11376_06 design guide clean rooms

7/31/2019 11376_06 Design Guide Clean Rooms

1/31

Clean Rooms

Design Guide

July 2007


2/31

This page has been left blank intentionally.


3/31

VESDA Clean Rooms Design Guide

i

Preface

Xtralis has produced this Design Guide as a reference, to be consulted when designing andspecifying VESDA fire protection solutions for Clean Rooms of all sizes. There is a significant firerisk within such environments due to the high volume of electrical equipment, continuous airsupply, operation of fully automated processes and presence of flammable/explosive substances.In addition to the fire risk, smoke detection is made more difficult by air movement caused by theforced ventilation required to maintain the sterile environment.

In this Design Guide we will discuss the relevant design considerations and makerecommendations regarding the most effective way in which to install a VESDA solution in theparticular Clean Room environment for which it is being designed.

The majority of Clean Rooms are typically found in the following facilities:

Semiconductor manufacturing plants. Electronic device manufacturing plants.

Pharmaceutical processing plants.

Research and development laboratories.

Important Note: The information contained in this Design Guide should be used in conjunctionwith specific local fire codes and standards as well as the guidelines providedin the VESDA System Design Manual

[1]. Where applicable, other regional

industry practices should also be adhered to.

Co yright 2007 Xtralis AG


4/31

Clean Rooms Design Guide VESDA

ii



5/31


iii

Contents

1. Background Information .........................................................................................................................11.1 Fire Safety Considerations in Clean Rooms.................................................................................11.2 Performance-Based Design..........................................................................................................11.3 Key Design Considerations...........................................................................................................21.4 Why Use VESDA Smoke Detection?............................................................................................3

2. VEWFD And EWFD ..................................................................................................................................42.1 Levels Of Protection......................................................................................................................42.2 The Effects Of Airflow ...................................................................................................................52.3 Detector Coverage Comparisons..................................................................................................5

3. Fab Area Ceiling Protection....................................................................................................................64. Under Floor Void (UFV) Protection ........................................................................................................7

4.1 Perforated Floor Void Protection...................................................................................................84.2 Solid Floor Void Protection............................................................................................................9

5. Sub-fab Area Protection........................................................................................................................115.1 Under Waffle Slab (UWS) Ceiling Protection..............................................................................115.2 Dry Coils/Return Air Vent Protection...........................................................................................145.3 Ancillary High Risk Area Protection ............................................................................................15

6. Air Handling Unit (AHU) Protection .....................................................................................................167. Object Protection...................................................................................................................................17

7.1 VESDA OEM Solution.................................................................................................................177.2 Utilities And Process Tool Protection..........................................................................................17

8. Truss And Ceiling Void Protection ......................................................................................................189. PROACTIV Fire Protection System......................................................................................................18

9.1 Single Site Monitoring.................................................................................................................199.2 Multiple Site Monitoring...............................................................................................................19

10. Commissioning, Service and Maintenance.........................................................................................20Appendix VESDA Smoke Thresholds Set-up Procedure For Clean Rooms ........................................21References .....................................................................................................................................................23


6/31


iv



7/31


1

1. Background Information

1.1 Fire Safety Considerations in Clean Rooms

The major fire risks and detection difficulties within Clean Rooms arise as a result of the following:

Utility and process tool equipment faults.

Electrical faults in cabling or other electrical and electronic equipment. The Over Head Transport (OHT) System. The presence of large amounts of flammable and explosive materials.

Potential rapid fire growth resulting from the delayed control of forced ventilation.

Rapid fire spread as a result of the air circulation used to filter out pollutants. Air movement dilution of and interference with the normal dispersion of smoke (Figure 1),

making its detection by conventional technologies extremely difficult.

Low air velocity smoke distribution.

High air velocity smoke distribution.

Figure 1 Examples of the way in which different air velocities affect the dispersionof smoke.

Important Note: It is recommended that smoke tests or Computational Fluid Dynamics (CFD)models be used to determine the optimum arrangement of the VESDAsampling pipes, based on predicted smoke movement paths with reference tothe above risks.

1.2 Performance-Based Design

The unique environments within Clean Rooms present a challenge to both early and reliable firedetection. There is a high likelihood that detection system performance will be dependent on airchange rate, air velocity in the particular area and the geometry of the area to be protected. Theflexibility of Performance-Based Design, while still following rigorous engineering processes, allowsthe fire protection system to be tailored to the specific requirements of each individual applications

environment, with the commercial drivers to manage the risks.


8/31


2

Detector spacing or, for a VESDA pipe, sampling hole spacing is traditionally dictated by localprescriptive codes and standards. In a more performance-based approach, each installation isassessed according to its specific environmental conditions. Sampling hole spacing and locationcan then be altered easily to suit the particular performance requirements

The Performance-Based Design approach is widely used since it can provide evidence to justifydivergence from prescriptive requirements, particularly in cases where there are practicallimitations or a need for an improved level of fire protection. There are some specific guidelines forthe use of Performance-Based Design and risk management concepts.

Examples of these codes and standards are listed below:

International Fire Engineering Guidelines (Edition 2005)[2]

.

British Standard BS 7974[3]

.

SFPE Engineering Guide to Performance-Based Fire Protection[4]

. AS/NZ 4360 Risk Management Standard

[5].

SFPE Handbook of Fire Protection Engineering Third Edition[6]

.

Performance-Based fire protection solutions can be made to comply with local and national codesand standards for buildings and life safety. Assessments of the environmental risks andperformance requirements, specific to the particular Datacom Facility, are conducted as part of thedesign process.

Note: The SFPE Code Officials Guide to Performance-Based Design Review[7]

is also a verygood source of information for Authorities Having Jurisdiction (AHJs) reviewing andassessing a VESDA system design for a Clean Room.

1.3 Key Design Considerations

The following should be considered when designing a VESDA system for a Clean Room:

1. What level of protection is required and how will fire safety be managed?2. Do local codes and standards recognise Aspirating Smoke Detection (ASD) technology for

primary protection or will it be necessary to employ Performance-Based Design principles todesign a VESDA system with point (spot) type smoke detector equivalent performance?

3. Which areas require protection (Fab, Under Floor Void (UFV), Sub-fab, Air Handling Unit(AHU), ceiling void and/or individual tools)?

4. Does the risk assessment, for particular areas, indicate the need for Early Warning FireDetection (EWFD) or Very Early Warning Fire Detection (VEWFD)?

5. What equipment represents a primary fire risk, process tools (e.g. wet benches, steppers,stockers, ion implanters, photolithography equipment etc), and where is it located within theClean Room?

6. Where are the AHUs, filter and pre-filter systems located?7. What will be the effect on airflow patterns and smoke detection of parameters such as ceiling

height, room geometry and equipment locations/dimensions?8. Are there likely to be future operational changes such as new equipment, a change in

perforated floor tile location etc?9. Will particular pieces of equipment require additional individual object protection?10. Is the integrity of the pipe network adequate with respect to being air-tight?11. Where fire suppression is also installed, how can the VESDA system be used to activate it?12. If implementing Xtraliss PROACTIV system, how will the interfaces of the VESDA system be

integrated with the Fire Alarm Control Panel (FACP)? Note that PROACTIV is not available inEurope and that you must ensure PROACTIV approvals requirements comply with your localcodes.

13. How are the fire protection and security systems to be integrated within the local emergencycontrol centre?

14. What do local industry codes of practice for Clean Room fire safety dictate?15. What do local fire codes and standards recommend (e.g. NFPA 318

[8]and BS 5295

[9])?

16. What are the relevant Performance-Based Design practices (e.g. the International Fire

Engineering Guidelines[2]

, BS 7974[3]

or those of the SFPE[4]

)?


9/31


3

Important Note: The arrangement of process tools and other equipment will affect the speedand direction of the airflow. It is strongly recommended, therefore, thatperformance tests be conducted once the system is installed. All such testsshould be carried out with the full cooperation of the facilities safety officers.

1.4 Why Use VESDA Smoke Detection?

It is essential that fire events in Clean Rooms be detected as early as possible, to avoid assetdamage, leading to costly business disruption, and to ensure occupant safety.

The limitations of point (spot) type smoke and heat detectors must be considered. Thecomparatively low sensitivity of point (spot) type smoke detectors, for instance, can mean that fireevents will not be detected soon enough in many cases. Rather than behaving as they normallywould, both smoke and heat from low energy fires will tend to follow the air streams created by theAHUs with the following consequences:

Air movement, filtering and the introduction of clean air during the air conditioning cycle will allcause smoke dilution. This consequence impairs the performance of point (spot) type smokedetectors.

The cooling effects of the air conditioning will decrease the temperature of the smoke plume.

This consequence impairs the performance of both point (spot) type smoke and heat detectors.

Note: The performance of point (spot) type smoke detectors may also be restricted by thevelocity of the air/smoke passing across the detection chamber and/or its temperature.

The Very Early Warning Fire Detection (VEWFD) capability of the VESDA system allows it tominimize the risks, discussed in section 1.1, and combat the detection difficulties caused by airmovement in the following ways:

VESDA detectors can be configured as both early warning and very early warning devices,thus, if both are required only one technology need be installed.

A VESDA system can detect fires very early, at the incipient (smoldering) stage. This providesstaff with an opportunity to investigate and take action, before the smoke contamination can

irreversibly damage process tools or the products being manufactured. The very early warning capability of the VESDA system also minimizes the rapid fire growth and

spread facilitated by the high air movement.

Since the security precautions necessary to ensure the maintenance of the sterile environmentwithin the Clean Room also complicate evacuation, very early warning will allow more time toexecute an evacuation.

A VESDA system has a better chance of detecting smoke that has been diluted thanconventional point (spot) type smoke detectors. This is due to the fact that air collected byseveral sampling holes, at different locations within the protected area, is being analyzedsimultaneously by the same detector (aggregated sampling effect).

A VESDA system actively draws air into its sampling holes which increases the chance ofsmoke being detected. Passive smoke detectors rely on smoke reaching them via diffusion orusing the thermal energy of the fire. Thus the detection of smoke by passive detectors would beless likely where air movement is being artificially directed as well as being cooled.

There is a comparatively low incidence of nuisance alarms with a VESDA system; a featurewhich will minimize the possibility of unnecessary evacuations.

In cases where fire suppression is to be included, as part of the overall fire protection system,the VESDA detectors wide sensitivity range of 0.005 to 20%Obs/m (0.0015 to 6%Obs/ft) meansthat appropriate alarm thresholds can be set for both early detection and, at a later stage in thefire event, the activation of the suppression release mechanisms.


10/31


4

2. VEWFD And EWFD

2.1 Levels Of Protection

Critical areas within a Clean Room require Very Early Warning Fire Detection (VEWFD), while

other areas may only require Early Warning Fire Detection (EWFD). In our experience, a VEWFDsystem is capable of detecting representative incipient fires, typically within 120 to 150 seconds,regardless of smoke location. An EWFD system, on the other hand, appears to take between 150and 180 seconds to detect fires of this size. VESDA systems can be designed to provide VEWFDor EWFD depending on the area in question.

The recommended levels of protection, for the various areas within a Clean Room, are presentedin Table 1.

Table 1 The type of protection necessary in various areas within a Clean Room.

Area Essential Recommended

Fab Area Ceiling

(EWFD)Under Floor Void (UFV)

Perforated UFV

Solid UFV

(VEWFD)

(VEWFD)

Sub-fab Area

UWS Ceiling

Dry Coils/ReturnAir Vent

Ancillary High Risk

Areas

(EWFD)

(VEWFD)

(VEWFD)

Air Handling Unit (AHU) (VEWFD)

Object Protection

Utilities & ProcessTools

Power Rooms

(VEWFD)

(EWFD)

Ceiling Void (EWFD)

Over Head Transport Utilities (VEWFD)

The VESDA Firealarm levels can be set to provide sampling hole sensitivities equivalent to point(spot) type smoke detectors. The Alertand Actionalarm levels are used as part of the emergencyprocedure controlled from the local Emergency Response Centre.


11/31


5

2.2 The Effects Of Airflow

Detector coverage is dependent on the air movement within the Clean Room area. Rather thanusing air change rate, which is a function of the volume of the enclosure, airflow within CleanRooms is represented by a quantity known as the average air velocity. Measurements of theaverage air velocity are usually made at the following locations:

0.3 m (1 ft) beneath the Filter Fan Units (FFUs). Under Waffle Slabs (UWSs). Across the Dry Coils/Return Air Vents.

Since all Clean Rooms vary in size and have differing ventilation arrangements (FFUs, AHUs orFan Towers), their average velocities will also vary. However the guidelines outlined in this DesignGuide are applicable to any Clean Room. The average air velocity can also vary widely from onearea, within the Clean Room, to another. This must be taken into account when designing aVEWFD system. Table 2 contains examples of a typical range for measurements of the average airvelocity.

Table 2 Examples of typical average air velocity measurements.

Average Air VelocityLevel

Below The FFUs (m/s(fpm))

At The DryCoils/Return AirVents (m/s (fpm))

Nominal AirChanges Per Hour

Low 0.15 (30) 0.75 (150) 25

Average 0.30 (50) 1.10 (220) 50

Medium 0.60 (120) 2.30 (460) 100

High 1.20 (240) 4.70 (940) 200

Very High 2.40 (480) 9.50 (1900) 400

Note: The average air velocity, in the locations listed above, can usually be obtained from theMechanical Services designers before beginning the Fire Protection System design.

2.3 Detector Coverage Comparisons

For the purposes of this Design Guide, the recommended area of coverage for VESDA detectors isbased on the following factors:

The fire size to be detected. Air velocity in the protected area.

Maximum Transport Time permitted by the ASPIRE2 pipe network modelling program.

Sampling hole spacing rules.

Throughout the remainder of this Design Guide, recommendations regarding the appropriateVESDA detector area of coverage and sampling hole spacing will be provided.

Table 3 illustrates the relationships between detector coverage and the factors listed above forVEWFD in comparison to EWFD.


12/31


6

Table 3 Comparison of the factors affecting detector area of coverage for VEWFD and EWFD.

Factor VEWFD EWFD

Level of Protection VERY High High

Detector area of Coverage Smaller Larger

Air Velocity Higher Lower

Sampling Hole Spacing Reduced Increased

Maximum Transport Time Shorter Longer

It is clear from the information presented in Table 3 that as the air velocity within the protected areaincreases, all other factors must be decreased in order to achieve VEWFD.

Important Note: All pipe network designs should be verified by Xtraliss Pipe Network ModellingTool, ASPIRE2

TM. This program will also determine which VESDA model

(LaserFOCUS, LaserCOMPACT, LaserPLUS or LaserSCANNNER) would bemost suited to the protection of a particular area.

3. Fab Area Ceiling Protection

The Fab area contains the manufacturing and process tools (Figure 2) which represent some of themajor fire risks outlined previously (Section 1.1).

Figure 2 Illustration of a typical Fab area in a Clean Room.

Circulating air, introduced through the ceiling and removed through the floor, will dilute smoke andaffect its dispersion pattern. Therefore, in comparison to conventional point (spot) type smokedetectors, ceiling mounted VESDA sampling pipes are a reliable and effective way of detectingsmoke under these circumstances.

Note: The sampling pipe network can be placed directly on the ceiling below the FFUs with thesampling holes facing downward.

The VESDA system is designed to replace point (spot) type smoke detectors. Recommendeddetector coverage, for a single VESDA detectors pipe network, is 500 m

2(5,400 sq.ft) (Figure 3)

with a sampling hole spacing of 6 m (20 ft). An increase from low to medium air velocity requires noalteration in the area of coverage of VESDA detectors due to the aggregation effect caused bysmoke entering a greater number of sampling holes when it is moved around more rapidly.


13/31


7

0

100

200

300

400

500

600

0.15 / 30 0.3 / 60 1.2 / 240 2.4 / 470

FFU Air Velocity (m/s - fpm)

VESDACoverage(m

2)

0

1000

2000

3000

4000

5000

6000

VESDACoverage(sq.f

t)

6 m x 6 m (20 ft x 20 ft) spacing

Figure 3 Recommended detector coverage for ceiling mounted sampling pipes in theFab area.

Important Note: Since all Fab areas vary with respect to layout and specifications, anassessment of each should be conducted to determine factors such as airchange rate, equipment location etc as part of the design process.

Note: Future Clean Rooms are expected to have higher Fab and Sub-fab area ceilings. This willmake the VESDA systems VEWFD capabilities even more crucial.

4. Under Floor Void (UFV) ProtectionThe Under Floor Void (UVF), used to house cable trays, is located between the Fab and Sub-fabareas. It also serves as a return air plenum, feeding air back to the AHU or return air shaft asshown (Figure 4). There are two possible configurations for the UFV: a perforated or solid (non-perforated) floor void.


14/31


8

Figure 4 Illustration of the Under Floor Void area.

4.1 Perforated Floor Void Protection

In this configuration, air exits directly downwards into the Sub-fab area through the perforated floortiles. These tiles form part of the Under Waffle Slab (UWS) ceiling structure of the Sub-fab area.

For a single VESDA detector, the recommended areas of coverage relative to the average airvelocity, measured at the FFU in an UFV with a perforated floor void, are shown (Figure 5).

0

100

200

300

400

500

600

0.15 / 30 0.3 / 60 1.2 / 240 2.4 / 470


VESDAC

overage(m

2)

0

1000

2000

3000

4000

5000

6000

VESDACo

verage(sq.f

t)


4 m x 4 m(13 ft x 13 ft)

spacing

Figure 5 Recommended detector coverage, for different average air velocities, in anUFV with a perforated floor void.


15/31


9

Both detector area of coverage and sampling hole spacing are reduced as average air velocityincreases, to maintain high VESDA detection performance. The sampling pipe network ispositioned 0.8 m (2.6 ft) beneath the ceiling of the UFV to ensure adequate mixing of smoke.

Important Note: In general, it is acceptable for sampling holes to face in the same direction asthe airflow. However, in cases where the airflow characteristics are likely tovary, it is necessary to orientate the sampling holes at an angle of 30 to thedirection of airflow. This neutralizes pressure difference effects at the samplingholes.

If conventional point (spot) type smoke detectors are being used to protect the Fab ceiling in a highairflow environment (FFU average air velocity of > 1.2 m/s (240 fpm)), the area of coverage perVESDA detector and sampling hole spacing within the UFV must be reduced to provide VEWFD.Detector coverage should be between 200 and 300 m

2(2,100 3,200 sq.ft)); sampling hole

spacing being 4 m (13 ft). Under these circumstances, the VESDA system should still detect Fabarea fire events which would not be detected by point (spot) type smoke detectors mounted on theFab ceiling.

Note: Due to the limited space within the UFV, the use of continuous, flexible HDPE piping isrecommended to avoid interference with equipment foundations and hookuppiping/ducting. When using this type of material, compression or electric fusion joints are

necessary.

4.2 Solid Floor Void Protection

In this case, the floor of the return air plenum is non-perforated so air must be returned via ReturnAir Vents similar to those in the Sub-fab area. These Return Air Vents are generally of smaller sizethan those in the Sub-fab. There is no waffle slab and the air is returned horizontally, hence, it isessential to protect the cabling and any equipment with its exhaust venting into this area (Figure 6).Sampling hole orientation should be at an angle of 30 to the direction of airflow.

Figure 6 Air sampling in a Fab with a solid UFV.

For a single VESDA detector, the recommended areas of coverage relative to the average airvelocity, measured at the FFU in an UFV with a solid floor void, are shown (Figure 7).


16/31


10

0

100

200

300

400

500

600

700

0.3 / 60 0.6 / 120 1.2 / 240 2.4 / 470


VESDACoverage(m

2)

0

1000

2000

3000

4000

5000

6000

7000

VESDACoverage(sq.f

t)

4 m x 4 m (13 ft x 13 ft)spacing

Figure 7 - Recommended detector coverage, for different average air velocities, in anUFV with a solid floor void.

The area of coverage per VESDA detector varies with the average air velocity to ensure theoptimum detection performance.

Due to the airflow pattern within the UFV, sampling holes are arranged in a zigzag or alternatingmanner as shown (Figure 8) to maximize coverage.

Airflow

SamplingPipe

SamplingHole

ReturnAir

Figure 8 Zigzag or alternating arrangement of samplingholes in a Fab with a solid UFV.


17/31


11

5. Sub-fab Area Protection

In its simplest form, the Sub-fab area serves as a return air plenum for the AHU. Quite commonly, italso houses the support equipment associated with the manufacturing equipment contained abovein the Fab area. Sometimes, in the case of LCD manufacture for example, a proportion of themanufacturing process actually takes place in the Sub-fab area. In other instances, automated

product transport conveyers and shuttles may also be present. Thus, there is a significant fire riskin this area which makes its protection essential. A typical Sub-fab area is shown below (Figure 9).

Figure 9 Illustration of a Sub-fab area.

The next two sections contain recommendations on detection methods suited to the protection of anumber of features common to Sub-fab areas.

5.1 Under Waffle Slab (UWS) Ceiling Protection

This is a common protection practice in Clean Rooms today. For a single VESDA detector, therecommended areas of ceiling detector coverage relative to the average air velocity, measured atthe FFU in a Sub-fab area, are shown (Figure 10).


18/31


12

0

100

200

300

400

500

600

700

0.15 / 30 0.3 / 60 0.6 / 120 1.2 / 240 2.4 / 470

VESDACoverag

e(m

2)

0

1000

2000

3000

4000

5000

6000

7000


VESDACoverage(sq.f

t)

EWFD

VEWFD


6 m x 6 m (20 ft x 20 ft) spacing 4 m x 4 m(13 ft x 13 ft)

spacing

Figure 10 - Recommended detector coverage, for different average air velocities, onthe ceiling of a Sub-fab area.

In order to provide VEWFD, special consideration is required in the case of Sub-fab areas wherethe airflow at various points varies significantly. For example, at the centre of the Sub-fab area airmovement may only be slight. Conversely, next to the Dry Coils/Return Air Vent the airflow wouldbe high in comparison to that at the centre of the Sub-fab area.

When designing a VESDA system to cover the entire floor space under the waffle slab (Figure 11),where the FFU measured average air velocity is 2 m/s (394 fpm), the following applies:

Each VESDA detector should cover an area no larger than 500 m2 (5,400 sq.ft). Sampling hole spacing should be no more than 6 m (20 ft).

All equipment and other objects must be a distance of at least 5 m (16 ft) from the DryCoils/Return Air Vent (D1 in Figure 11). Any closer than D1 would place objects in the regionwhere smoke detection is most difficult.

Unless otherwise stated, by local codes and standards, VESDA sampling pipes should be nofurther from the walls than D2 (Figure 11), that is, 5 m (16 ft).


19/31


13

Dry Coils

Figure 11 VESDA detector coverage on the ceiling of a Sub-fab area wherethe average air velocity at the FFU is2 m/s (394 fpm).

When designing a VESDA system to cover the entire floor space under the waffle slab (Figure 12),where the FFU measured average air velocity is > 2 m/s (394 fpm), the following applies:

For normal density coverage, each VESDA detector should cover an area no larger than 500 m2

(5,400 sq.ft).

For high density coverage (i.e. close to the Dry Coils/Return Air Vent) each VESDA detectorshould cover an area no larger than 200 - 250 m

2(2,200 2,700 sq.ft).

Sampling hole spacing should be no more than 6 m (20 ft).

All equipment and other objects must be a distance of at least 10 m (30 ft) from the DryCoils/Return Air Vent (D1 in Figure 12). Any closer places objects in the region within whichsmoke detection is most difficult.

Unless otherwise stated, by local codes and standards, VESDA sampling pipes should be nofurther from the walls than D2 (Figure 12), that is, 5 m (16 ft).


20/31


14

Figure 12 VESDA detector coverage on the ceiling of a Sub-fab areawhere the average air velocity at the FFU is > 2 m/s (394 fpm).

5.2 Dry Coils/Return Air Vent Protection

NFPA 318[8]

considers this a primary area of protection. Since this is the last point at which all airwill pass before being filtered, it is a critical location for detection in the Clean Room. Placingsampling pipes across the Dry Coils/Return Air Vent as shown (Figure 13) allows the air to besampled just prior to entering the air handling and filtering systems. This provides reliable and veryearly warning smoke detection.

Figure 13 Example of Dry Coils/Return Air Ventsampling.


21/31


15

The pipe network should be positioned 50 to 200 mm (2 to 8 inches) in front of the Dry Coils/ReturnAir Vent to minimize the possible adverse effects of the air pressure differences created byturbulent airflow at the surface of the vent. Sampling hole orientation, in cases where the airflowcharacteristics are likely to vary, should be at an angle of 30 to the direction of airflow.

For a single VESDA detector, the recommended areas of coverage (i.e. across the face of the DryCoils/Return Air Vent) relative to the average air velocity, measured 200 mm (8 inches) in front ofthe Dry Coils/Return Air Vent, are shown (Figure 14).

0

10

20

30

40

50

60

70

80

90

0.75 / 150 1.1 / 220 2.3 / 450 4.7 / 925 9.5 / 1870

Return Air Velocity (m/s - fpm)

VESD

ACoverage(m

2)

0

100

200

300

400

500

600

700

800

900

VESDACoverage(sq.f

t)

EWFD

VEWFD

1.2 m x 1.5 m (4 ft x 5 ft) spacing

1.2 m x 1.0 m(4 ft x 3 ft)spacing

1.2 m x 1.0 m(4 ft x 3 ft)spacing

Figure 14 Recommended detector coverage, for different average air velocities, atthe Dry Coils/Return Air Vent in the Sub-fab area.

5.3 Ancillary High Risk Area Protection

Since the electrical equipment within a Clean Room operates twenty-four hours a day seven days aweek, a power failure would be undesirable. Uninterruptible Power Supply (UPS) systems are usedto avoid such power outages, however, they also represent a significant fire risk. In-cabinet and oncabinet air sampling, of control cabinets and UPS systems, (Figure 15) offers the most effectivedetection of incipient fires.


22/31


16

In-cabinet Sampling

On cabinet Sampling

Figure 15 Example of in/on cabinet air sampling.

The protection of such areas as UPS systems and Server Rooms is identical to the protection ofany other Datacom Facility; refer to the Xtraliss Telecommunications and Data ProcessingFacilities Design Guide

[10]. Ceiling mounted detection, governed by local codes and standards such

as NFPA 72

[11]

, together with Return Air Vent and in/on cabinet monitoring provides an effective fireprotection solution.

Note: For duct protection, follow the guidelines in Application Note VESDA air sampling inducts

[12].

6. Air Handling Unit (AHU) Protection

The Clean Room environment is maintained by either a large powerful AHU or a number of smallerFFUs. Since the air being circulated by this equipment is filtered, fires within it would need tobecome reasonably large before they could be detected externally in the Fab or Sub-fab areas. Forthis reason, it is recommended that VESDA sampling pipe networks be placed above the FFUs or

inside individual AHU cabinets as shown (Figure 16).


23/31


17

Figure 16 Example of FFU air sampling.

7. Object Protection

7.1 VESDA OEM Solution

When designing for object protection with a VESDA system, the VESDA detector and samplingpipe network are installed on the process tool to be protected. Some manufacturers install VESDAdetectors as an optional feature on their equipment. Sampling pipe length should be as short aspossible to minimize Transport Time, thereby, increasing the speed of detection.

VESDA OEM solutions are usually explicitly designed, in conjunction with the equipmentmanufacturer, to provide the best possible detection performance for an individual process tool.Solutions can also be designed to meet specific requirements for a particular type of process tool.For more information contact your local Xtralis representative.

7.2 Utilities And Process Tool Protection

A large number of Clean Room utility and process tools are housed within free-standing cabinets.Either in-cabinet or cabinet exhaust air sampling provides the most effective method for thedetection of fire events within these cabinets. Sampling pipes are placed at the cabinet exhaust airduct and at critical positions inside the cabinet as shown (Figure 17).


24/31


18

Figure 17 Example of air sampling inside a process tool cabinet.

Note: The manufacturing processes, taking place within this type of equipment, can result inhigh background levels of debris and process byproducts.

8. Truss And Ceiling Void Protection

The Truss and Return Air Shaft contain large amounts of electrical equipment. By installing VESDAdetectors in the roof space, the overall protection of the entire facility can be increased. Locatingthem in the roof space also makes them easy to access.

Area of detector coverage and sampling holes spacing can be designed to comply with localstandards, allowing the VESDA system to form the primary detection system in the place of point(spot) type smoke detectors.

9. PROACTIV Fire Protection System

The PROACTIV system is an integrated VESDA fire management system, used to control anetwork of VESDA detectors and other associated devices. Relevant features are summarizedbelow:

PROACTIV is the only system capable of providing effective integration of VESDA ASDtechnology and point (spot) type smoke and heat detection.

The central monitoring of networked PROACTIV systems allows accurate and rapid control ofresponses, from multiple sites, at a single location.

PROACTIV can control multiple suppression devices, including gaseous suppression, inmultiple zones.

Flexible online/offline configuration via direct serial, modem or TCP/IP connections. The PROACTIV System Manager software (PSM) allows the configuration, monitoring and

troubleshooting of any PROACTIV system.


25/31


19

The PROACTIV system can function as a standalone system or as a networked system. Bothconfigurations are illustrated below.

9.1 Single Site Monitoring

A single PROACTIV system (Figure 18) can be connected to a range of approved detectors,warning systems, suppression control devices and other appropriate ancillary devices in a Clean

Room.

Figure 18 Example of a single PROACTIV fire protection system.

9.2 Multiple Site Monitoring

PROACTIV supports the networking of up to 32 systems, via the PROnet network, and multiplePROACTIV networks over local or wide area TCP/IP networks (Figure 19). This is achieved usingthe PROACTIV System Management (PSM) monitoring software.


26/31


20

Figure 19 Example of multiple site integrated monitoring using PROACTIVfire protection systems.

For more information about PROACTIV, visit our website at www.xtralis.com or contact your localXtralis office.

10. Commissioning, Service and Maintenance

Once the VESDA system has been installed, its performance and sampling pipe network integritycan be verified against the original ASPIRE2

TMdesign. A range of environmental parameters can

be input to determine Maximum Transport Times for each zone. Calculated Transport Timesshould be applied conservatively.

Smoke tests may also be conducted, as part of the commissioning process, to test systemperformance. Refer to Xtraliss latest revision of Clean Rooms Smoke Test Method[13] for furtherdetails.

The VESDA alarm thresholds, for optimum protection, can be determined according to theprocedure highlighted in the Appendix.

Note: It is expected that, while the Clean Room is being commissioned, there will be a greaterconcentration of particles in the air within the UFV than when the facility has been up andrunning for some time. In cases where the UFV contamination was high, it may beadvisable to change the VESDA detector filter(s) after commissioning. Once the UFVenvironment has stabilized, the procedure highlighted in the Appendix can be re-appliedto determine the appropriate ongoing alarm threshold settings.

System maintenance should be performed in accordance with local codes and the Clean Room riskmanagement requirements as well as the procedures outlined in the Maintenance section of theVESDA System Design Manual

[1].


27/31


21

Appendix VESDA Smoke Thresholds Set-upProcedure For Clean Rooms

Purpose

This Procedure aims to provide step-by-step instructions on the set-up of the VESDA smokethresholds for a Clean Room facility.

Apparatus

The equipment required is as follows:

VESDA detector; Sliding Windows High Level Interface (HLI) (VHX-200 ) x 1;

PC with VESDA Ssystem Manager (VSM4).

Average Background Level

The steps below need to be followed to obtain the average background level in the area of interest inthe Clean Room.

Step Action

1Use the Sliding Windows HLI to connect the VESDA detector to VSM4. Notesome VESDA detectors may be connected directly to VSM4.

2

In VSM4, highlight the detector under test. An example is shown below.

3

Go to the Device menu and select Start AutoLearn Smoke as shown below.


28/31


22

Step Action

4

Enter an AutoLearn period of a minimum of 1 day and then click on the Startbutton.

5 Obtain the Firealarm threshold upon the completion of the AutoLearn period.

6Divide the Firealarm threshold by 10 which will correspond to the averagebackground level in the environment.

Procedure

The steps below need to be followed to set the VESDA smoke alarm thresholds.

Step Action

1 Obtain the average background level from above.

2

Set the VESDA Alert alarm threshold to the background value plus

0.015%Obs/m (0.005%Obs/ft).

3Set the VESDA Action alarm threshold to the background value plus0.03%Obs/m (0.009%Obs/ft).

4Set the VESDA Fire 1 alarm threshold to the background value plus0.05%Obs/m (0.015%Obs/ft).

5Set the VESDA Fire 2 alarm threshold to the background value plus0.25%Obs/m (0.08%Obs/ft).


29/31


23

References

[1] Xtralis (2006) VESDA System Design Manual, Ed. 4.5.

[2] Australian Government State and Territories of Australia (2005) International Fire EngineeringGuidelines.

[3] British Government (2001) British Standard BS 7974: Application of Fire Engineering Principlesto the Design of Buildings.

[4] SFPE (2000) Engineering Guide to Performance-Based Fire Protection Analysis and Design ofBuildings

[5] Australian Standard AS/NZ 4360 (1999) Risk Management Standard.

[6] SFPE (2002) Handbook of Fire Protection Engineering, 3rd

Edition.

[7] SFPE & ICC (2004) The Code Officials Guide To Performance-Based Design Review.

[8] NFPA (2002) Standard for the Protection of Semiconductor Fabrication Facilities 318.

[9] British Government (1989) British Standard BS 5295: Environmental cleanliness in enclosed

spaces. General introduction, terms and definitions for clean rooms and clean air devices.[10] Xtralis (2007) Telecommunications & Data Processing Facilities Design Guide, Doc. No.11782.

[11] NFPA (2002) National Fire Alarm Code (72).

[12] Xtralis (2004) Application Note: VESDA Air Sampling for Ducts, Doc. No. 10847.

[13] Xtralis (2007) Clean Rooms Smoke Test Method, Doc. No. 11509.


30/31


24

Disclaimer On The Provision Of General System DesignRecommendations

Any recommendation on system design provided by Xtralis is an indication only of what isconsidered to be the most suitable solution to meet the needs of the common application

environments described.

In some cases the recommendations on system design provided may not suit the unique set ofconditions experienced in a particular application environment. Xtralis has made no inquiry norundertaken any due diligence that any of the recommendations supplied will meet any particularapplication. Xtralis makes no warranty as to the suitability or performance of any recommendationon system design. Xtralis has not assessed the recommendation on system design for compliancewith any codes or standards that may apply nor have any tests been conducted to assess theappropriateness of any recommendations on system design. Any person or organization accessingor using a recommendation on system design should, at its own cost and expense, procure that therecommendation on system design complies in all respects with the provision of all legislation, actsof government, regulations, rules and by-laws for the time being in force and all orders or directionswhich may be made or given by any statutory or any other competent authority in respect of or

affecting the recommendation on system design in any jurisdiction in which it may be implemented.

Xtralis products must only be installed, configured and used strictly in accordance with the GeneralTerms and Conditions, User Manual and product documents available from Xtralis. Xtralis acceptsno liability for the performance of the recommendation on system design or for any productsutilized in the implementation of the recommendation on system design, aside from the GeneralTerms and Conditions, User Manual and product documents.

No statement of fact, drawing or representation made by Xtralis either in this document or orally inrelation to this recommendation on system design is to be construed as a representation,undertaking or warranty.

To the extent permitted by law, Xtralis excludes liability for all indirect and consequential damages

however arising. For the purposes of this clause, consequential damage shall include, but not belimited to, loss of profit or goodwill or similar financial loss or any payment made or due to any thirdparty.

Recommendations on system design are provided exclusively to assist in design of systems usingXtralis products. No portion of this recommendation on system design can be reproduced withoutthe prior approval in writing of Xtralis. Copyright and any associated intellectual property in anysuch recommendations on system design or documentation remains the property of Xtralis.


31/31

www.xtralis.com

The Americas +1 781 740 2223 Asia +852 2297 2438 Australia and New Zealand +61 3 9936 7000Continental Europe +41 55 285 99 99 UK and the Middle East +44 1442 242 330

The contents of this document are provided on an as is basis. No representation or warranty (either express or implied) is made as to thecompleteness, accuracy or reliability of the contents of this document. The manufacturer reserves the right to change designs or specificationswithout obligation and without further notice. Except as otherwise provided, all warranties, express or implied, including without limitation anyimplied warranties of merchantability and fitness for a particular purpose are expressly excluded.

This document includes registered and unregistered trademarks. All trademarks displayed are the trademarks of their respective owners. Your useof this document does not constitute or create a licence or any other right to use the name and/or trademark and/or label.

This document is subject to copyright owned by Xtralis AG (Xtralis). You agree not to copy, communicate to the public, adapt, distribute, transfer,sell, modify or publish any contents of this document without the express prior written consent of Xtralis.

11376_06 design guide clean rooms

Documents