air filter calculation.pdf

6
Publication Series Viledon Filter Concepts for Gas Turbines – Overview and Field Report on Utility Value Enhancement with Three-stage Filtration Dr. Heiko Manstein Dipl. Ing. Andreas Rothmann Published in: VGB PowerTech 12/2009, pages 78 - 82 Paper presented at: VGB Conference "Gas Turbines and Operation of Gas Turbines 2009" June 24 - 25, 2009 in Mannheim

Transcript of air filter calculation.pdf

Page 1: air filter calculation.pdf

Publication Series Viledon

Filter Concepts for Gas Turbines –

Overview and Field Report

on Utility Value Enhancement

with Three-stage Filtration

Dr. Heiko MansteinDipl. Ing. Andreas Rothmann

Published in:VGB PowerTech 12/2009, pages 78 - 82

Paper presented at: VGB Conference "Gas Turbines and Operation of Gas Turbines 2009" June 24 - 25, 2009 in Mannheim

SR_VGB_engl_12-09 09.12.2009 11.15 Uhr Seite 1

Page 2: air filter calculation.pdf

Filter Concepts for Gas Turbines

2 VGB PowerTech 12/2009

Filter Concepts for Gas Turbines – Overview and Field Report on Utility Value Enhancement with Three-stage FiltrationHeiko Manstein and Andreas Rothmann

Authors

Dr. Heiko MansteinApplications Engineering Turbomachinery

Dipl.-Ing. Andreas RothmannHead of the Market Segment TurboMachinery Germany/Scandinavia Freudenberg Filtration Technologies KG Weinheim/Germany

Kurzfassung

Filterkonzepte für Gasturbinen – Übersicht und Praxisbericht

zur Nutzwerterhöhung mit dreistufiger Filtration

Effiziente Luftfiltration und ihr Beitrag zum wirt-schaftlichen Betrieb von Gasturbinen bieten ein breites Feld für stetige Weiterentwicklung. Der Nutzwert dreistufiger, hocheffizienter Fil-tersysteme für Gasturbinen-Ansaugsysteme wurde in einer Vorgängerveröffentlichung mit grundlegenden Betrachtungen vorgestellt.

Erfahrungen aus implementierten mehrstufigen Systemen belegen die Anwendbarkeit auf die derzeit gängigen statisch betriebenen Filter-systeme vor dem Hintergrund der Forderun-gen nach Partikelabscheidung, Koaleszenzver-mögen und Druckverlustverhalten. Zwar be-dingen mehrstufige Systeme den Nachteil eines anfänglich höheren Druckverlusts, er-möglichen aber durch fast vollständig vermie-dene Verschmutzung und erhöhte Anlagenver-fügbarkeit einen deutlichen, belegbaren Kos-tenvorteil.

Anhand einer beispielhaften Vergleichsrech-nung zum Kostennachteil eines Betriebssys-tems mit zyklischen Waschvorgängen und den Betriebskosten eines mehrstufigen Filtersys-tems wird der tatsächlich erreichbare Kosten-vorteil abgeschätzt und erörtert. Die Berech-nungen stehen in gutem Einklang mit den Er-fahrungen aus existierenden Anwendungsein-sätzen, die mit zwei Praxisbeispielen belegt werden. Bisher ungenutzte Leistungspotenzia-le können durch die Vermeidung von Ver-schmutzung zuverlässig erschlossen, Anlagen-verfügbarkeiten deutlich erhöht und zusätz- liche Betriebskosten verringert werden. Auch der Verzicht auf das Waschsystem erscheint in der Gesamtbetrachtung als weiterer Entwick-lungsschritt denkbar.

Introduction

Responsible deployment of the fossil resources at our disposal demands their maximally effi-cient and eco-neutral use while simultaneously factoring in their cost-efficient utilisation. To reconcile both these aspects is the paramount goal governing the development efforts of tur-bine manufacturers, system development engi-neers and system operators alike.

From a system operator’s viewpoint, it is pri-marily downtimes that are rated as lost profit, and must therefore be avoided. High levels of availability and long running times for the systems concerned are the declared goals. In particular, downtimes required for washing the compressor stages inevitably cause non-productive periods, which need to be mini-mised or altogether avoided. The aim of the washing routines is to remove any coatings and deposits on the blading. The cause for any such deposits will usually be inadequate in-take air filtration. Innovative concepts for en-hancing filtration quality by means of three-stage filter systems were presented in [1].

This paper provides an overview of filter sys-tems in current use and an approach for esti-mating the financial benefits accruing from system modification, an approach which is illustrated by examples from actual operation-al practice.

Principles of Particle Filtration

The demand for efficient air filtration for in-ternal combustion engines entails sophisticat-ed challenges in terms of adequate design for and implementation of the air intake systems being used.

An air filter system is required to significantly reduce the penetration of solid and liquid par-ticles into the turbomachinery, while coping with temporally fluctuating environmental conditions.

Capturing of air-borne particles (which may be dust particles or droplets) depends most particularly on the transport mechanisms ef-fective at the location concerned. Electrostatic, diffusion-, inertia-, and gravity-related effects are responsible for particle transport to the fil-

ter media, which are usually made of fibres. The adhesion forces operating between parti-cles and fibres in their turn are determined by the interaction of Van der Waals forces, and electrostatic and liquid-related effects, and en-able dirt particles to be permanently arrested. When developing filter elements, then, both these mechanisms, the transport and the adhe-sion mechanism, must be given due considera-tion in the optimisation process.

The concentration of air-borne dust particles is of crucial importance for the design of in-take air filtration systems. Over the past few years, measurements of dust concentrations have been continuously expanded, so now there is a broad data base available [2]. The temporal fluctuation band of the PM10 dust mass found here ranged from approximately 5 to 40 µg/m3 [3] in Germany during 2007, and depends largely on the season of the year, on the surrounding landscape and the degree of industrialisation obtaining at the place where the measurements are taken. In this context, PM10 denotes the dust fraction whose mean particle size (aerodynamic equivalent diame-ter) is 10 µm, with 50 % of it being arrested. This dust fraction exhibits a particle size dis-tribution that may well contain particles of up to 20 to 30 µm in diameter [4].

The fact that a very high and nonetheless lim-ited proportion of the dust fraction is being retained also means that there will always be particles penetrating the filter. As a conse-quence of particles passing through a filter stage, deposits are formed on the blading, which results in output losses at the gas tur-bine. In regions close to the coast, additional corrosion effects may be encountered, due to air-borne salt particles. The aim of develop-ment work on filter systems is accordingly to minimise precisely that proportion of the dust fraction which passes through the filter. Fil-tration quality is rated in terms of collection efficiencies for individual particle sizes or for the entire dust quantity in question [5].

Filtration Concepts for GT Applications

The temporal dust mass carried in determines the choice of a suitable intake air filter sys-

Page 3: air filter calculation.pdf

Filter Concepts for Gas Turbines

VGB PowerTech 12/2009 3

tem. The size of the particles of relevance for intake air filtration is typically to be found in a bandwidth of around 0.01 µm to about 3 mm, and at locations exposed to high indus-trial emissions an average mass concentration of up to 200 µg/m3 can occur [6].

It is only in a few regions (exposed to tempo-rarily extra-high dust concentrations) that re-generatively operated systems are actually necessary, which, following the principle of surface filtration, form a compact dust cake on the filter medium involved. Using the pulse-jet cleaning method, the dust cake can subsequently be shaken off the filter medium at pre-defined intervals. Filter systems of this kind are mostly in single-stage design and re-quire precisely harmonised and properly func-tioning cleaning systems. The drawback with these is a relatively high degree of particle penetration during and shortly after the clean-ing phase, since it is precisely then that the filtration-supporting effect of the dust cake is absent; it will only be built up again over the course of the cycle now commencing.

In most of the Earth’s regions, low to moder-ate dust masses are encountered, which can be very successfully stored and lastingly retained in the element’s filter medium using the prin-ciple of deep-bed filtration. If the dust-reten-tion capacity is exhausted, the filters con-cerned are replaced by new, non-loaded ele-ments during the system’s overhaul and stand-still times. The salient features of static filtration systems are these: arrestance of large particles in a pre-filter stage and storage of small particles in the fine-filter stage. Stat-ic air filter systems of this kind can be sup-plied in multi-stage design, which offers scope for optimizing the filter technology involved.

The task of a state-of-the-art design for intake air filters is to affordably reconcile the para-mount requirements posed for optimum sys-tem operation:

Maximised system protection = filtration –with maximum efficiency and at a consist-ently high level

High system availability = downtimes for –replacing the filters are rare and short

High system efficiency = low pressure drop –in the filter system

Reduction in unplanned downtimes of the –filter system = failsafe product quality

One feature inherent in all filtration is the phenomenon that initially high efficiency lev-els and the high dust storage volumes achieved during operation entail an initially high and steadily increasing pressure drop during the filter’s useful lifetime, and that is deleterious for optimum gas-turbine operation.

Optimisation work on intake air systems, fac-toring in some basic considerations regarding the comparison of two- and three-stage filtra-tion systems, has been extensively dealt with in a previous publication [1].

The Keystones of a Comparison of Two-stage and Three-stage Filtration

F i g u r e 1 illustrates the basic set-up of a two-stage filter system, in which the final cas-sette filter stage is protected by an upstream pocket filter stage. The core for top-quality filtration of a three-stage filter system as de-picted in F i g u r e 2 is the high-efficiency particulate air (HEPA) filter (Filter Class H11 as defined in EN1822) installed in the third filter stage. The upstream filter stages serve primarily to protect the HEPA filter in the fi-nal filter stage. The task of pre-filtration is frequently handled by a pocket filter of Filter Class F6 (as defined in EN 779), which in its turn is installed upstream of the intermediate stage comprising F8 or F9 cassette filters. Which filters to choose for these two stages

depends crucially on the environmental con-ditions concerned and the building constraints encountered.

A reduction in the number of pre-filter stages or of the Filter Class will directly affect the dust volume passing through the filter stages concerned, and thus the useful lifetime achiev-able for the HEPA filter in the final filter stage. Moreover, possible effects exerted by the ambient air’s humidity in conjunction with the dust arrested in the filter stages must be given due consideration, particularly when swellable or sticky dust constituents are in-volved. These might be responsible for an un-foreseeable rise in the pressure drop as a con-sequence of weather-related factors. Another task to be performed by the upstream filters is to act as coalescence stages. Minute droplets of fog contained in the intake air condensate at the fine fibres of the pre-filters and coa-lesce. Gradually, these droplets reach a flow-able state, and driven by gravity flow down so that they may exit at the filter’s base. When the intake air is humidified by means of evap-oration coolers, the effects described are en-countered even more often, and must be given particular attention in the run-up to filter sys-tem dimensioning. In this case, it is absolutely imperative to use a two-stage pre-filter sys-tem, in which droplets are arrested to a suffi-cient degree in the first filter stage so as to prevent the final HEPA filter stage being soaked through. Tests are also currently ongo-ing with filter systems comprising merely two filter stages at locations where only low hu-midity levels can be anticipated. Here, F7 pocket filters are used in the pre-filter stage upstream of the high-efficiency particulate air (HEPA) cassette filters of Filter Class H11. So far, however, the application results of these tests are not yet available. Of course, protection of the gas turbine is here assured to the same degree as in systems featuring two pre-filter stages. The results on the useful life-time for the final HEPA filter are being ea-gerly awaited.

Figure 1. Two-stage filter system comprising pocket and cassette filters.

Figure 2. Three-stage filter system.

1.00

0.99

0.98

0.97

0.96

0.95

P/P

GT,

i

0 1000 2000 3000 4000 5000 6000 7000 8000 9000

Operating time in h

Figure 3. Output loss due to compressor fouling and output recovery through “washing” in a two-stage filter system.

Page 4: air filter calculation.pdf

Filter Concepts for Gas Turbines

4 VGB PowerTech 12/2009

The principal focus of the considerations de-scribed in the previous publication was on two parameters, exhibiting both advantages and disadvantages [1].

The advantage offered by upgraded filtration is less fouling on the compressor section of the gas turbine.

In this context, the values compiled in Ta b l e 1 illustrate the improvement in arrestance achieved by the three-stage HEPA filter sys-tem as compared to a conventional two-stage fine-filter system. When a HEPA filter sys-tem is used, the number of particles responsi-ble for compressor fouling in sizes ranging from 0.3 to 0.5 µm can be reduced by a fur-ther factor of about 30, and the somewhat larger fraction of 0.5 to 1.0 µm in size by a factor of roughly 200.

The result of the empirical feedback so far obtained from the markets confirms these ba-sic considerations. By using upgraded filtra-tion technology featuring HEPA filters, the fouling customarily encountered on the com-pressor blading is avoided almost entirely. Feedback likewise confirms that online and offline washing can be completely dispensed with, creating the concomitant benefit that there is no longer any reduction in perform-ance between one washing routine and the next (otherwise an accepted fact), which in turn shows up immediately as enhanced per-formance figures at the machine itself. What is more, non-productive times, reduced out-put levels and fouling entrainment as side-effects of the washing routines are a thing of the past.

In the theoretical deliberations of Schroth et al., a lower output loss of the three-stage sys-tem was taken into account, which is outlined

as an example in F i g u r e 3 . The output loss has not materialized in actual practice. There is no particle penetration worth mentioning. Deposits on the turbine blading are being pre-vented almost entirely.

F i g u r e 4 shows the endoscopic image of the compressor blading’s surface in a Taurus 65 GT. After having been in operation for ap-prox. 9,000 hours without any washing rou-tine at all, no deposits whatsoever adhering to the blading are visible. This is an unequivocal success for the three-stage HEPA filter system featuring the combination F6-F9-H11.

The disadvantage of a three-stage filter sys-tem is, quite naturally (due to the contribution made by the third high-efficiency filter stage), the higher pressure drop that must be antici-pated in the filter system as a whole.

In F i g u r e 5 , a pressure drop curve typical for a three-stage filter system is compared to that of a two-stage system. Here, the hatched area marks the difference and/or increase of the pressure drop over the entire operating pe-riod and thus provides a benchmark for the mechanical energy yield actually achieved.

As experience has shown, the output loss of a gas turbine due to a higher pressure drop in the intake system can be estimated at about 0.1 % efficiency loss for each 50 Pa of in-crease in pressure drop, and the direct effects of an additional third filter stage can be evalu-ated. In the analytical summary, the advantage of increased machine availability must at the

very least compensate for the deleterious con-tribution of a higher pressure drop in the sys-tem, or even exceed it.

Typical Computation for a System

With an empirically based computation pro-gram, it is possible to use the operator’s par-ticulars on the status of the existing system to compare it with the modified, three-stage fil-ter system, and to analyze the advantages and disadvantages involved.

The particulars required from the operator are the following (here reproduced as an example; Ta b l e 2 ).

Further particulars given for the existing filter system serve to provide a comprehensive overview of the application task involved and are factored into the computation if neces-sary.

Based on the operator’s particulars and the typical pressure-drop curves, on which the computation is based, the anticipated average pressure drops are estimated as a characteris-tic mean value over the entire period of opera-tion for the individual filter stages. The results have been compiled in Ta b l e 3 .

The operating values for the new filter system are now estimated within the framework of the loadings stipulated by the operator (see Ta b l e 4 ) and then compared with the values for the existing system.

Table 1. Comparison of arrestances achieved by two- and three-stage filter systems.

Particle size µm Particles in ambient air per m3

Initial efficiency of filtration

Particle penetration per m3

Initial efficiency of filtration

Particle penetration per m3

0.3-0.5 20,000,000 ≈ 64 % 7,200,000 ≈ 98.9 % 220,000

0.5-1.0 4,000,000 ≈ 80 % 800,000 ≈ 99.9 % 4,000

1.0-2.0 300,000 ≈ 95 % 15,000 ≈ 99.999 % 3

two-stage (F6 + F8) three-stage (F6 + F9 + H11)

Table 2. Typical particulars serving as the computation basis.

Maximum power output 31 MW

Operating hours per year 5,600 h/a

Approx. volume flow at 20 °C 231,126 m3/h

Number of filter elements per filter stage 63 pcs

Loss due to fouling in 40 days 700 KW

Volume flow loading per filter 3,700 m3/h per filter

Figure 4. Endoscopic image of the compres-sor blading in a Taurus 65 GT.

Table 3. Computations for the two-stage system used.

Δp_Start[Pa]

Δp_End[Pa]

Characteristic mean value for pressure drop

[Pa]

Current filter system:

1st stage Pocket filter F6 70 250 130

2nd stage Cassette filter F8 120 400 213

Total 190 650 343

Page 5: air filter calculation.pdf

Filter Concepts for Gas Turbines

VGB PowerTech 12/2009 5

In this comparison, the difference between the characteristic mean values produces an anticipated pressure-drop increase of 493 Pa – 343 Pa = 150 Pa for the three-stage filter system.

Output Loss Due to Fouling up to Turbine Washing in the

Previous Filter System

Figure 3 shows an output curve typical for gas-turbine operation with washing routines, in the shape of a classical saw tooth curve. The output loss caused by fouling during the load period can be partly recovered by the washing routine. In a rough approximation, the hatched area below the curve can be re-garded as a triangle and its area adduced as a dimension for the output loss involved.

In order to take into due account the perform-ance curve actually encountered at the ma-chine (which frequently varies), the conserva-tive estimate adduces only one third x H x W of the area below the curve for triangle com-putation, to be on the safe side. This means that the machine’s output loss with the exist-ing two-stage filter system is in the computa-tion given a lower value than that actually oc-curring. The sub-areas computed within the period under review are added together, and adduced for system comparison. For this pur-pose, particulars from the operator on the number of washing routines within the period of operation are required.

The gas turbine’s output loss stated in the ex-ample is 700 kW within 40 days of operation. The system’s overall operating period is 5,600 hours, inside which 5.8 (mathematically de-termined) washing routines occur, which have

to be taken duly into account. (Operating time between two washing routines = 960 h). As a total result of cyclically encountered fouling, the gas turbine’s output loss in the period un-der review is:

P_Loss due to cyclical fouling = 1/3 · 0.7 MW · 960 h · 5.8 = 1299 MWh

Output Loss Due to Higher Initial Pressure Drop in

the Three-stage System

To start with, the output loss due to a higher initial pressure drop in the three-stage filter system is estimated. This is obtained as:

Output loss = GT output · output loss per 50 Pa of pressure increase · pressure drop

P_GT_3-stage_i = P_GT_i · C_DP_50Pa · (DP_3-stage_i/50 Pa)

where:

P_GT_3-stage_i Initial output loss of gas turbine in three-stage filtra-tion system

P_GT_i Gas turbine’s rating

C_DP_50Pa Efficiency loss coefficient

DP_3-stage_i Initial pressure drop in the three-stage filter system

With the typical values, the gas turbine’s una-voidable output loss due to the higher initial pressure drop in the three-stage filter systems is obtained as:

P_GT_3-stage_i = 31 MW · 0.001 · (150 Pa/50 Pa) = 0.093 MW

The gas turbine’s output loss caused by the three-stage filter system within the period un-der review is calculated as:

P_Loss due to 3-stage filter system = P_GT_3-stage_i · operating period in hours

P_Loss due to 3-stage filter system = 0.093 MW · 5,600 h = 520.8 MWh

Table 4. Computations for the three-stage filter system planned.

Viledon Type Δp_Start[Pa]

Δp_End[Pa]

Characteristic mean value for pressure drop

[Pa]

1st stage Pocket filter F6 45 250 113

2nd stage Cassette filter F8 100 200 133

3rd stage HEPA filter H11 220 300 247

Total 365 750 493

800

700

600

500

400

300

100

200

0

∆P in

Pa

0 1000 2000 3000 4000 5000 6000 7000 8000 9000

Operating time in h

3-stage2-stage

Figure 5. Comparison of the energy yield between a two- and three-stage filter system.

Figure 6. Image after approximately 9,000 operating hours without compressor washing routine.

Figure 7. Photo of the compressor stage after two months with F8 filters after the latest offline washing.

Figure 8. Photo of the compressor stage after four months with H11 filters without offline washing.

Page 6: air filter calculation.pdf

Filter Concepts for Gas Turbines

6 VGB PowerTech 12/2009

Evaluation and Discussion of the Results

Due to cyclically occurring fouling, the gas turbine operated with the conventional two-stage filter system and regular washing rou-tines loses approx. 1,300 MWh within the period under review. Taking an assumed re-muneration for electricity of 65 €/MWh, this corresponds to a financial loss of around 84,900 €. This loss can be avoided by install-ing a three-stage filter system and profitably eliminated.

On the one hand, installation of a three-stage filter system increases the pressure drop, which in our example corresponds to an output loss for the gas turbine of approx. 520 MWh or a financial loss of around 33,800 €. On the other hand, this investment eliminates the two-stage system’s output loss, which was hitherto una-voidable.

Avoiding the costs caused by compressor fouling, of: = + 84,900 €

Accepting the additional costs incurred by a higher pressure drop: = – 33,800 € ––––––––––

Within the period under review, a total cost advantage of = + 51,100 € is thus obtained.

An estimate, plus a cost comparison, must be performed individually for each system in question. Note that further deleterious aspects have not yet been taken into account here, particularly the irreversible degradation of the gas turbine due to fouling. In spite of regular washing routines, the original performance level is impossible to reach again. With three-stage filtration, this effect is almost entirely avoided. Nor have the costs for cleaning agents and their disposal been factored into this computation. Effects improving the over-all result, such as its no longer being neces-sary to add to the running time equivalent op-erating hours for the start-up procedures re-quired after washing, and the increase in the gas turbine’s availability levels, have likewise not yet been factored in.

Typical Examples

The innovative filter concepts with upgraded filtration quality have been undergoing appli-cations-engineering trials since 2003, and cur-rently comprise more than 30 intake air sys-tems from all front-ranking gas turbine manu-facturers. One representative example each for a three-stage filter system ( F i g u r e 6 ) and the continuation of a two-stage high-perform-ance system ( F i g u r e 7 , 8 ) with a final HEPA filter stage are adduced here to demon-strate the expectations posed for the relevant performance capabilities.

Typical Example of Three-stage Filtration in HEPA Quality

System type 2 x Taurus 65 GT

Place of installation Cardboard factory

Systems rated at 6.3 MW

Total volume flow per system 67,000 m3/h

1st filter stage 28 x F6 pocket filter T60

2nd filter stage 28 x F9 MaxiPleat MX 98

3rd filter stage 28 x H11 MaxiPleat MX100

Typical Example of Two-stage Filtration in HEPA Quality

System type 2 x Alstom GT13E2

Place of installation South East Asia

Systems rated at 165 MW each

Total volume flow per system 1,500,000 m3/h

1st filter stage 360 x F8 MaxiPleat MX 95

2nd filter stage 360 x H11 MaxiPleat MX100

Summary and Outlook

The empirical feedback and the results pre-sented demonstrate that installing upgraded

filtration quality in the air intake systems for gas turbines definitely makes sense, with con-comitant benefits for the users concerned.

By means of systematically analysed feedback from actual applications, characteristic pres-sure-drop values can be computed for existing filter systems, which permit adequate com-parisons with planned modifications towards upgraded filtration quality. Upgrades to three-stage filter systems already implemented with high-efficiency filter elements in the final stage of filtration enable frequent compressor-stage washing routines to be dispensed with. Note that modularized filter systems, which can be installed upstream without any struc-tural modifications and substantially increase filtration quality while only requiring very lit-tle additional outlay, have proved their worth in actual operation. Obviating the need for compressor-stage washing, and increased sys-tem availability, usually compensate many times over for the slightly higher pressure drop in the upgraded filtration system. Trials in ac-tual practice with two-stage intake air systems that reach a filtration quality equivalent to that of HEPA filters are currently in the test phase and may well prove to be a further step towards optimization, targeting maximally cost-effi-cient system operation. The option for doing without a washing system opens up yet anoth-er step towards cost optimization for turbine manufacturers.

References

[1] Schroth, T., Rothmann, A., and Schmitt, D.: Nutzwert eines dreistufigen Luftfiltersystems mit innovativer Technologie für stationäre Gas-turbinen. VGB PowerTech 10/2007, 48-51

[2] European Environment Agency, http://dataservice.eea.europa.eu

[3] Umweltbundesamt mit Daten der Messnetze der Länder and des Bundes, http://www.umweltbundesamt.de

[4] Mattenklott, M., and Höfert, N.: Gefahrstoffe 69 (2009) No. 4 April, P. 127

[5] Schmidt, E., and Löffler, F.: Filternde Abschei-der. In: Brauer, H. Handbuch des Umweltschut-zes and der Umweltschutztechnik, Vol. 2, Chapter 5. Berlin: Springer 1996

[6] United Nations Environment Programme, Glo-bal Environment Outlook, UNEP/DEWA/GRID-Europe, GEO Data Portal, Concentra-tions of Particulate Matter (PM10) 2005. h

Freudenberg Filtration Technologies KG69465 Weinheim/GermanyTel. +49 (0) 6201/80-6264 | Fax +49 (0) 6201/[email protected] | www.viledon-filter.com

SR_VGB_engl_12-09 09.12.2009 11.15 Uhr Seite 2