CFD designs on ATEX
16 Jul 2004
Gas turbine installations commonly involve the housing of the gas generator unit within a confined acoustic enclosure.
As well as being designed to attenuate noise levels, the enclosure also provides ventilation for controlling convective heat rejection and diluting gas fuel leaks.
The potential for fuel leakage means that such enclosures are covered by the ATEX Directive and the Health and Safety Executive's Guidance Note PM84 on the control of safety risks for gas turbine installations.
Although it is not a specific legislative requirement to comply with these guidelines, they provide plant operators and local HSE inspectors with a framework for assessing whether the operation of a plant is satisfactory from a general health and safety standpoint. It is therefore appropriate for installations to adhere to the guidelines as far as is reasonably practicable.
In the detection of gas leaks it is important to balance the ventilation system so that leaks can be diluted and removed from the enclosure, preventing build up to explosive levels, but that dilution is not so great that the true leak magnitude is masked at the detection point.
The philosophy is that leaks that can't be detected are diluted, and those that can't be diluted are detected.
In order to establish the thermal distribution within the enclosure airspace and potential gas fuel leak distributions and concentrations, it is important to be able to examine the through-flow inside the enclosure.
In situ testing within gas turbine enclosures (anemometer and smoke bomb tests, for example) does provide a means of providing overall flow patterns inside the enclosure. However, the results are either qualitative, cold-flow based or provide only highly localised information.
This is mainly because personnel access is highly restricted, for safety reasons, when the engine is running. In any instance, however, gas cloud sizes cannot be easily quantified by in situ testing for the purpose of comparing against the criterion in PM84.
Computational fluid dynamics (CFD), on the other hand, has the capability to model gas leaks in enclosures. In fact, PM84 states that CFD is the best tool for predicting ventilation and gas leakage in larger enclosures. This is because of its capacity to allow quantitative assessment against the criterion, and because it enables the modelling, analysis and understanding of design impact before changes are made, or even before a plant is built.
This ability of CFD was put to the test in a recent case study, completed by Fluent Europe, to investigate the conformance to the revised PM84 guidelines at a particular power plant site in the UK.
This was a consequence of changes in the ventilation system occurring after 1 July 2003.
Firstly, a representative model of the inside of the enclosure was built using Fluent's CFD preprocessor GAMBIT. The gas generator, intake, diffuser, pipework and enclosure walls are all represented realistically in order to provide as accurate a solution as possible.
A series of outlet ventilation ports are positioned on the roof incorporating extraction fans that establish a negative gauge pressure inside the enclosure. The supply is from a number of inlet vents positioned strategically on the side walls.
Notice the complexity of pipework around the engine that could restrict ventilation penetration into the gas fuel manifold area. A mesh - of around 3.1 million cells - was applied to the numerical domain as part of the CFD calculation.
The level of heat rejected from the engine casing was calculated from a heat balance across the engine and applied realistically to the casing surface on the model.
The commercially available CFD software, FLUENT, was used to perform the calculations and, in conjunction with site measurements of internal pressure and temperature, was used initially to establish confidence in a baseline operating case prior to leak introduction.
Key flow features such as areas of flow stagnation, depth of penetration from the ventilation inlets and temperature distributions around the engine were identified. The variation in average to peak static temperature variations at the exit planes was also identified.
For the safety case, a range of leak source flow rates, directions and locations were modelled in an attempt to create the largest leak cloud possible. The leak flow rates were modelled based on the detector alarm threshold levels in the 3-10% LEL (lower explosive limit) range. A more sensitive setting for the exit sensor would obviously result in an earlier detection and, consequently, a smaller gas cloud at the point of detection.
The leaks were modelled one at a time, with the sources located on the underside of the turbine region where the gas fuel rings are positioned, the distance to the exit vents is largest, the air temperature is high and the diluting ventilation flow has restricted access. Additionally, predicted regions of relatively high ventilation air residence time and low local air velocity were used to assist in placing the leak sources in their worst possible position.
A two-species mixture model was used to examine diffusion of the gases and a subroutine was written to calculate the sizes of the gas clouds.
The sizes of the steady state 50% LEL gas clouds for the worst leak case considered were predicted at the 3, 5 and 10% LEL detection levels. These were determined to be 0.083% and 1.61% of the net enclosure volume for the first and last cases respectively.
The pictures (below) show how the sizes that the 50% LEL gas clouds would reach before they were detected, depending on the exit detection levels for this worst considered leak case.
In this instance, it was decided to install detection capable of reaching the 3% LEL level, rather than using a less sensitive limit coupled with localised dilution or ventilation system modifications.
Detection at the 3% LEL level ensured compliance with the PM84 guidelines in that the cloud size was less than 0.1% of the net enclosure volume. An analysis of the fuel-air mixture at the exit plane was also conducted to determine the level of mixing and assist in determining a desirable pick-up point for the sensors.
This case study showed the importance of CFD for quantitative study of potential gas fuel leakages, enabling an informed decision to be made on compliance with the ATEX Directives now in force.
Mark Keating is Senior CFD Engineer with Fluent.
<b>The letter of the law</b>
The ATEX Directive (ATEX standing for Atmosphere Explosive) is one of a series of new measures introduced under article 100a of the Treaty of Rome. Unlike previous hazardous area directives, it is a 'New Approach Directive' and is thus mandatory.
Besides considering potentially explosive concentrations of gas, vapour or mist in the air, ATEX also now takes into account dust in the air. Furthermore, the sources of ignition covered have been expanded to include both electrical and mechanical sources. Any control equipment that is used to ensure the safe operation of equipment in a hazardous area also comes within its scope.
There are in fact two ATEX directives. ATEX-100A (94/9/EC) concerns equipment and protective systems for use in potentially explosive atmospheres, while ATEX-137A (1999/92/EC) concerns the safety and health of workers potentially at risk from exposure to explosive atmospheres. Both came fully into being on 1 March 1996, with a transitional period that ended on 30 June 2003.
From 1 July 2003, all new equipment within the scope of the directive must carry the CE mark and comply with the directive's requirements. Workplaces already in use before July 2003, but modified before July 2006, must meet requirements from the time the modification takes place. From 1 July 2006, all relevant workplaces must comply with the directives.
The 'UK Dangerous Substances and Explosive Atmospheres Regulations (DSEAR) 2002' and 'Equipment and Protective Systems for Use in Potentially Explosive Atmospheres Regulations' provide a UK legislative framework for the implementation of the EU ATEX Directives 137A and 100A, respectively.
HSE Guidance Note PM84, 'Control of Safety Risks for Gas Turbines used for Power Generation', is the most pertinent set of current guidelines covering the design and operation of gas turbine installations. It makes specific and practical recommendations for operational functions such as alarm and engine trip conditions, and the maximum acceptable volume of explosive mixtures of gas and air.
PM84 declares limits on the size of the 50% LEL (lower explosive limit) gas cloud before detection, as less than 0.1% of the net enclosure volume, where 100% LEL represents the fuel-air mixture just rich enough to be ignitable. In this way a degree of conservatism is built into the criterion. Additionally, PM84 requires that the sensor alarm threshold be set at 'ideally less than 5% and no more than 10% LEL'.
