Inlet Filter DP

C

Thread Starter

Cris

What are the importance of Inlet Filter DP in GT?

If choked we may not have enough air flow. If icing formation, similar situation but what are the other reason or what are importance of having it?

Looking for details explanation. Thanks all for providing valuable information.
 
I'm not quite sure I understand the question. I think you are asking why inlet filter differential pressure is monitored and used to alarm, shutdown or even trip a unit.

I have seen several sites that ignored high inlet filter differential pressure alarms that resulted, initially, in low turbine performance (the axial compressor is "choked" as you say so the mass flow of air is limited which reduces power production), and then several of the inlet filters actually ruptured. The sand and dirt that had been trapped in the filter elements was then sucked into the axial compressor, and in two cases there was foreign object damage to the compressor resulting in tens of thousands of dollars in damage to the axial compressor, and tens of thousands of dollars in lost generation, and in one case, hundreds of thousands of dollars in lost steam production which caused the adjacent chemical processing facility to lose several days of production runs because the standby boilers weren't able to be put in service because of lack of maintenance.

Many, but not all, GE-design heavy duty gas turbines have implosion doors on the inlet filter ductwork. The function of these doors is to open if the differential pressure gets too high in order to prevent collapsing filters, or worse, collapsing inlet duct walls. (Vacuum can be very powerful!) When these doors open under high dp, they allow unfiltered air into the machine, and many times the opening results in an automatic shutdown of the unit. But, at least it prevents filters rupturing, or worse, inlet duct work from collapsing. I have not personally seen collapsed inlet duct work, but have seen photos from two sites where it occurred, one where the implosion doors were rusted shut and could not open.

Does that answer the question about why it's important (critical) to monitor inlet filter differential pressure?
 
We just don't have enough information. It seems you are not experiencing icing of the IGVs, just the inlet filters; is that correct?

What have you done to try to troubleshoot the problem?

What were the results of the efforts?

Have you spoken to the turbine OEM or packager?

Have you spoken to a filter manufacturer/supplier?

I've been to many gas turbine sites on coastal areas which experience fog many nights of the year and in the winter the ambient hovers around 2-5 deg C or even lower on occasion. These sites do not experience icing because of the orientation and/or placement of the unit and it's inlet duct. Some sites which have dusty ambients during the day do experience some high inlet filter dp's when the humidity is high, but I've not seen icing.

In my experience, icing of the inlet filters is the result of an un-natural source of humidity in the inlet air stream. For example, if there are evaporative cooling towers nearby and at certain times of the winter the wind blows the vapors from the cooling towers into the air filters I have seen icing. These sites have used the self-cleaning inlet air filter mechanisms to loosen the ice to prevent complete blockage (meaning the pulsing mechanisms have to be run extensively when conditions arise).

You have not provided enough information for us to be able to help with the problem.

How long has this problem persisted? Since commissioning or since some change on site or in the type/style of inlet air filters being used?

Was the inlet filter house replaced recently and then the problem started?

Is there some un-natural source of humidity/vapor that is blown into the inlet air stream?

What type of filters does the you have? Square or rectangular pleated filters? Circular pleated filters?

Are pre-filters used on the inlet air filters?

Is the site dusty at times, or is there a cement plant nearby? An unpaved road or roads?
 
Some GE industrial Gas turbines have both air filter DP and compressor inlet DP installed. I can understand we can see the increase of air filter DP if air filter is fouling. Can we observe the increase of compressor inlet DP as well for air filter fouling?
 
>The compressor inlet DP sensor is used
>as part of the airflow calculation, not
>for filter fouling indication.

Does that mean you may see the increase of filter DP for both air filter and compressor inlet filter if air filter is fouling?
 
If there is filter fouling it will increase the compressor inlet DP reading. Note, there may be 2 different sensors on the inlet, the 2nd one being called "compressor bellmouth DP" - I don't remember the actual difference between the two. You would have to look at the piping schematic to determine what DP each sensor actually measures.
As far as the air flow calculation, I don't remember the equation. You might be able to get it from Toolbox. If not, GE considers it proprietary.
 
Have you looked at where the two "pressure" taps for the compressor inlet DP sensor are in the inlet?

In my estimation, if the high- and low-pressure taps for the compressor inlet DP sensor are DOWNSTREAM of the inlet filters, then the the compressor inlet DP sensor will not indicate dirty inlet air filters.

In other words, if the differential pressure being measured by the compressor inlet DP sensor does not include ambient pressure (UPSTREAM of the inlet filters) then it won't measure dirty filters.
 
Cris,

As mentioned by previous comments, operation with high differential pressure on the inlet filters may lead to performance loss, inlet duct collapse, admission of dust and foreign objects. I would add to these bent filters, high vibrations, danger of compressor stalling.

Normally, you have 2 mechanisms to prevent this to happen, a pulse cleaning system that is producing shock waves by rapid release of high pressure air and droplet catchers in front of the filters.

The droplet catchers are very important for the intake system. They are designed to condensate the humidity in the air before it reaches the filters. Unfortunately they may also be a cause of the problem. If improperly designed, the droplet catchers may become clogged with ice and reduce the amount of air. In one of the sites where I went, the air humidity is very high during winter and the ambient temperature is very close to the dew point. Hence, ice formation occurs when water droplets are in contact with any metal surface or already formed ice.

As a conclusion, you need to make sure your filter cleaning system is working properly, the droplet catchers are clean and in good condition and, in case ice tends to form on them, install an anti-icing system in front of them.

Also, as CSA suggested, check if you have implosion doors/flaps and make sure they are in good condition. The site I mentioned before doesn't have them and we were very close to a nasty disaster 2 winters ago.

One more thing, it is important to take care of your filters as well as possible. Not only that in case of high DP you might damage your compressor/turbine, but neglecting them is reducing their life-time (e.g. wet filters tend to collect dust that can not be blown by the pulse system), they are quite expensive and take a lot of effort to dismantle/replace.
 
E

Emmanuel Iyoho

New air inlet filters were installed and when the unit was started. the air inlet dp jumped to 4.7" H2O on startup. It increased to a generator trip point of about 8.9" H2O. We have since then replaced the filters with brand new ones but are still having an initial differential pressure of 4.7" H2O.
 
There are 2 things you should check to make sure your DP reading is correct. First, that the low-pressure tapping point (downstream, after the filters) is clear. The pressure is measured sometimes through a piece of pipe welded inside, with a vertical hole. If this is the case, make sure there are no deposits there. You can blow the pipe from outside to make sure it is clean.

Second, check the calibration of the DP transmitter. It might be that your actual DP is good but the transmitter is not working properly. A consistent drift in the span calibration, for example, might be a good explanation for the huge jump you had.

There is another probable cause for your problem, the pre-filters are not clean. Can you tell us what kind of filter-house you have (producer, model)?

And, of course, there is the problem of using the appropriate type of filters for replacement, but I presume you ordered them according to the OEM documentation.

Just as an example (Frame 9e, Donaldson filter-house, new filters, DP transmitter re-calibrated, everything cleaned, pre-filters changed): DP ~0.1"H2O on turning gear, goes up with the speed to 1.34 at FSNL, decreases after synchronization, goes no higher than 1.4 at Base Load.
 
Top