Can anyone in the Instrumentation and Control department help me with this issue. The Stop/Ratio Valve (SRV) of the GE NUOVO PIGNONE 5001 gas turbine opens on firing but immediately after firing, it closes and causes the flame to go out. this continues until the unit trips. Please can someone with experience about this problem help me out with the actual issue and how to solve it.

 

<b>Not enough information.</b>

What alarms are being annunciated when this is occurring--specifically, what alarms are present and active during purging, and what alarms are annunciated during firing, in order of annunciation?

Is anyone watching the Trip Oil pressure? Are you able to trend the Trip Oil pressure switch inputs?

What is happening to the Hydraulic Pressure during purging and firing?

You want answers. We need much more information than you have provided.

 

Hai Teekay,

Is there any alarm like "STARTUP FUEL FLOW EXCESSIVE TRIP"?

If YES, then servo valve not maintaining the p2 pressure. It could be a servo valve problem.

Have you confirmed SRV operation by stroke checking?

Is there any hunting on trip oil or hydraulic oil systems?

Have you performed any maintenance before the problem started?

take care
G.Rajesh

 

Good day Rajesh,

Thanks for your response.

We recorded START UP FLOW EXCESSIVE TRIP once during one trial, but it didn't come again. but that was before we calibrated the new SRV & GCV servo valves that we changed. Afterwards, we were still having irregular hunting of the SRV which was making the flame go out & having FAILURE TO IGNITE or LOSS OF FLAME after some time of igniting and going off due to the SRV hunting. Eventually the SRV, GCV & IGV stopped moving on calibrator mode stroke tests. The trip oil & hydraulic oil pressures are ok during the operations. When we checked the small hydraulic filter before the GCV & SRV servo valves, it was clogged. So we cleaned & changed it then the SRV & GCV moved again on calibrator mode for a while & later stopped again.

A point i want you to also know and confirm for me is that we were previously using MOOG servo valve G771K200, but the new ones we installed were MOOG servo valve G771K201A. So can you please confirm for me by experience if the two models are interchangeable without issues. This particular turbine unit had been on forced outage for about 3 months before this recent action. Thanks for your help. You can send me an email if necessary and I will also like to have yours so I can send you more details if you like.

 

Check out the Moog website yourself, there is lots of information there. G771K200/G771K201A, from Moog website G771Kxxxx where xxx is described as (Assigned at the factory) and last x is described as (Factory Identification (Revision Level)), so I don't think this is a problem. What is important is the Type No. which should be on the nameplate as well. It will look something like H10FOFM4V14, this is the No. that really specifies the servo functionality (Flow Rate,Operating pressure,etc) The new valve you put on must have the same Type No. as the one that came off, or you could have problems.

Hope this helps.

 

Hai Teekay,

>>When we checked the small hydraulic filter before the GCV & SRV servo valves, it was clogged. So we cleaned & changed

Did you clean and fixed back or totally replaced the hydraulic line filter? (better to replace with new, cleaning may not give sufficient result)

>>it then the SRV & GCV moved again on calibrator mode for a while & later stopped again.

So do you mean to say, during stroke checking both SRV & GCV not responding now?

How is the performance about GCV?

Did you replace both servo valves? We hope you may not miss the correct servo polarity.

We are using MOOG servo valve Model:G771K200 and type: HI9FOFA4VI4 for both SRV & GCV. As Mr.Johnston said, please look the servo type.

There will be some very remote chances for hydraulic cylinders leaks.

You can contact me at [email protected].

3 months forced outage is long time, let us hope Mr.CSA can share his valuable experiance to solve your issues.

Take care
G.Rajesh

 

G.Rajesh,

The originator of this thread does not seem to want to provide the information requested of him. For the SRV to close suddenly after firing there is likely an alarm, or more likely, several alarms which are present during purging and firing and at least one which occurs after flame detection and causes 20FG-1 to be de-energized because the Speedtronic believes the turbine should be tripped.

This is typical of many sites--they just ignore alarms because they don't understand how what the alarm text message means intuitively (so it must not be important) and there are probably so many alarms present and annunciated during normal operation that they have become de-sensitized to all alarms. They also don't know how to 'work backwards' from an alarm message to understand what caused the alarm to be annunciated. They just want someone to tell them all the possible reasons which might cause the SRV to close some time after flame is actually detected (if we read the posts correctly) and then hopefully one of the possibilities will look like something they might be experiencing and then they will be heroes for solving their problem.

The originator did not provide enough information in his original post, nor has he provided any requested information. He's not really interested in learning anything, just in being told how to solve the problem.

Either there is low Trip Oil pressure to begin with (and many Speedtronic panels will actually allow purging and firing with low Trip Oil pressure and will then trip the unit on low Trip Oil pressure when the firing timer expires--YES!!!), or there is some other trip condition being detected by the Speedtronic control system and an alarm (more likely multiple alarms!) is being annunciated to indicate what the problem is, but, then again, if there are always lots of alarms and one doesn't know how to sort through them and understand the sequence of events and one is de-sensitized by the continual annunciation of alarms (Process and Diagnostic), and one doesn't know how to troubleshoot alarms then one is just looking for someone to intuit what is happening at the site and guess at problems.

We already know the site replaced two servo-valves (and we already know that replacing servos is a very common--and mystically incorrect--troubleshooting procedure) and we have no idea about LVDT calibration or servo-polarity checks. We have no idea what the regulator gains are, or what the null bias current values are.

There's just too much we don't know, and will likely never know.

Leave this one be until such time as we get more detailed information--as has already been requested above.

 

G.Rajesh,

One more thing; when you take a thread off-line we all lose the benefit of learning from the process and the outcome.

And that's one of the best things about control.com, when people provide feedback! And, when we learn from the process during troubleshooting.

 

Are you sure that oil is have good quality ?

 

Sorry all, I've been offnet & offline for a while due to location constraints hence the delayed responses.

Good day CSA,
I appreciate your response and candor, however I sense an undertone of some sort of personality attack on me. I will take it that you misunderstood me. I will like to first of all say that I'm very new to this forum & that was in fact my very first post, so I haven't fully understood how the issues are treated here. I'm also just starting out in this profession and as such have very little experience so when I encountered this challenge which I have been working with some other colleagues to solve, I was told about this forum and I quickly came to seek help from experienced professionals such as you and others. So Yes I was looking to hints or information that will help me identify the real problem and get the solution either directly or indirectly and not necessary trying to be a hero as you said. That's bashful and misrepresenting my intentions.
I will however still appreciate all the help I can get from you and others who have more professional experience and as such a better understanding of these issues.

Concerning the matter at hand, I didn't take a record of all the alarms that came up as we had attempted several startups without success and had to acknowledge and clear some of the alarms to get a ready to start on the next attempt, however I can remember some that I took note of.

1. On the first attempt, we had a START UP FUEL FLOW EXCESSIVE TRIP, but after calibrating the servovalves, we didnt have that again.

2. During firing, one of the alarms we usually overlooked was LVDT FEEDBACK TROUBLE

3. We usually get a few DIAGNOSTICS one I took note of is DIAGNOSTIC ALARM <C><Q> and some VOTER MISMATCH on several inputs which I overlook cos I don't know what to make of them yet and I don't know their effects and I assume they are harmless. I didn't take specific note of them so I cant quote them exactly now.

4. We had several GAS RATIO VALVE NOT FOLLOWING ALARM during firing also and most times the unit tripped after this alarm.

5. On some attempts, it was FAILURE TO IGNITE or LOSS OF FLAME.

6. We also recorded GAS CONTROL VALVE NOT FOLLOWING

7. On one attempt after changing the servovalves again to older models which we took from another unit which was on forced outage, the unit fired not smoothly though and was proceeding to FSNL, it suddenly tripped at about 80% TNH on HYDRAULIC TRIP OIL PRESSURE LOW but we still had GAS CONTROL VALVE NOT FOLLOWING.

8. The last attempt we to start up, we got to FSNL but the GCV and SRV were not steady, there were fluctuations and the Frequency was hunting however it stayed like that for about 5 hours and later tripped on CONTROL MODULE OVERSPEED TRIP.

9. Other alarms seen which I dont understand are: IOMA POWER SUPPLY OUT OF LIMITS P15 also TCQA 4-20mA INPUT mA2 OUT OF LIMITS

Finally your comment as quoted below is true:

> We already know the site replaced two servo-valves (and we already know that replacing servos is a very common--and
> mystically incorrect--troubleshooting procedure) and we have no idea about LVDT calibration or servo-polarity
> checks. We have no idea what the regulator gains are, or what the null bias current values are.

So i will appreciate further enlightenment on these as I'm interested in learning as well as getting the problem solved.

Thank you for your time and I will appreciate your continuous professional support. I will ensure to keep every discussion on the forum for all to benefit from as you rightly pointed out. I look forward to your response. Cheers

 

We thought so, but we took oil sample for a test and the result said the oil is good.

 

Teekay,

I will take it that you took my comments personally, which was not intended. But, no matter what you call it, overlooking alarms is the same thing as ignoring them. Just resetting and continuing to push the START button without understanding every alarm is just wishing for good luck. Sometimes it works; sometimes it doesn't. Albert Einstein said it best, "Doing the same thing over and over again and expecting different results is the definition of insanity."

I'll bet the alarm printer for the operator interface isn't even connected or working, so you can't retrieve the alarm printouts to get a chronological list of alarms. (If you can, then you could post the chronological list and someone here might be able to help.)

The Mark V has a Trip History Display which can be printed after any turbine trip which can be used to help understand what might have caused the trip, as well to collect a list of alarms before and at the time of the trip. It's very unfortunate that GE doesn't block trip alarm messages which occur after the turbine has already tripped. And, alarm text messages don't always tell the operator in very specific terms that it results in a turbine trip. A technician or operator needs to learn how to troubleshoot alarms, and make their own list of alarm messages which can result in a turbine trip. (Any time L4T transitions from logic "0" to logic "1" the turbine will be tripped. So, if you want to know most of the conditions (with the exception of hardware trips) which result in a trip, work backwards in the sequencing from L4T.)

You haven't confirmed the Model numbers of the servo's you installed matched the Model numbers of the servo's they replaced. If the Model numbers aren't the same, you may have exacerbated a pre-existing condition by using incompatible servo-valves.

We don't know what kind of Speedtronic panel you have so we can't tell you how to find the servo regulator gain and -null bias values.

If you can use the manual positioning feature of AutoCalibrate to move the SRV and GCV to several positions and they are stable at those positions (say, 15$, 25%, 40%, 55%, 70%, 85%, and 95%, for example) then the problem is likely not with the servo or the servo regulator settings (gain and null bias). If the valve isn't stable when being manually controlled, then the number of possible problems is many, including servo regulator gain and/or null bias settings, leaky actuators, etc.

If the positions under manual control are stable, then you need to investigate the inputs to the automatic references. The SRV is positioned to control the pressure between the SRV and GCV (called P2 pressure) as a function of turbine speed. If the speed pick-up feedback is unstable or the P2 pressure feedback is unstable then that could be causing lots of problems.

If the gas fuel supply pressure is restricted (choked), say by dirty filters or clogged strainers, then the SRV may have problems maintaining P2 pressure. If the gas fuel supply pressure is not stable then the SRV will fluctuate to try to maintain stable P2 pressure.

If the P2 pressure feedback is unstable or if there are multiple P2 pressure transducers and the three feedbacks are not closely matched then that could cause SRV instability, as well.

During acceleration after firing and warm-up, if the rate of acceleration is too fast then the GCV will close to try to reduce the rate of acceleration. If the turbine speed feedback is unstable during acceleration then the GCV will likely be opening and closing which will also have an effect on the P2 pressure, which will cause the SRV to close and open to try to maintain P2 pressure equal to the P2 pressure reference.

If the unit reaches rated speed and can be synchronized and is being operated in Droop speed control mode in parallel with other generators and the speed of the turbine and generator is not stable (which would mean the generator frequency was also unstable) then the grid with which the unit is synchronized is "soft" and/or the grid frequency is unstable, or there is some problem with the turbine speed feedback.

Most of the above conditions will have some alarm, either Process or Diagnostic, associated with it. But, if they are being overlooked (ignored)....

As for the IOMA Power Supply Out of Limits alarm, that is not unusual for certain PROM revisions, but could also be problematic. IOMA refers to the "I/O Master" and if the power supply to the I/O Master is not within bounds then there could be problems with the I/O being controlled by the I/O Master. (You would have to use the I/O Report and the I/O Configurator to know what I/O is being controlled by the IOMA).

Since you are just starting out, I would caution you to not overlook (ignore) alarms just because your more experienced colleagues seem to do so. It's a very bad practice. Learn to use the Alarm List and Dynamic Rung Display and the CSP Printout to troubleshoot alarms. I believe a former contributor, markvguy, wrote some replies to threads to teach people how to troubleshoot alarms on Mark V turbine control systems. control.com has a very good 'Search' feature at the top right corner of every page.

Best of luck with your challenge in your new profession. When you write to a forum like this for help, be prepared to provide detailed information. If you could solve the problem without help from others, then you wouldn't need to provide the detailed information others might require to help you. If you feel the information requested is irrelevant, then you should state why. If you have done any troubleshooting you should state the results of the troubleshooting in your original posting.

 

Dear Teekay,

this is an interesting thread you have. I will attempt to help as others have here have, but understand as others have noted we need your help to supply as much information about your site, so the comments we offer are useful. My assumptions are this is a Frame 5 gas turbine, using the combination SRV/GCV valve assembly. You run at least gas fuel, not sure about liquid. You did not specify the control system for the turbine but from your post I am thinking its a MKV.

1) You need to understand that ALL alarms are important. That means diagnostic and regular alarms. You need to have a printout so that you can identify the timing of each alarm. You need to use the tools to identify the diagnostic alarms and understand them and their cause.

1) You mention that there are lots of alarms. Each alarm needs to be investigated and understood as to why it is occuring. Understand that alarms are not normal, none, any, nada. Alarms indicate a problem, failure, or abnormailty that needs to be addressed or repaired. I do understand that there may be times an alarm is present, like loss of a thermocouple for instance. But you must be sure that if an alarm is present it will not impact the turbine operation. If alarms are ignored then things don't work, and all of our suggestions here are useless. Like CSA, I have heard from sites that "oh yeah we always get that alarm". Please understand that is not acceptable,(in my mind anyway).

2) From reading all your posts the feeling I get is your gas valves are not moving properly. In one post you say you found the pencil filters plugged for the servo valves. Simply put that is very bad, that means your oil was and possibly still is contaminated. Without clean oil at the proper temperature you cannot expect the gas valve hydraulic system to work properly. In another post you say that you changed servos from another unit, but that the servo part numbers did not match. This will also cause problems. As Bob noted each servo is slightly different, they are designed for different pressures and flows, and you cannot and should not expect them to operate the same.
Again from the alarms you provided they all lead me to believe that your gas valves, both SRV and GCV are not following the reference provided to them. If that is the case then you will see instability in your P2 pressure and uncontrollable flow through the GCV. This will lead to failure to ignite, loss of flame trips, unstable speed or unit output, and lastly even overspeed events.

Now I understand as others do that things are difficult in parts of the world. You can't just run down to the corner store and get new servo's and a refill of your lube oil tank. But without the right parts and clean oil you won't be able to make things work or operate correctly.

My suggestions.

1) Have your oil tested by a laboratory for varnish potential, additive degradation etc. If it is not in good condition you need to understand why, replace the oil and prevent a reoccurance.

2) Either have your servos tested, or replace with new of the proper part number.

3) Test your hydraulic system. Verify proper operation of all pumps, pressure regulators, switches and trip valves.

4) Verify correct wiring and polarity of all servo valve and LVDT wiring.

5) Run a calibration check of the SRV and GCV. Verify proper feedback from each LVDT. Verify the travel of each valve is smooth and matches the reference. Verify that each valve travels the correct amount for what is written in the control specification.

6) Verify that your incoming gas pressure is at specification and is steady.

After all this I might be ready to attempt to restart the machine. Please continue to write back and let us know how you are coming along.

 

Well said, MIKEVI!!!

It does seem to be the case at too many sites that alarms are just interpreted as nuisances and not viewed as the helpful indications they are or should be.

It is also unfortunate that GE does not do a better job of documenting alarms, or providing written documentation in the manuals provided with the control systems for how to understand and troubleshoot alarms. If they did that, then most likely people wouldn't feel so intimidated by alarms.

It's also unfortunate that if a unit trips on, say, generator differential, that any subsequent trip conditions which occur but are not the original cause of the trip (such as the generator differential in this example) aren't blocked, such as the ever-misleading and -present 'Hydraulic Protective Trouble' or 'GCV Fault.'

Many sites don't understand that the last alarm reported (at the top of the list or printout) is the most recent alarm to have occurred, and that the last alarm reported may not be (usually isn't, in fact) the original cause of the trip. They just look at the list of alarms and assume the last alarm reported (at the top) is the original cause of the trip.

And lastly, it's also unfortunate that many people working on GE-design heavy duty gas turbines (from OEM TAs to site technicians and operators) don't have the proper training to understand how to troubleshoot alarms and read alarm printouts or alarm displays. And for this I blame supervision and management for not ensuring their personnel have proper training to support reliable unit commissioning and/or operation.

It's really a credit to GE that their turbines (most of them, anyway) are so robust that they can take the abuse they are subjected to when people just keep pushing the START button time and time again hoping for a successful start, synch, and load. And, it's also a credit that the control systems do a reasonably good job of protecting the turbines in spite of poor operating practices.

I understand that many site supervisors are reluctant to provide training to their personnel because those personnel will very quickly market that knowledge to another site for more money and then the investment has vanished. That also says something about site supervision and management, as well. First that they don't treat their personnel well enough for them to consider staying on after receiving valuable training. Second that they will poach personnel from other sites who have invested in training.

But, I digress--while waiting for mechanics to finish a job they were supposed to finish eight days ago. So my work will be squeezed because there is now a mad rush to get the unit returned to service because there wasn't sufficient time put into the schedule to account for unfamiliar mechanics brought onto a site for the first time (at a discount to more experienced mechanics) without proper supervision and direction. But, it's a weekend and the pay is good. Ahh, the life of a controls TA!

 

Thank CSA for your response. I appreciate your support. I agree I took it quite personally but not anymore and yes I am guilty for ignoring/overlooking some alarms mostly because I came into the system to meet the units running with so much alarms coming up and being ignored because they don't cause any noticeable harm and some of them I don't know what they mean or what their impact may be or why they come about (especially the diagnostics, IO Limits, voter mismatch and they likes) but that will be a learning point for me and I will do well to keep records of and investigate every alarm that comes up subsequently.( You were right about the printer not being connected also and I'll look into that as well).

Concerning the servo valve issue at hand, sadly I don't have a chronological record of the alarms, so i cant provide that now. However I confirmed that the servo valve models were different, taking a cue from Johnston and G.Rajesh's posts, i checked and found that the former servos were G771k200 TYPE: S19P0FA4V14 and the new ones were G771K201A TYPE: H04FOFA4V14.

Eventually we changed the new servo valves to the older model gotten from another turbine unit in our station which is on forced outage due to ratchet system trouble. The pencil filters were also replaced with new ones and the main hydraulic filters also. Lube Oil tests were taken, but results said the oil is still ok. After this change, the unit was started successfully to FSNL however fluctuations of the SRV and GCV were still there causing the frequency to hunt on the FSNL state as well. The unit is currently on load in this condition and since its on droop system, the other units synchronized to it are also experiencing these fluctuations. However I just noticed that as the unit got to a load value of 13MW (its a 25MW capacity turbine), the SRV, GCV and frequency hunting stabilized. It was reported that the PCV (Pressure Control Valve) at the gas skid no longer works, so gas supplied is now being regulated only by manual valves on the skid which does not offer any dynamic control. I'm beginning to think that might have been part of the problem. What do you think?

You mentioned the servo regulator gain and null bias values..which I don't know how to check and how to adjust, but we use a Mark V Speedtronic TMR Panel, so i will appreciate information on how to check and adjust them if need be for proper functionality of the servo valves.

I will search and study on how to troubleshoot alarms from the former posts of MarkVguy as you recommended using dynamic rung display and CSP.

Thank you all for your professional support. I look forward to more interaction & knowledge sharing.

 

Dear MikeVI,

I appreciate your concern and your inputs. They were all very on point. I have just posted a reply to CSA's recent comment on this post which contains information that I meant for you too. Pls read it and let me know your thoughts and any further inputs. Thanks for your cooperation.

 

Dear Terkuma Ivande,

thank you for the reply. I think you comment of the gas yard pressure control valve is a great observation and very well could be contributing to the problem.

I would urge you to read and investigate any source of information regarding the machines you maintain. The more you know and understand, the less you will need us.

I will let CSA comment on the gain and null bias values for the MKV, my knowledge of that system is very small compared to others, it was slightly before my time, not to date anyone here! My thought would be to get servos that match what the system is designed for. The gain and null bias numbers may be a problem if someone changed them trying to correct a problem before that was unrelated, but that then begs the age old question of "when did this start?" Many times I think things are changed or modified as part of the turbine control system to attempt to fix the real problem. I'm sure you agree that the PCV in the gas yard is just one possible example.

In short, there is a wealth of help and information here, especially from MR. CSA who spends time here I only wish I had. And has knowledge I could only dream of now. But for me (and CSA as well I think) the desire to help people is proportionally equal to the time they have spent trying to help themselves.

 

Teekay/MIKEVI,

The servo gain and null bias can be obtained from the I/O Configurator on the operator interface (I don't think we know if it's an <I> running IDOS or a GE Mark V HMI running MS-Windows and CIMPLICITY--but we should, be told, that is). And you will need to use the operator interface that is used to download settings and information to the Mark V with the problem (sometimes there are multiple operator interfaces which can communicate with multiple Mark Vs on a site). If there are multi-unit operator interfaces you will need to find the one used to make changes to the Mark V in question.

Once you start the I/O Configurator, you will need to click on the TCQA card target and scroll to the Servo output screens. There are eight servo outputs on a Mark V, and the SRV and GCV are usually assigned to the first two, SVO1 and SVO2, so you need to fin the screens for SVO1 and SV02. You will then need to look for the fields labeled GAIN and NULL BIAS. The SRV will likely have a couple of fields labelled GAIN, write down the exact name and value of both fields. Don't forget to get the NULL BIAS current values for SVO1 and SVO2.

Once you have the information recorded, just exit the I/O Configurator without saving any changes (you shouldn't have made any changes, but choose the target at the bottom which says 'Exit Without Saving' anyway).

I'm making a presumption the SRV and GCV are assigned to SVO1 and SVO2. If they aren't, then they will be assigned to SVO7 and SVO8. You can be certain by looking at the I/O Report for the unit; the technicians usually have a printed copy of it to refer to. If not, there is a file named TC2KREPT.TXT in F:\UNIT1. You can look in that file for the information, but it's not easy to find. The file is an ASCII text file, and it can be viewed using any file viewer/editor. Search for SVO1 and SVO2 and you will likely find the SRV and GCV are assigned to these outputs. This will just ensure the information you retrieve from the I/O Configurator is correct.

 

Dear Mike,

Thanks a lot for your time. I definitely need to read and study more about the systems I'm operating with. I will do so. I however do appreciate your support. The unit in question has currently being shut down for further investigations about the unsteady P2 pressure causing the hunting. I will publish the findings when it is resolved.

Cheers

 

Thanks a lot CSA for your time and support.

I'll keep you posted on current developments as we troubleshoot the unit further. I saw the regulator gain and current bias in the IO Configurator. We are using Mark V HMI running Windows with Cimplicity.