AA Compressor Related Problem

K

Thread Starter

kh000

In frame 6B machine, Some times Sudden Temperature drop of 25 to 30 Deg C observed in Atomising Air Temp. and at the same time AA pressure starts hunting and condensate also observed from AA1 drain. GT was running on Gas Fuel at that time. It remains with same condition for 2 to 3 days or more and after that all three parameters get normal on its own. The same phenomenon has been observed 3 to 4 times in past. No alarm appears in mark during this action.

Pl. help in analysis.

(As per my knowledge, main problem is AA compressor discharge pressure hunting which in turn is causing AA temperature drop resulting in condensate from AA1 drain.)
 
khoriwal000,

What fuel is being burned when the problem occurs?

Typically, the AA Compressor discharge air temperature is not monitored--ONLY the Inlet (suction) air temperature. The air temperature entering the AA Compressor should be approximately 225 deg F--above boiling point, so that no condensate is present. The Inlet air temperature should not be above approximately 275 deg F; if it's above 300 deg F the AA Compressor can be damaged--severely. AND if condensate/water gets into the AA Compressor it can also severely damage the Compressor.

The Atomizing Air Pre-Cooler Temperature Regulating Valve (usually VTR2-1) is misadjusted or is not working properly and that causes entrained humidity in the axial compressor discharge air (CPD) to condense, which is the source of condensate. If there is coolant in the condensate you are seeing at the Inlet to the AA Compressor then the cooler bundle may be leaking (not likely, but possible).

AA Compressor discharge temperature is not the problem--it's never the problem (which is why it's not typically monitored); refer to the P&ID for the Atomizing Air system. It's the AA Compressor inlet air temperature that's the problem, and misadjustment or a non-operating temperature regulating valve is the usual culprit. The AA Compressor runs at something like 14000 or 44000 RPM--so if someone is standing in the vicinity when it fails they could be seriously injured, or worse. Again--high AA Compressor inlet air temperature and/or a slug of water/condensate can damage the Compressor--and will eventually do so if not corrected.
 
K
sir,

ANS-fuel being burnt is NG during the abnormality.

sir, according to your explanation, mis-adjustment of VTR2 below 225degF-->condensate generated-->same condensate entering the AA compressor suction and causing problem.

sir, if cooler tube is damaged and leaking, then air (>7kg/cm2) should go inside water (4-4.5kg/cm2) circuit (to a higher point or vent). How can it be a cause of this abnormality?
 
sir,

>sir, if cooler tube is damaged and leaking, then air
>(>7kg/cm2) should go inside water (4-4.5kg/cm2) circuit (to
>a higher point or vent). How can it be a cause of this
>abnormality?

I said it wasn't likely, but it has been known to happen--particularly if the unit is frequently started and stopped?

And have you confirmed the operation of the temperature regulating valve?

Could there be another source of condensate?
 
K
Sir,
VTR2 valves auto operation is not in function and AA temp is maintained by manually closing it or opening it.

But during above said abnormality AA temp was near to 112-115 deg C.

Another source of condensate can't be seen by me.
 
khoriwal000,

The only sources of water moisture in the axial compressor discharge are:

--humidity (including fog) and rain (which may be ingested through broken filter elements or leaky inlet filter house structure joints/seams)

--evaporative cooling (which affects humidity)

--leaks from AA Pre-cooler tube bundle/nest

--water from Off-line Water Wash (not properly drained after rinse)

Any water or moisture (humidity; fog; evaporative cooling) which enters the axial compressor through the bellmouth/IGVs will be heated as it passes through the compressor, and will be cooled as it passes through the AA Pre-cooler. If it's cooled excessively, it will condense.

Condensate can collect in piping low points, which should be equipped with continuous blow-down orifices (very small holes drilled in small pieces of pipe which are provided (or should have been provided by the turbine packager) to continuously allow any condensate (small amounts of condensate!) to be blown out of the piping. MANY sites believe the continuous blow-downs are leaks, and close the manual valve upstream of the orifice--which completely defeats the purpose. Condensate can pool in low points, and remain there until the pool becomes so large that it gets pushed along, into the Main AA Compressor inlet, which can have BAD consequences.

Water remaining from an off-line water wash/rinse, can also pool in low points if not properly drained, or allowed to exit through the continuous blow-down(s).

Because the continuous blow-down drain orifice(s) are VERY small, they are easily plugged with rust and other debris. They should be cleaned during every maintenance outage--but RARELY are. And, the manual isolation valves upstream of the continuous blow-downs MUST be left open at all times, unless the orifice gets worn (which it will) and causes excess leakage of hot air into the compartment. These orifice assemblies are almost always overlooked, and almost never checked/replaced during maintenance outages.

You have never said how much water/condensate you have witnessed coming from the drain, and where the drain was located, and if you had to open a manual valve to view the condensate running out of the drain. (All questions which would be great to have the answers to!).

It's presumed you have a T/C (or T/Cs) mounted in the Main AA Compressor inlet providing indication to the Mark*. If not, how are you monitoring Main AA Compressor inlet temperature?

A manually-operated VTR2-1 is likely the culprit. If the unit is being unloaded and loaded during normal operation, and operators are making manual adjustments to the temperature regulating valve, are they logging the changes? Are they logging the pre-change temperature, and then checking the post-change gemperature (say 15 minutes or so after the change)? Because there is a LOT of air flowing through the AA Pre-cooler, and it can take a long time for any manual change to take effect.

If they are manually changing the temperature regulating valve position while having an operator at the HMI tell them what the temperature is, it's VERY likely the change is EXCESSIVE and over time will go too low or too high. Patience is the key to making adjustments to VTR2-1 (or even VTR1-1, for that matter). There is a lot of latent heat in the heat exchangers and the effect of adjustments can't usually be properly determined for 15 minutes or so.

If the unit is started and stopped frequently (once a day; two or three times a week) and operators are manually adjusting VTR2-1 during starting and operation--you can be very confident that misadjustments are causing problems. And could lead to pooled condensate in the piping, which may or may not be being drained through the continuous blow-downs.

People are ALWAYS saying that VTR2-1 (and VTR1-1) are not working--but that's because they don't know how to properly adjust the automatic mechanism. Here's a link to the Manual for the valves typically used on GE-design heavy duty gas turbines:

http://robertshawindustrial.com/Manuals/MAN160_RT1010.pdf

The applicable information is on Page 4, Section III, and Figure 7.

They are VERY simple and robust valves--IF THE CAPILLARY TUBE IS PROPERLY SHIELDED AND NOT ABUSED DURING MAINTENANCE OUTAGES. The capillary tube is RARELY properly installed during construction, and maintenance personnel drop things on it, step on it, and even use the top of the bellows mechanism as a step (which operators have also been known to do). When the capillary breaks, the liquid inside leaks out (with a nasty odor), and then it can't work properly.

But MOST people assume the handle on the side of the actuator is how it is to be adjusted for automatic operation--which it is NOT.

Now, here's the thing about VTR2-1. It's automatic setpoint should be approximately 225 deg F. When the unit is shut down, VTR2-1 will CLOSE to try to increase the AA Pre-cooler exit air temperature to 225 deg F. And, when the unit starts up, it will be closed--restricting the flow of water to the Pre-cooler. And, it's slow-acting--so it's <b>VERY</b> common for an 'AA TEMPERATURE HIGH' Process Alarm to be annunciated for a few minutes (10-20 minutes usually) during start-up as VTR2-1 moves to increase the flow of cooling water to the Pre-cooler to reduce the exit air temperature. This is NORMAL, and because VTR2-1 is a slow-acting valve, it will almost always occur (if the unit is starting from a "cold" condition). And if people don't wait and allow the actuator to come to "equilibrium", they can make an excessive manual adjustment, walk away from the valve, and then find condensate coming from the Main AA Compressor inlet blow-downs/drains.

These are NOT high-tech valves--but they ARE adequate for the application, though VTR2-1 does cause nuisance Process Alarms on start-up. Trained and experienced operators should recognize this and be accustomed to it--and if the alarm doesn't clear in 20 minutes or so after reaching rated speed, then take appropriate action. If they have HMI indication of Main AA Compressor inlet air temperature, they should see the temperature dropping slowly when the alarm is first annunciated, and continuing to drop as the unit runs for a few minutes, until the alarm clears as the temperature drops to normal (approximately 225 deg F).

One has to remember: These valves have been SUCCESSFULLY used for DECADES on GE-design heavy duty gas turbines. They are robust and adequate for the application. When they were first used, the turbine control systems did not have the I/O capability to operate an electric- or pneumatic control valve for these applications (VTR1-1, and VTR2-1). And, GE designs build on proven processes and schemes, without unnecessarily complicating designs (except for GE Belfort). Very large units (F-class and above) now use pneumatic control valves for controlling Main AA Compressor inlet temperature which get a 4-20 mA signal from the turbine control system--but these valves require a source of clean, dry instrument air and they are not cheap. The Robertshaw temperature regulating valves are self-actuated, and require no external control signal or source of pressure/force. When properly installed and maintained and operated, they are virtually set-it-and-forget-it devices. Yes; there can be nuisance Process Alarms--but, I don't know of one process that doesn't have something similar. And, for a properly configured unit, this should be the <b>ONLY</b> nuisance Process Alarm during a start-up. As such, it should be easily recognizable, and dealt with (if somewhat of a nuisance).

Again, we don't know how you're checking for condensate. We don't know how you're monitoring Main AA Compressor inlet temperature. We don't know how the valve is being manually adjusted during start-up and loading/unloading. And, it will need to be adjusted if operation is switched from natural gas to liquid fuel, and after switching back from liquid to natural gas (the air flow-rate through the Pre-cooler should change as a function of the fuel being burned; more is required for liquid fuel, and less is required for natural gas fuel). We don't know if the continuous blow-down drains are working/clear. We don't know if the unit has evaporative cooling, or some kind of inlet air cooling. We don't know how the unit is prepared for off-line washing, and how it's prepared for operation after off-line water washing/rinsing.

But, again--and, as you have noted, there are limited options for water/moisture in the inlet air flowing into the Main AA Compressor. VERY limited options. If you're seeing condensate (and condensate that's free of cooling water solution), chances are very high that it's coming from the AA Pre-cooler condensing moisture from the axial compressor discharge air because VTR2-1 isn't working properly, or is mis-adjusted. If the temperature regulating valve is being operated manually, and the unit is being started and stopped, or loaded and unloaded, frequently, and/or if fuel is changed frequently or often, then a manually-operated VTR2-1 is going to take a LOT of manual adjusting. And, if it's not adjusted and then the results are checked 10-15 minutes after adjustment, then there are likely going to be issues with air temperature and moisture condensation.

Please write back to let us know what you determine!
 
Top