by Paul Sennett

There are several ways to maximize the performance of gas turbines — including ensuring that the air filtration system is optimized for the local environment. Paul Sennett suggests that care and effort taken to specify tailored filtration systems is well worthwhile.

The fluctuating cost of fuel over the last 12 months has put the spotlight clearly on saving energy wherever possible. But equally the scarce availability of power throughout the world has also highlighted that any improvement in efficiency will have an immediate impact on improving the gap between demand and supply. In a gas turbine power plant (whether a large power station or an on-site facility) there are many ways of improving the output or reducing the fuel consumption, but one which is often overlooked is to improve the combustion air intake system through better design or better filtration.


There are numerous components in a gas turbine power plant which affect fuel consumption and plant efficiency. Despite its relative simplicity, the air intake can have a direct and significant effect on the efficiency of the turbine itself.

‘Pulse jet’ filters are designed to displace the outer coating of dust. Photos courtesy of SPX
Click here to enlarge image

There are many different types of air filter and filtration systems. Rather than explain the many variations between filters, one can start by looking at a simple filter design in order to understand the main selection criteria.

In its simplest form, a filter acts as a selective barrier for the combustion air. It allows air and dust particles of a certain size through, and stops particles which are larger. Various filter grades are available, which progressively allow less and less contamination through. However, (again expressed very simplistically) as the filters become more efficient and less dust gets through, it is more difficult for the air to penetrate the filter. This resistance to the air flow results in a pressure drop and it is this which directly impacts the performance of the turbine. The higher the pressure drop, the higher the fuel consumption (for the same power output), or the lower the power output (for the same fuel consumption). In fact a pressure drop reduction of 50 Pa can result in approximately 0.1% improvement in machine power output. Not much in the grand scheme of things, but effectively something for nothing.

Clearly, to take this argument to its conclusion, no filter at all will produce the lowest pressure drop, which will provide the highest turbine efficiency. But this will only be true for a short period before the turbine requires a major and unnecessary overhaul.

So it’s clearly a balance between filter efficiency (how much dirt it lets through) and pressure drop (how much resistance is offered to the air flow). The skill of the filter designer is to make an air filter which utilizes the best, most suited media (and so is as efficient as possible), whilst minimizing extra pressure drop caused by the other components.

Taking a cartridge (so-called ‘pulse jet’) air filter as an example, the necessary structural components such as the inner and outer protection grids will have an effect on pressure drop, as will the media pleat sharpness. Multiply these apparently minor features by the many hundreds of filters which may be in place even in a modest sized gas turbine power plant and you can see that the overall effect may be significant. This then is the filter manufacturer’s contribution to your power plant’s efficiency.


Often a pre-filter will sit in front of a final filter, protecting the more expensive item. Clearly this adds to the system’s pressure drop, but helps by removing the large particles before they reach and contaminate the more expensive final stage.

However, the pre-filter is sometimes an unnecessary stage — highlighting that every location is different. In particular, pulse jet air filters are often protected by wrap-around pre-filters. This extra layer can be essential in heavily industrial areas, ensuring the hydrocarbon contamination is collected at an early stage and does not reach — and restrict the porosity of — the final filter. They can often be washed and therefore reused.

But they can also have a negative effect on the final filter. The pulsing action, which displaces the outer coating of dust, is reduced when the filter is covered in a blanket. The dust can get trapped between the pre-filter and final filter and increase the pressure drop unnecessarily.

Eliminating the pre-filter may not be desirable though. However, some power plants have found that they can change the layout of the filter house and incorporate pre-filters earlier in the housing, rather than needing to wrap them around the filter itself.

Figure 1. There are many different ways to capture particles in a filter
Click here to enlarge image

This extra stage may sound cumbersome, but can reduce pressure drop initially and, best of all, can ensure that the pulse jet filters continue to operate at a lower pressure drop for longer.


Although the cost of retrofitting the filter house may sound unattractive, again the benefit will be in improved pressure drop, which will increasingly show a payback if fuel costs continue to escalate. Careful design at this stage too can have an impact on pressure drop — as well as many other aspects of performance.

Changing the design of the filter housing, particularly where the local environment has changed, may bring a reduction in pressure drop
Click here to enlarge image

As mentioned before, filtration is always something of a compromise. A filter house redesign is another opportunity to build on that compromise — a complete change from static filtration to pulse jet filtration is often considered (and vice versa). This recognises that the environment may have changed since initial installation and so there may be new benefits from different filtration, which were not anticipated at the original build.


Air filter manufacturers have a number of tools to improve the filtration characteristics and to deliver better performance at lower pressure drop. As a simple explanation of a filter design, we need to understand the basic principles of air filtration. There are many different ways to capture particles in a filter.

The sieve effect is the commonly understood process, where the dust particle is caught between fibres, where the gap between the fibres is smaller than the particle itself. This technique is a common method of dust collection in filters and is certainly the easiest to visualize and understand.

However, there are a number of other techniques, which when the filter media is changed can become more or less effective. A couple of simple examples are electrostatic effect and diffusion effect.

Inevitably small fibres contain some electrostatic properties. The manufacturing process either generates this naturally (friction as the fibre is drawn) or it is artificially added. The charge on a fibre will attract the opposite charge on a particle, which might otherwise pass by the fibre. This effect can be useful in attracting small particles, which would normally travel through the filter and into the turbine.

Some particles will not be travelling with the air flow. Instead they follow a more random path driven by Brownian motion (molecules bombarding them from different directions and causing them to change course). Eventually they collide with a fibre and are held there by intermolecular forces.

There are other principles of filter operation as well, which really goes to highlight the complexity of the science behind filtration. For that reason, the purchase of air filters for a power plant (even a small power plant) should never be considered like a purchase from a catalogue. Not only do these relatively inexpensive devices protect a major asset, but any changes can sometimes have unforeseen consequences. Therefore it is very clear that the best course of action is either to understand the filter science at a deep level or to take the advice of the filter manufacturers by undertaking a true audit of the power plant.


If you have a power plant which is more than ten years old, you should call in a filtration expert to analyze your current requirements. If the local area has changed at all during this time, the filtration chosen at the beginning of its life may not offer the best performance today.

The two scenarios on page 38 highlight how simple changes can make a big difference to the economics of the power plant. These changes are more common as process plants change by adding or decommissioning certain parts and therefore have an immediate effect on the surrounding levels/types of pollution.

The best method of establishing the most suitable filtration at a point in time is to conduct a detailed audit and rely on the expertise of the filtration people. Many gas turbines have been shipped over the years with a ‘standard’ filter package — just as the filtration requirements between Siberia and Saudi Arabia are different, so it can be true that the filter requirements for two turbines sitting on either side of a steel mill can also be different.

A good air filter manufacturer will offer a consultative approach to filter selection. This involves the acquisition of local data, a realistic understanding of what the filters will do and then an understanding of the turbine operator’s objectives from his filter selection. This personalized recommendation then offers the operator the very best chance of meeting not only his pressure drop (fuel) requirements, but also filter lifetime, filter burst performance, filter efficiency, filter cost, and so on.

Figure 2. A consultative approach to filter selection will bring improved understanding to the real needs of the power plant
Click here to enlarge image

And this should not be a one time activity. Over the life of the power plant, things will certainly change in the local environment. Localized construction will not only effect the atmospheric pollution during the building, but also after construction the air contamination will be different from before.


Another good reason to accept the consultative intelligence of the filter-maker is to understand what new technology may help to improve fuel efficiency. New filters are being designed almost every day. An improvement in the balance between filter efficiency and pressure drop is the goal of most filter manufacturers. Major steps forward in media design are helping to produce filters which get ever closer to this target. A step change in performance is unlikely, but gradually the filter-makers hope to get closer to the turbine operators’ needs. But as the cost of fuel increases, the filter-maker will always have to stretch further to help out.

So next time your power plant requires new air filters, ask a filter manufacturer to recommend a new filtration solution. Your new filters might bring you the benefits of improved fuel consumption or improved output, but also, you may well help to reduce your power plant’s costs. Above all, a new filter solution could bring you more output for no extra cost, something for nothing, which is an offer you rarely hear these days.

Paul Sennett is with Dollinger Power Systems, part of SPX Dehydration & Filtration, Ocala, Florida, US.

Scenario 1

Imagine an example where a power plant was built at the beginning of an industrial development. The air intake filtration was initially selected to cope with the high levels of dust expected in this dry and arid location. However, over time the area grew with far more heavy industry arriving in the locality. The resultant hydrocarbons emitted from other process plants in the area, started to stick to the filter media and consequently so did the dust. The filter’s pressure drop increased significantly as the dust could not release from the surface and the turbine’s efficiency decreased.

A relatively simple change of filter (utilizing a different filter media) at the next outage ensured that the hydrocarbon contamination had a much reduced effect on the filter media, the dust did not stick as much to the media. Additionally an early stage, washable pre-filter kept the sticky hydrocarbons away from the final filter. The pressure drop took longer to increase and the power plant saw a reduction in fuel consumption for the same output.

Scenario 2

Imagine another scenario where an on-site power production facility at a metal processing plant has an air intake which initially faces open fields. But the next development phase of the process plant will result in the open fields being built upon. The air intake now consumes air from the production facility itself. The contamination can include metal fines, processing additives, smoke, steam, hydrocarbon exhaust and many other contaminants which were never anticipated in the first stage.

The result is that filters, which should take 12 months between changes, are now completely exhausted within only a few months. This not only brings the extra expense of new filters, but also the inconvenience of unscheduled maintenance to arrange for them to be changed. Although the power plant is relatively small, moving it would be out of the question, so the operator experiments with different filters. The balance of pressure drop, filter efficiency and suitability for the installed environment is not a simple balance and after two to three attempts the operator finally goes to a filter manufacturer for advice.

In this instance the solution is not simply a change of filter, but rather a change of the filtration type and the air intake — moving from standard static filters to filters which could self-clean and remove some of the contamination regularly by themselves.