- 5 December 2007 -

Perfect air filtering for gas turbines

Paul Sennett from SPX looks at how engineers should optimise the filtering of air for their gas turbines - through improved pressure drop, regular cleaning and maintenance, and by taking a consultative approach - ensuring that all environmental factors are taken into account before a filter is specified.

In 1791 an Englishman called John Barber filed a patent for the first gas turbine. Subsequent designs over the next 150 years finally resulted in an engine that could produce far more energy than it consumed. But it's really the last 40 years, with the availability of relatively low cost natural gas and oil, that has seen the true transformation of the technology, with the result that much of the world's power today is generated by gas turbines.

Gas turbines burn fuel with enormous volumes of air to produce energy via a generator. Of course the air is simply pulled from the atmosphere and so needs to be cleaned to prevent damage to the turbine from atmospheric pollutants. Although water and dust (of natural or anthropogenic origin) are present everywhere, the quantity of these contaminants can vary dramatically in a location throughout the same region, let alone in different parts of the world. Therefore choosing air filters to clean the air is not a simple, consistent job, but a complex activity involving a number of considerations and experience.

To compound these issues further, gas turbines may have a working life of 30 years or more. The local environment may very well change during that period - even change two or three times. Local construction, mineral extraction, agriculture or weather patterns can all potentially affect the level of pollution present.

Perfect air filtration

Clearly the need for effective air filtration to protect the turbine is important. But protection is not the only consideration for a turbine operator. Perfect air filtration would bring an ideal state to each of the following:

. Pressure drop over the filters - the lower the pressure drop the higher the output from the turbine (or the lower the fuel consumed for the same output);

. Filtration efficiency - the higher the efficiency the cleaner the air and therefore the lower the damage and wear that occurs on the critical turbine components like blades;

. Maximum burst pressure - to ensure the highest factor of safety, the final filter should be tested to destruction, especially in dust loaded, high humidity conditions - this worst case scenario tells the operator what level of safety he has should things go wrong.

Improved output

Everybody wants to get something for nothing, and with gas turbine air filter choice this is often quite possible. The pressure drop of the filter combination has a direct correlation with output from the turbine or fuel consumed. Put simply, if the pressure drop is reduced, turbine output increases with no increase in fuel consumption.

A pressure drop reduction of 50 Pa will result in approximately 0.1% improvement in machine power output - effectively an increase of 450kW of output on a 450 MW turbine - for free.

Changing to better designed filters, which offer lower pressure drop, not only saves money, but increasingly is being seen as one of the factors to help save the planet.

However, reducing pressure drop has to be achieved all the while keeping the filters' efficiency constant; otherwise the filter lets more dust through to the turbine and results in greater damage to the turbine components. There is no point in improving output if it has been achieved at the cost of greater maintenance. A knowledgeable filter supplier knows how to balance filter design and media selection to get the highest efficiency with the lowest pressure drop.

Reducing pressure drop to save money is one thing, but any reduction in the quantity of fossil fuel burnt also has a direct impact on the carbon dioxide emissions from a power station. Changing to better designed filters, which offer lower pressure drop, not only saves money, but increasingly is being seen as one of the factors to help save the planet. Considering that a combined cycle power station might be producing 400 g of CO2 per kWh of electricity produced, it is quite clear why improved turbine power output (and therefore a reduction in fossil fuel consumption) will help to clean up the atmosphere.

Making it cleaner

Poor, low efficiency air filtration results in damage to critical turbine components, but often this damage can be reduced by regular, on-line, high pressure washing. But power washing the turbine blades can drive contamination into the second and third stages, resulting in the need to clean these stages with the turbine off-line.

Increasingly, this is causing power companies difficulties as this maintenance can interfere with the flexibility of the turbine's output. An example might be an unexpected high demand, where spot electricity prices are high, but the turbine is unavailable to take advantage of this due to a wash or an unplanned outage to clean the later stages. Preventing the turbine from getting dirty in the first place (with better, higher efficiency filtration) ensures that the turbine is completely flexible and available for those high profitability opportunities.

To put this into context, a filtration system in a gas turbine with standard F7 (EN779) final filters will allow 15,000 g of dust through to the turbine in one year. Using the same (consistent) example, but with higher efficiency H11 (EN1822) filters, the quantity of dust reaching the turbine will reduce to only 27 g per year. The need to wash is virtually eradicated and the power station is free to take maximum financial benefit from its flexibility.

This benefit is compounded further when you consider the reduction in turbine performance due to fouling of the turbine blades. A 4-5% reduction in turbine efficiency is not uncommon before a wash takes place. Changing to an improved filtration system may eliminate all of this fouling and therefore produce a turbine with more constant output.

However, higher filter efficiencies will inevitably mean higher pressure drops and therefore lower output or higher fuel consumption. That's the decision facing the turbine operator and perfectly highlights the essential need for considered and calculated filtration, ensuring that the filters not only meet the local environment, but also the commercial requirements of each individual power plant.

Preventing a 'burst'

Without a doubt, the worst situation for a filter is for the pressure to get high enough that the media can no longer sustain it and for the filter media to tear - a burst. All of the contamination held within the filter is then deposited straight into the turbine.

High efficiency final filter after a “burst test” to determine the maximum sustainable pressure. Picture © SPX Dehydration & Filtration.

Clearly with planned maintenance this should never happen as the filters would be changed before they reach such a critical state. But it does happen, too often.

By-pass doors will generally open if the pressure in the filter housing becomes too high. But if the filter bursts before this pressure, then the operator is left with zero safety factor.

Good performance filters will burst at pressures of 2-3 times the by-pass door operating pressure. Anything less (especially a burst pressure of less than the by-pass door) is simply risking the integrity of the expensive turbine.

Higher filter efficiencies will inevitably mean higher pressure drops and therefore lower output or higher fuel consumption. That's the decision facing the turbine operator and perfectly highlights the essential need for considered and calculated filtration.

Turbine manufacturers are also considering the effect of high pressures under high humidity. Tropical weather can influence the performance of air filters and this too needs to be considered (and modelled) in order to build the factor of safety for the turbine operator. This is especially true when unexpected increases in dirt loading take place in short time frames - the effect on the pressure drop can be a sudden increase. Ensuring that filters can cope with this is the only way of ensuring that factor of safety.

A consultative approach

Summing all of this together highlights the complexity of choosing the perfect filters for a gas turbine application. For that reason, buying air filters should not be considered a 'catalogue' purchase.

Understanding the balance that the turbine operator wants to place on each aspect, means that the air filter manufacturer needs to provide a consultative approach to filter selection. The best filter companies will ensure that they acquire data from the local environment, match this to their detailed knowledge of the long term performance of their filters and then use a computer modelling program to establish how long a certain filter combination will last.

This should not be static. Over the life of the power station, the local environment is almost certain to change. Construction of housing, commercial buildings and roads are classic changes which affect the levels of pollution in the air. These changes will have an effect on filtration performance and so a regular review of this (as well as a review of the objectives of the power station) is essential to ensuring optimum performance throughout the turbine's life.

Meeting the specification

The easiest way to meet the challenges and complexities of all these variances is simply to meet the specifications or standards in the filter industry for power generation applications. Unfortunately no such standardisation exists.

Help is on the horizon though. A European standard for the use of air filters in gas turbines is now in the early stages of being drafted. This will set a benchmark specific to gas turbine filters, which will help operators to make decisions and at least demonstrate the difference between high performance and under performing filters.

Stepping up to the challenge

So with these various levels of complexity, it is not that surprising that the power companies and turbine or compressor operators are turning to the filtration industry for support. Each time the filters need changing the discerning operator should be asking: "Has the environment changed during the time that the filters have been in place?", "Do I need my turbine to be more flexible and so will I benefit from higher efficiency filtration?", "Am I confident that the filters I have will cope with extreme pressures if an unforeseen incident happens and contaminates the filters prematurely?". The answers to these will determine whether the operator should use the consultative capabilities of the filter manufacturer. If the filter manufacturer cannot answer these questions, then it's probably time to look for a new filter supplier.

Contact
Paul Sennett
Paul.Sennett@airfiltration.spx.com
www.spxdehydration.com