inlet Temperature, which is the only AA temperature I know of that's monitored by Speedtronic panels. That setpoint should be approximately 120 deg C (250 deg F). There is a temperature switch that should alarm if the temperature is above 135 deg C (275 deg F) to protect the Main AA Compressor.

The Main AA Compressor inlet air is cooled by the AA Precooler, and there is a temperature regulating valve to control that temperature. Sorry; I can't recall the device number for that valve, but it's usually a Robertshaw temperature regulating valve, and it's adjustment is not intuitive, so you should find the manufacturer's instructions and read them to understand how to adjust it properly.

The Main AA Compressor inlet air temperature should be above the boiling point of water (to prevent any condensation!), and 75 deg C is not above boiling point. I don't know where you obtained the 90-105 deg C setpoint, but that's also marginally low. The setpoint for the AA Precooler temperature regulating valve is the setpoint for the Main AA Compressor inlet temperature; check the Device Summary for the proper value.

If the AA Precooler cools the air below the condensation temperature, then lots of water will collect in the piping, and that's not good for the Main AA Compressor, or the AA Booster Compressor. There is a drain on the shell of the AA Precooler. Since it's in a very low point in the piping I would suspect you will find lots of water in the shell if you open the precooler shell drains.

The water you are seeing may be coming from improper valve settings during Off-line Compressor Water Washing; we don't know. I wouldn't suspect that there could be such accumulations of water from an incorrectly set AA Precooler Temperature regulating valve. But, certainly, a low Main AA Compressor inlet temperature will cause condensation, and therefore rust.

And, when did we learn there was water in the lube oil? I don't know what that issue has to do with the issues we're discussing in this thread.

>The pressure that we got was only
>0.1 bar. Insufficient pressure compared
>to rated capacity of 0.7 bar.

Unfortunately, I can't comment on what pressure you might actually see from the AA Booster Compressor as the factory-installed gauge doesn't provide a very good indication.

<b>Please provide the rated current draw and HP from the AA Booster Compressor motor.</b> Presuming the voltage of the motor is 440 VAC, at 14 amps, the power consumed by the motor is approximately (440 * 14) * 3^0.5 = 14 HP, minus electrical losses.

I don't think you're making good conclusions on the working of the AA Booster Compressor, and I don't understand the "need" to insert a blind downstream of the compressor when testing. You may have more information (from the motor's nameplate) than you have shared here, but we don't have that information and can't reach the same conclusion as you.

Again, without the above information we can't know that the conclusion you have reached about the AA Booster Compressor is correct.

>The LFBV has LVDT. It was half closed
>when firing started. Our FSR during
>firing according to the sequence is 24
>%. It was increased from 18.8% according
>to the manual due to recommendation from
>GE long long time ago.

The LFBV should be closing during the start of firing, and it usually goes very nearly closed during firing and acceleration. This is because the High-pressure Liquid Fuel Pump is driven by the Accessory Gear, and it's driven by the turbine shaft, and the turbine is only at about 10% speed during firing, so the pump flow output is low requiring the LFBV to be very nearly closed to obtain flow. As turbine speed, and pump speed increases, the flow will increase, but the Speedtronic will adjust the LFBV to make the actual flow equal to the required flow; the components are sized for these conditions. (So, again, no need to change firing speed!)

As firing proceeds, the LFBV should close further from the 50% you noted. Can you record data using Short-Term Trending or one of the VIEW tools and upload that data to www.SpeedyShare.com or some other free hosting site and then provide the link to the data in this thread? We would need to see all the parameters previously requested, plus LFBV position (usual FSL).

One critical parameter we haven't discussed is the incoming liquid fuel pressure. It should be stable and approximately 4 barg (approximately 60 psig) during firing and operation. If the pressure is not stable, that can cause problems. Unstable fuel supply pressure can cause the flow-rate to oscillate during firing. If it's too low, that can cause problems achieving proper check valve cracking pressure.

Again, I keep going back to the fact that you can run on liquid fuel with "no problems", it's just starting on liquid fuel that's problematic.

>The VA 13 (fuel gas purge valve) was opened.
>The condition was really bad. The valve
>was badly corroded. This is the sign of
>water wash water damaging the valves. We
>have plans to replace valves with
>stainless steel ones.

I see this a lot; people think they don't have to manually close/open any valves when performing Off-Line Compressor Water Washes, and that's just not true. There are several critical manual valves which must be changed to protect water/detergent from getting into areas it's not supposed to get into.

>How to we check the
>liquid fuel and purge air check valves?
>How do we determine the cracking
>pressure?

Testing check valves involves applying pressure on one side of the valve and then on the other side of the valve. The cracking pressure (opening pressure) is listed in the Device Summary; for the Liq. Fuel Check Valves that pressure is usually approximately 6.9-8.6 barg (approximately 100-125 psig). This pressure is required for proper "mechanical" atomization of the fuel, particularly during firing. You will need to apply pressure to the check valve in the forward direction, slowly increasing the pressure. When the pressure exceeds the cracking pressure (which is sometimes stamped on the side of the valves) then there will be flow through the valve.

When you apply pressure to the check valve in the reverse direction, there should be NO flow. You should increase the pressure to approximately 15-10 barg to be certain that the check valves don't pass/flow in the reverse direction.

It would be very helpful to know what the servo current, FQLM1 value, FQR, and FQROUT values are just before the firing timer expires. The LFBV should have had a chance to close sufficiently to make the actual flow equal to the flow reference, so the numbers should be stable and tell us if the actual flow is being achieved or not.

It would also be helpful if you could look in the I/O Configurator, in the TCQA tab, for the LFBV (usually Servo-valve Output #3) and tell us what the Regulator Gain setting is, and what the Null Bias Current value is.

 

I have located a Mark V CSP for a Frame 9E; the Liquid Fuel FSR is FSR1, and it is multiplied by (TNH/100) to derive FQR, and FQROUT.

The liquid fuel flow divider feedback is FQL1, and it's multiplied by KFQLM1 to derive FQLM1.

I believe the servo regulator on the TCQA card uses the reference, FQROUT, and compares it to FQL1. FQLM1 is just the scaled flow divider feedback, scaled to a mass flow-rate (#/sec, or kg/sec).

So, the values we need to know are FAL, FQL1, FQR, FQROUT, FSL, TNH, FSR and FSR1. I would also like to know the value of K2F during the liquid fuel firing (it's a Control Constant, but some machines use one value for the gas fuel firing timer and another value for the liquid fuel firing timer). We need to know those throughout the firing sequence, from the start of firing through the end of firing. You can use VIEW1 to obtain this information on a once-per-second basis and then post the information to this thread.

There are explanations of how to use VIEW1 in the Mark V Maintenance Manual, GEH-5980. It's pretty straightforward (relatively simple), but it will require a second <I> if you have <I>s or you will need to open a command prompt window on an HMI if you have an HMI.

 

CSA,

I take my hat off to you for the relentless guidance. It is a quality that should be emulated by others. The least i could do is to ensure prompt reply for the knowledge of not only me but for other forumers.

Back to the topic, regarding the exhaust temp when running on gas and distillate, when we are running on gas the exhaust average temp is between 540 - 565 Celsius. When running on distillate, the exhaust temp is slightly higher between 564 - 572 Celsius.

The AA booster data is accurate. There isn't any typo. Both AA booster motors when running at both direction produced the same amp reading (GT1: 8-9 Amp, GT2: 14-15 amp). We are planning to blank off GT1 check valve at the discharge of the AA booster pump and tap in external air pressure at the inlet of the AA booster pump and regulate the pressure until we get a reading of 0.5-0.7 bar. Your feedback on this is appreciated.

Regarding the correct temperature setpoint of the AA, the correct setting is 107 Celsius at all time. Alarm will be annunciated when the AA temp reaches 135 Celsius. The day when the water was found leaking out of the FA 4, water wash wasn't performed days. Furthermore, the waterwash valve (normally open as can be seen in PID MLI 0425) was closed and the telltale drain after that valve was opened to ensure there wasn't any water passing from the waterwash valve. Our AA temp is at 75celcius, this is definitely an issue we need to discuss about.

The nameplate info of the motor is as follows: Model: 5ks2560PA1002A
HP: 20
RPM: 2955/2965
Volts: 380/415
Phase: 3
Hz: 50
Typ: KS

On the nameplate, there is also "HTR LDS 230V and 60W".

I will run VIEW1.exe the next time the test is conducted. Btw CSA, can temperature of the liquid fuel be a factor? If it is working when we C/over from gas to distillate at FSNL,can the liquid fuel temp be a factor of it being atomized? I look forward to your reply. Thanks again CSA.

>> Testing check valves involves applying pressure on one side of the valve and then on the other side of the valve. The cracking pressure (opening pressure) is listed in the Device Summary; for the Liq. Fuel Check Valves that pressure is usually approximately 6.9-8.6 barg (approximately 100-125 psig). This pressure is required for proper "mechanical" atomization of the fuel, particularly during firing. You will need to apply pressure to the check valve in the forward direction, slowly increasing the pressure. When the pressure exceeds the cracking pressure (which is sometimes stamped on the side of the valves) then there will be flow through the valve. <<

>> When you apply pressure to the check valve in the reverse direction, there should be NO flow. You should increase the pressure to approximately 15-10 barg to be certain that the check valves don't pass/flow in the reverse direction. <<

Regarding the testing of checkvalves, our instrument air max pressure is only 8bars. The pressure which you state is 15-10 barg. How else can we do it?

I have just checked the I/O configurator. The LFBV is servo output #4. Current bias is 2.67 and current gain is 3.5

 

A Main AA Compressor Inlet temperature setpoint of 107 deg C (225 deg F) is still above boiling, which would prevent condensation. I would suggest the water you are seeing in the piping is condensation, not necessarily from water washing. (The only reason water washing was suggested was because you brought it up; but it's still very common for valves not be properly set before and after off-line water washing.)

You didn't include the most important reading from the motor nameplate: the rated current! (The HTR LDS information is probably for the anti-condensation heater leads in the motor, which are to be energized when the motor is NOT running.)

Again, I keep going back to the fact that the unit transfers to liquid fuel from gas fuel while running, though we have no idea what the exhaust temperature spreads are on the two fuels.

Fuel temperature might make a difference if it's below 10 deg C or so. Most units located at sites with ambients cold enough to cause the liquid fuel to be difficult to atomize (especially during starting) usually had electric heaters to lower the liquid fuel viscosity by raising its temperature.

I think you are fighting several problems. Liquid fuel systems are very complicated, and many of the devices and parameters are neither controlled nor monitored by the Speedtronic. I'm speaking about things like liquid fuel supply pressure (upstream of the LFBV); individual liquid fuel supply pressures to the fuel nozzles; Main AA Compressor inlet temperature; AA flow-rate; etc. All of these things play a very important part in how well the liquid fuel system operates.

And, the Booster AA Compressor may also be part of the problem.

Troubleshooting is a logical process; working to find or eliminate each possible condition. Sometimes, it can take a long time because there is more than one problem.

You may also find that the amount of liquid fuel being admitted during firing (starting) is either slightly high or slightly low. I like to make sure everything else is as it should be before making recommendations to change the firing FSR.

As for testing devices with pressures above your instrument air system pressure, you would need to have a source of higher pressure that can be regulated as necessary. Most sites use something like a cylinder of compressed nitrogen and an accurate pressure regulator and gauge.

 

CSA,

The rated current is 27.4/25.7 amp. I will get back to you about the check valves cracking pressure test. Thanks CSA

 

Okay, so it looks like the Booster AA Compressor motor is only drawing about half of rated current so that tells us it's not working very hard which tells us the Booster AA Compressor isn't working very hard.

Now if that's because you have put a blind downstream of the compressor is not clear. A proper test would be to measure the current while the machine is firing, under normal conditions with some CPD pressure at the inlet to the compressor and no blind in the discharge of the compressor.

 

CSA,

We have already tried that as well on GT1. The amp measured was at 8amp only. I am looking forward to testing the unit on distillate again and take a complete data from VIEW1 from the start up until the failure to ignite alarm is annunciated. I will then post it here.

 

RY,

I didn't understand that from any of your previous posts, but it does add credence to the theory that there is a serious problem with the AA Booster Compressor. Again, the amount of current drawn by a motor is directly proportional to the amount of work done by the device the motor is supplying the torque to (after any mechanical losses, such as a gear box, etc.). If the current drawn by the motor is only 8A, and the nameplate is approximately 24-25A (slightly higher or lower, depending on voltage) then the AA Booster Compressor isn't working very hard.

I think if you look very closely at the P&ID where the AA Booster Compressor is shown (Atomizing Air?) I believe you should see that there is a continuous blow-down orifice somewhere very close to the blower cavity, and it's there to allow any condensate to be drained from the blower cavity. You have already noted a problem with the setpoint of the AA Temperature regulating valve and the presence of a lot of water in the AA piping, and I believe that water is not good for the AA Booster Compressor.

I still maintain that when a properly working roots-type blower is rotated in the reverse direction that it will draw excessive current, and the fact that the two blowers at your site don't is very unusual.

I would agree that it would be unusual for two AA Booster Compressors to both fail at roughly the same time with roughly the same behaviour, but stranger things have happened.

Since this turbine has a Mark V, the AA Booster Compressor may be more than 10 years old. It's not used very often, and so it might be worth removing it and sending it out for refurbishment.

But, having more data will be excellent. And, it may even allow us to narrow down the possibilities even further.

 

Here are the test results from the 2nd unit.
The data taken is as follows:<pre>
## Unit Pointname Scale
-- ------ ------------ -----
1 T2 TNH %
2 T2 TNH_RPM rpm
3 T2 L20FL1X LOGIC
4 T2 L33FL1C LOGIC
5 T2 L20CFX LOGIC
6 T2 FQLM1 kg/s
7 T2 FQROUT %
8 T2 FAL %
9 T2 FQL1 %
10 T2 FQR %
11 T2 FSL %
12 T2 FSR %
13 T2 FSR1 %
14 T2 K2F sec
15 T2 KFQLM1 kg/s%

12-MAY 01:53:53 10.00 300 0 1 0 0.00 -25.00 70.88 0.00 0.20 0.55 0.00 2.00 60.0 0.1184
12-MAY 01:53:54 10.00 300 0 1 0 0.00 -25.00 72.92 0.00 0.20 0.57 0.00 2.00 60.0 0.1184
12-MAY 01:53:55 10.00 300 1 0 0 0.00 3.00 72.92 0.00 0.00 0.56 0.00 0.00 60.0 0.1184
12-MAY 01:53:56 10.00 300 1 0 0 0.00 4.40 14.81 0.00 1.40 14.66 17.16 13.98 60.0 0.1184
12-MAY 01:53:57 10.00 300 1 0 1 0.00 5.35 14.81 0.00 2.35 46.87 23.47 23.47 60.0 0.1184
12-MAY 01:53:58 10.00 300 1 0 1 0.00 5.35 -9.57 0.00 2.35 79.52 23.47 23.47 60.0 0.1184
12-MAY 01:53:59 10.01 300 1 0 1 0.00 5.35 -6.23 0.00 2.35 99.90 23.46 23.47 60.0 0.1184
12-MAY 01:54:00 10.01 300 1 0 1 0.00 5.35 -6.23 0.00 2.35 99.95 23.47 23.46 60.0 0.1184
12-MAY 01:54:01 10.02 301 1 0 1 0.00 5.35 -4.01 0.00 2.35 99.83 23.46 23.47 60.0 0.1184
12-MAY 01:54:02 10.02 301 1 0 1 0.00 5.35 -4.01 0.00 2.35 99.86 23.47 23.46 60.0 0.1184
12-MAY 01:54:03 10.02 301 1 0 1 0.00 5.36 -5.34 0.00 2.36 99.87 23.47 23.47 60.0 0.1184
12-MAY 01:54:04 10.03 301 1 0 1 0.00 5.36 -5.34 0.00 2.36 99.87 23.47 23.47 60.0 0.1184
12-MAY 01:54:05 10.04 301 1 0 1 0.00 2.36 -4.77 52.72 2.36 0.64 23.47 23.47 60.0 0.1184</pre>
When we entered firing sequence, the alarm "start up fuel flow excessive trip: was annunciated 2 seconds after entering firing sequence and the unit coasted dwn and b4 going up again for another round. Same alarm was annunciated. There wasn't pressure reading in the flow divider. The flow divider did not rotate.

The unit 2 has not started up on diesel for quite some time. All this while we did not dare to run this unit on diesel bcoz the spread has been good.

Looking forward to seeing your feedback CSA. Thank you.

 

RY,

As has been said before, you are probably battling more than one issue here as liquid fuel systems are very complicated.

From the data provided, it would appear the LFBV is going fully closed in an effort to try to get the actual flow-rate to be equal to the flow-rate reference (presuming the LFBV is properly calibrated).

But for some reason, there was no flow. And you also indicated there was no pressure at the flow divider.

The 'Start-up Fuel Flow Excessive' alarm can be generated when there is air in the liquid fuel supply piping, causing there to be "spikes" in the flow-rate, sometimes very high spikes as liquid fuel and air flows through the flow divider. Unfortunately, one of the problems with VIEW1 is that it only gets data at a 1 Hz rate (once per second) and misses momentary spikes.

Another possibility might be that there was no pressure whatsoever at the inlet to the high pressure liquid fuel pump. You said there was no pressure at the liquid fuel flow divider (I presume someone was monitoring the gauge on the flow divider selector valve), and it should have registered some pressure--at least the pressure of the liquid fuel forwarding skid/pumps. With the LFBV closed, if there were any liquid fuel at the inlet of the high pressure liquid fuel pump it should be somewhat "magnified" by the pump, yet you indicate there was no pressure indication.

One of the selector valve positions is usually the pressure upstream of the high pressure liquid fuel pump, and another should be the pressure downstream of the pump. Pressure should have been seen at the selector valve position for pump inlet pressure. It's not clear if the selector valve was moved through the various positions during the firing or just left at one position, but some pressure should have been seen at any position if there was actually liquid fuel under pressure at the suction of the high pressure liquid fuel pump.

I would suggest transferring from gas fuel to liquid fuel just before shutting down, either at some low load or at FSNL. Then shut down while operating on liquid fuel. Then try starting on liquid fuel. This should ensure that all air is out of the liquid fuel lines and the lines are all full of fuel.

If the exhaust temperature spreads are higher on liquid fuel than on gas fuel, you should be able to use the flow divider selector valve to help determine where the problem might be. A combustor/fuel nozzle with a pressure that is higher than the average of the others might indicate a plugged nozzle. A combustor/fuel nozzle with a pressure lower than the average of the others might indicate a leaking purge air check valve (and that could be verified by liquid fuel flowing out of the Tell-tale Leakoff.) If there is no fuel flowing out of the Tell-tale Leakoff, then the problem might be a liquid fuel check valve that is opening at a pressure below the design cracking pressure.

But, the data you provided is kind of a step backwards from the original post. It appears that there was no liquid fuel pressure/flow from the data.

 

CSA,

These are the values for GT1<pre>
## Unit Pointname Scale
-- ------ ------------ -----
1 T1 TNH %
2 T1 TNH_RPM rpm
3 T1 L20FL1X LOGIC
4 T1 L33FL1C LOGIC
5 T1 L20CFX LOGIC
6 T1 FQLM1 kg/s
7 T1 FQROUT %
8 T1 FAL %
9 T1 FQL1 %
10 T1 FQR %
11 T1 FSL %
12 T1 FSR1 %
13 T1 K2F sec
14 T1 KFQLM1 kg/s%

23-MAY 23:43:38 10.00 300 1 0 1 0.00 3.00 16.33 0.00 0.00 0.68 0.00 60.0 0.1184
23-MAY 23:43:39 10.02 300 1 0 1 0.00 5.39 16.33 0.00 2.39 14.61 23.82 60.0 0.1184
23-MAY 23:43:40 10.04 301 1 0 1 0.00 5.39 -9.17 0.00 2.39 31.38 23.82 60.0 0.1184
23-MAY 23:43:41 10.05 301 1 0 1 0.00 5.39 -9.17 0.00 2.39 48.10 23.82 60.0 0.1184
23-MAY 23:43:42 10.07 302 1 0 1 0.00 5.40 -9.80 0.00 2.40 64.55 23.82 60.0 0.1184
23-MAY 23:43:43 10.09 302 1 0 1 0.00 2.40 -9.80 3.43 2.40 75.30 23.81 60.0 0.1184
23-MAY 23:43:44 10.11 303 1 0 1 0.31 2.41 -20.06 3.31 2.41 74.20 23.81 60.0 0.1184
23-MAY 23:43:45 10.11 303 1 0 1 0.31 2.41 -20.06 2.37 2.41 73.08 23.81 60.0 0.1184
23-MAY 23:43:46 10.12 304 1 0 1 0.28 2.41 -15.50 3.79 2.41 73.26 23.81 60.0 0.1184
23-MAY 23:43:47 10.13 304 1 0 1 0.46 2.41 -18.82 1.27 2.41 73.04 23.81 60.0 0.1184
23-MAY 23:43:48 10.15 304 1 0 1 0.46 2.41 -18.82 3.02 2.41 74.84 23.82 60.0 0.1184
23-MAY 23:43:49 10.17 305 1 0 1 0.32 2.42 -17.55 3.86 2.42 73.68 23.82 60.0 0.1184
23-MAY 23:43:50 10.18 305 1 0 1 0.32 2.43 -17.55 2.57 2.43 72.75 23.82 60.0 0.1184
23-MAY 23:43:51 10.19 306 1 0 1 0.38 2.43 -17.05 3.35 2.43 72.14 23.82 60.0 0.1184
23-MAY 23:43:52 10.21 306 1 0 1 0.38 2.43 -17.05 3.46 2.43 72.95 23.81 60.0 0.1184
23-MAY 23:43:53 10.22 307 1 0 1 0.42 2.43 -18.70 3.04 2.43 72.18 23.81 60.0 0.1184
23-MAY 23:43:54 10.22 307 1 0 1 0.37 2.44 -18.69 3.70 2.44 72.46 23.82 60.0 0.1184
23-MAY 23:43:55 10.24 307 1 0 1 0.37 2.44 -18.69 2.86 2.44 71.62 23.82 60.0 0.1184
23-MAY 23:43:56 10.25 308 1 0 1 0.25 2.44 -18.89 2.59 2.44 72.94 23.81 60.0 0.1184
23-MAY 23:43:57 10.27 308 1 0 1 0.25 2.45 -18.89 3.82 2.45 73.20 23.82 60.0 0.1184
23-MAY 23:43:58 10.29 309 1 0 1 0.47 2.45 -20.16 0.00 2.45 72.27 23.81 60.0 0.1184
23-MAY 23:43:59 10.30 309 1 0 1 0.47 2.45 -20.16 2.46 2.45 72.20 23.81 60.0 0.1184
23-MAY 23:44:00 10.31 309 1 0 1 0.45 2.45 -21.95 3.70 2.45 71.82 23.81 60.0 0.1184
23-MAY 23:44:01 10.32 309 1 0 1 0.45 2.45 -21.95 1.36 2.45 71.91 23.81 60.0 0.1184
23-MAY 23:44:02 10.34 310 1 0 1 0.16 2.46 -19.43 2.29 2.46 71.87 23.81 60.0 0.1184
23-MAY 23:44:03 10.34 310 1 0 1 0.36 5.46 -14.59 3.57 2.46 71.52 23.81 60.0 0.1184
23-MAY 23:44:04 10.36 310 1 0 1 0.36 5.46 -14.59 1.44 2.46 71.46 23.81 60.0 0.1184
23-MAY 23:44:05 10.37 311 1 0 1 0.44 2.46 -19.28 2.91 2.46 72.04 23.81 60.0 0.1184
23-MAY 23:44:06 10.38 311 1 0 1 0.44 2.47 -19.28 3.70 2.47 71.62 23.81 60.0 0.1184
23-MAY 23:44:07 10.39 312 1 0 1 0.46 2.48 -17.30 3.17 2.48 71.52 23.82 60.0 0.1184
23-MAY 23:44:08 10.40 312 1 0 1 0.46 2.48 -17.30 2.21 2.48 70.70 23.82 60.0 0.1184
23-MAY 23:44:09 10.42 312 1 0 1 0.38 2.48 -18.17 3.37 2.48 71.15 23.81 60.0 0.1184
23-MAY 23:44:10 10.43 312 1 0 1 0.38 2.48 -18.17 3.37 2.48 71.08 23.81 60.0 0.1184
23-MAY 23:44:11 10.45 313 1 0 1 0.28 2.48 -19.28 2.56 2.48 70.48 23.81 60.0 0.1184
23-MAY 23:44:12 10.45 314 1 0 1 0.36 2.49 -21.94 3.14 2.49 71.63 23.81 60.0 0.1184
23-MAY 23:44:13 10.46 314 1 0 1 0.36 2.49 -21.94 3.73 2.49 71.05 23.81 60.0 0.1184
23-MAY 23:44:14 10.46 314 1 0 1 0.35 2.49 -15.72 2.91 2.49 71.66 23.81 60.0 0.1184
23-MAY 23:44:15 10.48 314 1 0 1 0.35 2.50 -15.72 2.62 2.50 71.73 23.81 60.0 0.1184
23-MAY 23:44:16 10.49 315 1 0 1 0.46 2.50 -15.83 3.80 2.50 70.78 23.82 60.0 0.1184
23-MAY 23:44:17 10.50 315 1 0 1 0.46 2.50 -15.83 3.04 2.50 69.94 23.82 60.0 0.1184
23-MAY 23:44:18 10.51 315 1 0 1 0.36 2.50 -20.52 2.25 2.50 70.60 23.82 60.0 0.1184
23-MAY 23:44:19 10.52 316 1 0 1 0.04 2.50 -18.61 3.23 2.50 70.16 23.81 60.0 0.1184
23-MAY 23:44:20 10.54 316 1 0 1 0.04 2.51 -18.61 3.70 2.51 70.48 23.82 60.0 0.1184
23-MAY 23:44:21 10.54 317 1 0 1 0.21 2.52 -14.72 2.47 2.52 69.50 23.82 60.0 0.1184
23-MAY 23:44:22 10.55 317 1 0 1 0.21 2.52 -14.72 3.06 2.52 70.64 23.82 60.0 0.1184
23-MAY 23:44:23 10.56 317 1 0 1 0.30 2.52 -19.45 3.73 2.52 70.48 23.81 60.0 0.1184
23-MAY 23:44:24 10.57 317 1 0 1 0.30 2.52 -19.45 3.37 2.52 69.85 23.82 60.0 0.1184
23-MAY 23:44:25 10.59 317 1 0 1 0.40 2.52 -19.28 2.56 2.52 69.16 23.81 60.0 0.1184
23-MAY 23:44:26 10.60 317 1 0 1 0.40 5.52 -19.28 2.64 2.52 70.07 23.81 60.0 0.1184
23-MAY 23:44:27 10.61 318 1 0 1 0.46 2.52 -16.80 3.69 2.52 69.72 23.82 60.0 0.1184
23-MAY 23:44:28 10.63 319 1 0 1 0.44 2.53 -16.28 3.38 2.53 70.07 23.82 60.0 0.1184
23-MAY 23:44:29 10.63 319 1 0 1 0.44 2.53 -16.28 2.67 2.53 68.98 23.82 60.0 0.1184
23-MAY 23:44:30 10.63 319 1 0 1 0.44 2.54 -22.18 2.77 2.54 70.01 23.81 60.0 0.1184
23-MAY 23:44:31 10.64 319 1 0 1 0.44 2.54 -22.18 3.84 2.54 69.98 23.81 60.0 0.1184
23-MAY 23:44:32 10.64 320 1 0 1 0.40 2.54 -18.06 3.25 2.54 69.77 23.82 60.0 0.1184
23-MAY 23:44:33 10.66 320 1 0 1 0.40 2.54 -18.06 2.60 2.54 68.97 23.82 60.0 0.1184
23-MAY 23:44:34 10.67 320 1 0 1 0.31 2.54 -18.72 2.81 2.54 70.93 23.82 60.0 0.1184
23-MAY 23:44:35 10.68 320 1 0 1 0.38 2.55 -17.50 3.83 2.55 69.68 23.82 60.0 0.1184
23-MAY 23:44:36 10.70 321 1 0 1 0.38 2.55 -17.50 3.52 2.55 69.19 23.82 60.0 0.1184
23-MAY 23:44:37 10.71 321 0 1 0 0.00 -25.00 3.11 0.00 1.70 25.91 15.96 60.0 0.1184
23-MAY 23:44:38 10.72 321 0 1 0 0.00 -25.00 3.11 0.00 0.52 0.70 4.80 60.0 0.1184
23-MAY 23:44:39 10.72 322 0 1 0 0.00 -25.00 75.00 0.00 0.16 0.76 1.44 60.0 0.1184
23-MAY 23:44:40 10.73 322 0 1 0 0.00 -25.00 75.00 0.00 0.16 0.73 1.44 60.0 0.1184
23-MAY 23:44:41 10.74 322 0 1 0 0.00 -25.00 74.20 0.00 0.05 0.73 0.43 60.0 0.1184
23-MAY 23:44:42 10.75 322 0 1 0 0.00 -25.00 74.20 0.00 0.02 0.71 0.13 60.0 0.1184</pre>
Appreciate your feedback. Thank you.

 

RY,

This thread was started because of a possible problem with Atomizing Air pressure/flow from the Booster Atomizing Air Compressor during starting on liquid fuel.

The data for this unit seems to indicate that the LFBV was controlling liquid fuel flow-rate during firing properly without closing fully (presuming the LVDTs were properly calibrated). I'm not thrilled with the servo current value, but the actual flow vis a vis the required flow seems to be good, so the servo current is kind of irrelevant at this point.

Since the unit(s) can transfer fuel at rated speed and run on liquid fuel without any problems, and since the problem we are trying to resolve is an inability to start on liquid fuel, we need to concentrate on what is required to establish flame when starting on liquid fuel. That is fuel, spark, combustion air, and atomizing air.

The spark seems to be present. Combustion air is there (from the axial compressor discharge with the IGVs at the closed position).

The actual fuel flow-rate seems to be matching the required fuel flow-rate. It may just be that the liquid fuel flow-rate needs to be increased (by increasing firing FSR)--but that's not clear.

Sufficient atomizing air pressure/flow is still a question--at least in my mind. The results of the testing, based on my experience, indicate something is not right with the blower/compressor. But, I'm not on site and can't see exactly what's happening or how the testing is being done or the condition of the equipment.

You can try increasing the liquid fuel firing FSR by a half percent (0.5%), try starting again. If that doesn't work, try increasing it another 0.5% and try again. I wouldn't go much higher than that.

If that doesn't work, it's likely the Booster Atomizing Air compressor. You have indicated the presence of large amounts of water in the Atomizing Air Piping as well as plugged continuous blow-down orifices. I don't believe that the Booster Atomizing Air Compressor tolerates water very well. The current drawn by the motor driving the Booster Atomizing Air Compressor seems to be much less than the motor's nameplate rating. Either the motor wasn't sized properly (it's oversized for the application--not likely as an oversized motor costs more than a properly-sized motor!) or the torque required by the motor isn't near rated because the blower/compressor isn't working as hard as it should be.

The fact that you seem to have the same testing results for two Booster Atomizing Air Compressors on two units is confounding, but stranger things have happened.

I'm out of ideas at this point. Again, the things necessary for firing on liquid fuel are:

Fuel
Combustion Air
Spark
Atomizing Air

You haven't tried increasing the fuel flow-rate, and in my opinion the atomizing air is suspect.

Just be careful with increasing the firing FSR value. You should ensure proper purging between liquid fuel firing attempts, maybe even let the unit CRANK for a couple of minutes or more between attempts. Pooled liquid fuel in the combustion wrapper or combustors or the turbine or exhaust can be very explosive if flame is established after several failed attempts. The False Start Drain Valves are installed just for this purpose, but they are usually not very well maintained or monitored during liquid fuel firing.

Just be careful is all I'm trying to say.

Please write back to let us know how this progresses.