GE Frame 9E Bleed Off Valves Passing

Gas turbine main filter replacement activity was carried out in foggy conditions, and after that off line water wash not carried out. After one day, compressor efficiency was reduced. Later it was identified that GT generating 4Mw less as compare to before filter replacement activity.

Thereby, GT was again shutdown for off line water wash activity. And after startup, nothing changed. Compressor was generating less MW. And then, we checked bleed off valves through Temperature gun and found 3 out of 4 bleed valves are passing.

Why this happened? It looks strange. Is it due to dust or some other factors? Need support.

The biggest reason for axial compressor bleed valves failing to close completely is moisture in the air used to actuate the valves. MOST GE-design heavy duty gas turbines use air from the axial compressor discharge to actuate the axial compressor bleed valves. That air can have a lot of moisture in it just from ambient humidity, and, if the isolation valves to prevent moisture from getting into the actuator air supply piping are not properly closed prior to an off-line axial compressor water wash then even MORE moisture in the form of water can get into the actuator air supply piping and cause the actuator to fail to close completely.

And, some installations failed to put low point drains in the actuator air supply piping to be opened briefly after an off-line axial compressor water wash to drain water from the piping.

Finally, there is supposed to be a filter, a poro-stone filter, in the actuator air supply piping, and in the bottom of that filter canister there is supposed to be a continuous blow-down orifice (really just a small piece of pipe with a plug on the end and a very tiny orifice drilled into the pipe). This continuous blow-down orifice is intended to provide a continuous path--through the orifice--for small amounts of moisture to be blown out of the supply piping and the poro-stone filter. However, many people mistakenly believe the small amount of air/moisture exiting the orifice to be a leak and they subsequently remove the orifice and plug the tap in the filter canister. OR, since the filter canister is usually made of carbon steel it rusts and rust then blocks the orifice. So, in either case, the continuous blow-down is rendered useless, so no moisture is removed from the actuator air supply piping.

And, almost NO site ever opens the poro-stone filter during maintenance activities to remove rust or check the condition of the poro-stone filter. So, rust and dirt from the filter canister can also sometimes make their way in the axial compressor bleed valve actuator.

A few sites have switched the axial compressor bleed valve actuator air supply to dry instrument air from the plant compressed air system. This is an excellent idea, if the plant has one, and it can accommodate the require flow to close the valves (the actuators only require air to close the valves once during a start).

So, check the condition of the actuators and you will likely find rust and/or debris has made their way into the actuator, along with water and moisture, and that has caused the actuator to bind and not fully close.

This is not an unusual problem, especially the lack of attention to manual isolation valves during off-line axial compressor water washing, and if the site is located in a humid ambient location, and/or uses any kind of inlet air cooling (evaporators; chillers; fogging; etc.). It's a good idea to make sure there are proper low-point drains in the axial compressor bleed valve actuator air supply piping, and they are used after axial compressor off-line water washes, and if possible to switch the axial compressor bleed valve actuator air supply to plant air supply from a tap off the axial compressor discharge.

Please write back to let us know what you find.

We have carried out online water wash and Gas turbine MW restored back to original. Is there any possibility that dust has entered the compressor and after online water wash, fouling removed?

Firstly, This actually shows that some dust or oil is entering the compressor and causing fouling on compressor blades and after online water wash it removes.

Secondly, is there any possibility that filters are leaking unfiltered air?

Thirdly, the quality of filter since the initial dp last time was more as compare to this time. Delta of dp was 17mm h2o less as compare to last time.

Please suggest keeping in view above scenario.

Nothing is easy with you. And, the information you provide seems to be inconsistent, too.

Axial compressors do foul--and dirt/dust can cause the fouling. But not all dirt/dust is the same. There is dust from roads; there is dust from nearby cement plants, or gravel pits. There is soot from lorries and cars. There are contaminants from nearby oil- or petrochemcial plants, or refineries, or fertilizer plants. Any hydrocarbons in the air ingested into the axial compressor will also contribute to fouling of the compressor. They can be most evident on the IGVs (Inlet Guide Vanes), and if present prior to an off-line compressor water wash, they should be manually removed. If there is a L.O. leak of the #1 bearing, and that oil is being ingested into the axial compressor, it will also contribute to fouling of the axial compressor. So, there are a lot of possibilities for fouling and contaminants to get into and cause fouling of the axial compressor. And all sites are different, and all contaminants are not the same. Being near a road or highway with high traffic and lot of lorries (diesel-engined trucks and buses) can cause problems. Or, of the roadway is dusty. Some desert environments have lots of very fine dust that can cause fouling of the filters and the axial compressor, and if if rains heavily during some times of the year filter elements can become swollen and not allow much air to get through--especially if the filters are already dirty. And, then they rupture. (Many gas turbine inlet air filter/ducts have implostion doors which are designed to open before filters rupture. Sometimes the opening of the implosion doors cause a turbine shutdown, or a turbine trip--because once they open, they allow unfiltered air to enter the axial compressor. But not all units have implosion doors, and few implosion doors are ever tested to make sure they operate smoothly when they need to.)

But what makes dust/dirt/soot even worse for fouling of an axial compressor is if the ambient is humid, or if any kind of inlet air cooling is used on the turbine. The humidity mixes with the contaminants and as they pass through the axial compressor, the "plate" or deposit on the compressor blades (rotating and stationary) and reduce the air flow through the axial compressor.

And filters don't prevent ALL dust and dirt and contaminants from getting into the axial compressor. All filters have a minimum particle size rating, and if particles (dust, dirt) are below that size the filter can't block them. There are different kinds of filters for different expected airborne contaminants and conditions. Do you know the nature of contaminants at your site, and do you know the particle rating of the inlet air filters at your site?

Yes. Filters can always leak, usually if not installed correctly. And the differential pressures you cite, going from high to low, suggest one or more filters may have ruptured--meaning the dp was so high across the filter element(s) they ruptured and most of that dirt on the filters made its way into the axial compressor, contributing to fouling. That would explain the differential pressure decrease from one reading to the next.

It's very easy to find ruptured filters--or to find leaks in the inlet filter housing and inlet duct work. On a bright sunny day, go inside the inlet filter house and inspect the filters, looking for daylight. Look at all the seams and joints of the metal inlet filter house for signs of daylight. Anywhere you see daylight, that's a potential leak, allowing unfiltered air to get into the axial compressor. And, if there's humidity in the air (either from rain or fog or just high ambient humidity or inlet cooling) any dust/dirt/soot will mix with the humidity and cause fouling of the axial compressor, which will reduced compressor efficiency and cause the gas turbine-generator output to go down.

A lot of inlet filter houses were constructed using stitch welding on the seams of the metal enclosures--meaning the seams were only welded in short lengths. The sections in between, over time, can enlarge and allow unfiltered air to enter the inlet air duct-work and the axial compressor. So, many sites have to re-weld the seams, continuously, to stop the leaks.

Hope this helps--and I hope I considered all the things you want considered.

I want to add, before you ask, that there is NO SET RECOMMENDATION for when to perform an off-line water. It depends entirely on the type of contaminants (dust; dirt; soot; hydrocarbons; etc.), the amount of humidity in the air, and the condition of the filters and the attention paid to them during installation and the condition of the inlet filter structure and ductwork. All of these things can vary from site to site, and even from turbine to turbine on the same site.

Some sites will orient the turbine specifically so that prevailing winds will not blow nearby dust and contaminants into the inlet air filters. Some sites don't do this.

I was at a multi-unit site where one of the unit was sited near a dusty road with lots of truck traffic and it needed more frequent off-line water washes and filter changes than the other units. And, the site staff wanted to know why....

Some sites can go a little longer than other between off-line water washes because they don't need to restore the lost power so quickly. So, the decision of when to perform an off-line water wash also depends on the site's tolerance for performance loss--if it's a big deal, then they will need to be done more often. If not, they don't need to be done as often.

And, lastly, because an off-line water wash requires many hours to complete, there is lot generation during that time. And, sometimes, sites can't always afford the lost generation--and if the unit exhausts into an HRSG (Heat Recovery Steam Generator--a boiler) and the steam is needed for a nearby plant/process, sometimes the unit can't be shut down as often or for as long because of the need for the steam.

So, every site needs to choose when to perform an off-line water wash based on their site ambient conditions (which can and often do change throughout the year) as well as their appetite or tolerance for lost power generation--and lost steam generation if that's a factor, also. Making the decision very, Very, VERY site-specific.

The GE-design heavy duty gas turbine control community here at has been around for more than 15 years. Many topics--very similar to this--have been discussed MANY times before. And, they can all be read by using the cleverly hidden 'Search' field at the far right of the Menu bar of every webpage. (It's recommended to use the Search 'Help' function the first couple of time as the syntax is NOT like the most popular World Wide Web search engines.) You can find lots of information on lots of topics, mostly controls-related, some not so controls-related (like this thread--which has nothing at all really to do with controls, though we like to answer as many questions as we can, sometimes we have to just say this doesn't qualify).

Also, if you like the information provided in response to your questions there is 'Thumbs-up' icon you can click on to show your satisfaction if you don't want to reply with a short message to say thanks for information. There's also a 'Thumbs-down' icon you can click on to say you didn't provide the information helpful or useful--but if you do use that icon we do ask that you tell us why the information was not useful or helpful so we can improve the quality of our responses.

And, as you search past threads and topics (and we hope you do!), you will find that many original posters provide feedback if they found the information/answers useful, or not, as the case may be. This let's others who read the threads--both now and in the future--if the information was helpful or not. The amount of feedback provided to the threads here on is far and away higher that usual for most other on-line forums--and that lets people know if the information provided was useful, or not. As well as the posters--who provide their knowledge and experience free of charge. So, we like to encourage all new people to the site to provide feedback, and if you don't like the information that was provided because it wasn't useful or applicable--please let us know because we want to provide good, useful and helpful information. If you can take a few minutes to post a question, we feel you can take a few minutes to leave a sentence or two if you found the information useful or helpful--and help others now and in the future.

Thanks, and welcome to! Hopefully you'll learn a lot here, and even provide the benefit of your knowledge and experience when you can!
Its unusual to us also.

The main concern is we have checked the filter house twice and no abnormality was found.

Moreover, we have witnessed higher DP replaced back in Dec 2016 and now in Nov 2017 we have witness less DP. A difference of 17mmH2O was observed with Latest has produced less overall DP.

Is it possible that Evaporator Media filter is degrading and its particles are entrying the compressor. Evaporator media is installed back in 2010 and were not replaced yet.