We have two GE 7 FA machines normally operating at Baseload with IGV servo currents at +/- 1 to 3 %.One of the machine's servo currents increased almost 4 months back to -7 to -12% at baseload and running without any alarms.

The machines were selected to part load as per loading scheme and operating without any abnormality.After 10 hours,it was observed that the Unit-2 servo currents slowly increasing and the hdyraulic pressure dropped gradually.

The hydraulic pressure dropped to 106 bar from 112 bar and afterwards,the servo currents increased to -30 to -50% and maintained there.After 6 hours,the machine was selected to base load and found no change in the currents. Hydraulic pumps C/o was done.found no difference.IGV actuator was checked for any oil leaks but found no leaks.

Machine was responding normally to the load setpoints but the servo currents are higher.

The following are our observations.

1) IGV should open to 86% at baseload and normally operated at 85.7 to 86%.But now operating at 84.9 to 85.5% only.

2) There was an abnormal noise observed on both Hydraulic pumps and Maintenance checked it.They doubted of some valve passing/accumulator trying to support the system.

3) No other servo's of GCV's and SRV's are affected.They are having no problem.

Our doubts are :

1) When the Servo suicide function will occur.

As per software configuration,Fbksucide and curr_suicide are enabled.

Fbksuicide 5.0 short servo current output if position exceeds range limits (%)

Curr_suicide 5.0 short servo current output if current error exceeds the limit in (%).

TMR_diff limit 25.0

So if the CSGA_R,S & T (not voted)are matching ( or near),then is it okay to run the machine? what exactly causes the suicide function.

We had GT trips on Exhaust Temp due to IGV servo suicide before 1 year but that time we noticed some diagnostic alarms and IGV position fault.Now there are no alarms or faults.

Did anybody experience the same problem before and continued operation like this.

2) Why IGV opening is lower when compared to before at baseload ? CSRGV is 86 %.

3) If the problem is due to hyd.pressure drop then it should cause some change in other Servo's.But it did not happen.So can we rule out this concern?

Thanks in advance for your replies.

sudheer

 

A lot of things to respond to here. And a lot of things which weren't mentioned. Proper operation of servo-valves and their actuators requires design hydraulic pressure, properly charged and operating accumulator, clean filters (main hydraulic filter and "last-chance" servo/actuator filters), properly adjusted hydraulic pump compensators, properly operating air-bleed check valves, and properly adjusted relief valves.

I have been to many sites in the last few years where the pump compensators and the relief valves were improperly adjusted and were causing excessively high flows through the hydraulic pumps. This was proven by measuring the current drawn by the Aux. Hyd. Pumps (the AC motor driven pump) which was found to be in excess of nameplate rated current. When the air bleed check-valve and relief manifold was accessed it was also obvious that relief valves were passing.

When the pump compensators and the relief valves were properly adjusted, everyone involved commented on how much quieter the hydraulic pumps were operating compared to the as-found condition.

First, if you are attempting to draw any parallels between servo current changes and subsequent changes in hydraulic pressure, that would be a mistake. While a fluctuating servo current could induce pressure fluctuations (presuming the hydraulic accumulator isn't valved-in properly or isn't charged properly), it's extremely unlikely that a change in continuous servo current value(s), particularly by one processor, would induce a change in operating hydraulic pressure. Extremely unlikely.

Second, as for actual IGV position versus reference IGV position, if the actual position isn't equal to the reference then the servo-valve output regulator will change the servo current to try to make the actual equal to the reference. If there's not enough hydraulic pressure to keep the IGVs open to the required (reference) position, the regulator will try to drive the IGVs to the required position by changing the servo current, but it's not a proportional-plus-integral controller so it won't drive the actual to be equal to the reference. From the information you provided, it seems the servo outputs are trying to do just this. Negative servo current increases the flow of fuel or air and the fact that the servo currents are increasing in the negative direction implies that the regulators are putting out more negative current to try to drive the IGVs closer to the reference position.

But something is not "cooperating" in this endeavor. And, low hydraulic pressure is one obvious culprit.

I also have seen some very unusual regulator gains and null bias current values in many of these same machines, and we don't know what values are in your machines versus the values in the Control Specification drawing. Low hydraulic pressure can manifest itself in many different ways depending on the configuration of the machine. And if the regulator gains and null bias current values aren't per the Control Specification, then all manner of unusual things can happen.

In my personal opinion, the IGV actuator requires more torque to control the position of the IGVs than the SRV- or the GCV actuators require to control the positions of those valves. The IGV actuator is much larger than either the SRV or GCV actuator. The IGVs have a much larger total surface area than the plugs of either the SRV or GCV, even taking the springs of the SRV and GCV into account. So, low hydraulic pressure would likely, in my mind, primarily affect IGV position first.

You didn't mention anything about the hydraulic accumulator valving, its pressure, or the adjustment of the hydraulic pump compensator(s) or the relief valves. Usually, when hydraulic pressure is low, the first thing one does is to try to adjust it back to rated, but you didn't indicate that had been attempted. Just that an assumption had been made that any drop in hydraulic pressure would be equally seen in all hydraulically-operate devices. Which may, or may not, be true.

As you noted, you will likely have Diagnostic Alarms before the suicides are actuated.

 

Thanks Mr. CSA for the response.The machine tripped after 22 hours of operation with the Servo currents fluctuated.

The servo currents increased to -97 to -100 % and then the machine tripped on the "Exhaust over temp Trip".

This time we did not get any diagnostic alarms.

The Hydraulic pump compensators were not adjusted as the pressure was normal when started the standby pump. It slowly reduced only.

After the machine trip, the sound from the hydraulic pump's and accumulator has gone and they are very silent again.

My question was:

1) The servo currents maintained at -10 to -15 % for a long time and they slowly started to increase and the hydraulic pressure decreased at the same rate.

Could there be any leak in servo valve so that enough Hydraulic oil is not flowing to the IGV actuator.

2) When we started both pumps also,there was no change in servo currents or IGV mismatch. What could be the reason?

Thanks,
Sudheer

 

I don't know where you are monitoring the servo currents, or if you're adding the three values together, or what. It's very confusing. It would really help if you would tell us where you're reading these values from and if you're adding the three processors' values together; in other words, how you're arriving at these numbers.

Each processor should, under normal conditions, be putting out approximately -2.67%, +/- 1.33%, so from -1.33% to -4.0%. That's from each processor. If you added all three together, the sum should be approximately -8.0%, +/-4.0%, so between approximately -4.0% to -12.0%.

What are the null bias current values in the Toolbox file for the IGV servo-valve output?

What are the servo current gain values in the Toolbox file for the IGV servo-valve output?

Have you verified the wiring/polarity of the current being applied to all three servo-valve coils from the three processors?

So, the kinds of numbers you are reporting are not what I would normally expect to be seeing. Especially the -97% to -100% numbers.

There certainly seems to be something amiss. Please review the Piping Schematic diagrams (P&IDs) for the Trip Oil System, the IGV System, and the Hydraulic Oil System. The high pressure hydraulic oil goes through a "dump valve" (I forget the device number of the dump valve), the position of which is controlled by 20TV-1, and then to the servo-valve. There may be some problem with some portion of the system ahead of the servo-valve, or even after it, that's causing the servo-valve to appear to behave badly. I'm not saying it's not the servo, but they get incorrectly blamed for a lot of problems.

I believe for F-class turbines the servo-valve is visible on the outside of the actuator, so if it was "leaking" then there would likely be a mess around the actuator.

If you're talking about internal leakage, that would be possible, and I don't know how one would detect that, short of replacing the servo-valve.

The noises emitted by the axial-piston hydraulic oil pumps used on GE-design heavy duty gas turbines are usually pronounced when there is high flow-rates through the pumps. In general, the hydraulic system is pretty static, because when the turbine is operating at a stable load and output, the fuel valve and IGV positions are not changing so there is very little flow of hydraulic oil. When there is high flow, the pumps have to work harder (and the current drawn by the motors increases--hint, hint!) and the noises get much louder and pronounced. That's also why the valves on the hydraulic accumulator have to be in the correct positions, in particular, the "bleed" or drain valve must be fully closed during normal operation/configuration. And, of course, the block (isolation) valve must be open. And, the bladder(s) must be properly charged.

So, from this latest description it would sound like the flow through the pump(s) had increased, significantly.

If there is some problem with 20TV-1 or the dump valve is controls and the hydraulic oil wasn't getting to the actuator and the Speedtronic had to increase the servo-valve current to try to keep the IGVs at the reference angle. There were likely some Diagnostic Alarms about the high servo currents, but that's just an educated guess.

I have also seen intermittent operation and poor adjustment of the hydraulic relief valves cause problems with hydraulic pressure. The intermittent problems were also seemingly linked to oil temperature problems.

It's just good troubleshooting practice to try to eliminate all possible causes of a problem which might cause the servo-valves or the Speedtronic servo current outputs to appear to be the problem. I've seen, <b>literally</b>, hundreds of servo-valves needlessly replaced over a couple of decades-plus. And, they're not inexpensive. Nor is the lost generation that's necessary to replace them.

The manufacturers of turbine lube oils have changed their formulations in recent years. The result is an improvement in the performance of the lubricating oils, for the turbine bearings and auxiliaries. However, this change has had a negative impact on the oils when used as the hydraulic medium through servo-valves. GE has published a TIL (Technical Information Letter) on this, TIL-1528, and I think the latest revision is -3. (And, no; I can't send anyone a copy of the TIL; I don't have one, I've only recently seen a report about it. You'll need to get it from GE, or perhaps someone who has an electronic copy can offer to send it to others.)

I believe that BP-Castrol have also developed a new turbine lube oil to address the problems caused by the previous formulation change(s), and I think it's called Perfecto XPG. It's supposedly been used with good results on a couple of site in the UK/Europe.

Oil condition is very important, and for some odd reason in my perception (<b>not</b> by any data-based analysis!) it seems to be particularly critical to F-class machines. Maybe it's just because they're so big (output) and they are generally more "visible" in a power system that reliability issues are more magnified. Or that F-class units are so expensive that people demand more of them and complain louder about problems with them. I don't know what it is, but they seem to be more "susceptible" to or "affected" by oil problems, and, again, that's just my personal perception. (Maybe it's because I'm not fond of F-class turbines; who knows?) So, it would behoove you to review the above TIL and make some investigations into the issues it raises, for future reliability.

 

Here's a link to some more info about the lube oil and some experience with it:

https://absolute.castrol.com/files/342312_GE_Frame_6FA.pdf

 

---I don't know where you are monitoring the servo currents, or if you're adding the three values together, or what. It's very confusing. It would really help if you would tell us where you're reading these values from and if you're adding the three processors' values together; in other words, how you're arriving at these numbers.

I mentioned the servo currents from the R, S, T and the voted outputs.

Thanks,
Sudheer

 

Sudheer

by your explanation data I catch that your igv feedback is pushed below set point reference so that servo current keep at relatively negative value. my opinion is this problem could be caused by : lower hydraulic pressure (your hyd pressure 100psi drop) , too high null bias at servo valve, LVDT or even mechanical restriction on igv movement.

do you have information what is actual IGV angle when feedback display is 84.5 DGA?

I think if you have chance you have to take more data ( ref , feedback, servo current, actual IGV pos) by performing IGV simulation step by step open & close IGV. (of course at this case you will use aux hydraulic pump). have you increase hydraulic pressure? Is there any chattering phenomena at servo valve tubing to IGV cylinder actuator?

If you can share trip log data I think it will help as well

 

Hi,

One day after the replacement of the servo valve, again the fluctuation in the servo currents started. We checked all other parameters and found normal. The next day servo currents became normal on it's own and were maintaining normally.

After 2 days,We had an opportunity to take a shutdown for some other job and it was used to do IGV stroking.

During this time, it was found that the reason for this the metallic tags (plant identification tags) which were recently applied to the LVDT's. They were obstructing the movement and they even got bent. So they were removed and now there is no other problem.

Thanks to All.
Sudheer

 

Thank you very much for the feedback!

 

Can you please explain how the metallic tags affected the servo valve?

These were applied to LVDT's,so if there is any fluctuation in the LVDT's reading should also effect the IGV position indication.

But from above posts, there is no indication of IGV position disturbance?

Can you please clarify more?

Thanks,
GT-ops