BG1 2006

Applying HVAC building
services calculations
A BSRIA Guide
Model Demonstration
Project
By David Churcher, John Sands and Chris Parsloe
BSRIA Guide BG 1/2006

Supported by
www.bsria.co.uk
See Guide to HVAC Building Services Calculations, BG 30/2003
ACKNOWLEDGEMENTS
BSRIA would like to thank the following sponsors for their financial support in the
preparation of this Guide:
Department of Trade and Industry
N G Bailey
FaberMaunsell
WSP Group
EMCOR
Hoare Lea & Partners
The research project was undertaken under the guidance of a project steering group
drawn from industry representatives, all of whom gave valuable contributions in kind
to the project:
Steering Group chair
CIBSE
EMCOR
FaberMaunsell
Fulcrum Consulting
Hoare Lea Partners
HVCA
London South Bank University
Mecserve
N G Bailey
Newham College
SummitSkills
WSP Group
Bryan Franklin
Hywel Davies
Leon Hewer
Mike Campbell
Andrew Ford, representing DTI
Graham Cossons
Gavin Crook
Derrick Newson
John Missenden
Lester Bentley
Roland Edkins
Gary Mann
Tony Barton
Richard Tudor
Contributors from BSRIA included John Sands, Chris Parsloe (on behalf of BSRIA), and
David Churcher.
This publication has been produced by BSRIA as part of a contract placed by the
Department of Trade and Industry. The contract was let under the Partners in
Innovation programme, which provided part-funding of collaborative research. Any
views expressed in this Guide are not necessarily those of the Department.
The authors have sought to incorporate the views of the steering group, but final
editorial control of this document rested with BSRIA.
The Association of Consulting Engineers (ACE) is now known as the Association of
Consultancy and Engineering.
The references to the CIBSE Guides are those originally quoted in BG 30/2003.
Subsequent editions of the CIBSE Guides may have changed the references.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in
any form or by any means electronic or mechanical including photocopying, recording or otherwise without
prior written permission of the publisher.
BSRIA 70206
September 2006
ISBN 0 86022 661 1
Printed by Multiplex Medway
MODEL DEMONSTRATION PROJECT

BSRIA BG 1/2006
CONTENTS
Page
1 Introduction to the building process
2 The demonstration project
3 Outline and detailed proposals stage
11
4 Final proposals/production information stage
24
APPENDICES
A Arrangement drawings for outline HVAC design
43
B Outline design heat gain calculations
48
C Outline design heat loss calculations
53
D Outline design
57
E Final proposals/production information
63
F Condensation risk H4
64
G Executive summary from engineering design
69

BSRIA BG 1/2006
1 INTRODUCTION TO THE BUILDING PROCESS

Overview of the building process
The building process covers the complete project from inception
to successful handover, either to the developer for later fit-out or
to the end-user for occupation. This encompasses a number of
stages and can vary with the type of project. There are many
models of the building process, and it is outside the scope of this
project to review all of them. The Royal Institute of British
Architects Plan of Work stages (RIBA, 1999), shown in Table 1, are
used as the model for this report. The Plan of Work breaks
projects down into a series of work stages (A to M).
The 2002 edition of the Association of Consulting Engineers
Agreement B2 (2002 Conditions of Engagement for Mechanical and
Electrical Services Engineering uses the same work stages to define a
project. These are also shown in Table 1.
Stage
Appraisal stage
C1
Strategic brief
C2
Outline proposals
C3
Detailed proposals
C4
Final proposals
C5
Production information
C6
Tender documentation
Tender action
Mobilisation/project planning
Construction
After practical completion
Predesign
Design
C7
Construction
C8
Mechanical and electrical designers may be appointed, either to

provide a full design and calculation service, or to produce a
performance specification for development by an m&e
contractor. This is covered in more detail on page 3. In all cases
there should be clear lines of accountability within the project
team. In a building project this is traditionally determined by the
architect.
Project management is a core requirement to make sure that:
Cost estimates are calculated from a properly defined

specification of what the completed building must provide
this is usually called a functional specification or a
performance specification
contracts for design work, building work and supply of
materials and components are awarded according to the
best value rather than lowest price
decisions regarding variations to the project are made
according to whether they provide functions necessary for
the building to perform in the way the client requires.
ACE Conditions (2002)
Note that the RIBA Plan of Work is being revised and is likely to
be issued in 2007.
Project management techniques particular to building services

are explained in the BSRIAs Project Management Handbook for
Building Services, AG 11/98.
Table 1: RIBA Plan of Work stages.

RIBA Plan of Work
Work
Title
stage
There are three critical success factors for projects to remain

under good control and to increase the likelihood of
providing excellent value for money. These are:
The completed building provides the functional requirements

stipulated in the brief
the cost budget is met
the programme is met
quality levels are achieved
the building can be safely maintained, operated and
decommissioned.
Project inception
The need for a project is determined by business or policy
requirements that are identified and justified well in advance
of design or construction work. This is done through the
business case. One option always open to the client is to do
nothing.
The purpose of the project is to satisfy the requirements
defined by the client according to the business (or policy)
needs. This covers both commercial clients, such as
developers and public sector clients, such as National Health
Service trusts or local authorities. These needs will define:
What the completed project is for

the deadline by which it must be delivered
the maximum amount it can cost
the quality threshold it must reach.
If the business or policy needs are achieved then the client will
receive value for money and the project will be considered a
success. Business or policy needs must not be confused with
achievement of technical specifications (for example,
providing a specified internal temperature in an office space)
which are a means to achieving business needs, not ends in
themselves.
Analysis of claims and litigation in respect of building services
has shown that 45% of successful claims are due to errors in
design concepts and parameters (Griffiths & Armour, 1999).
The importance of fully and correctly understanding a clients
business needs cannot be overstated.
Initial understanding of client needs can be changed by
interpretations made by others (for example the architect or
the surveyor), particularly if the building services engineer
becomes isolated from the client and end-user. It is therefore
important that a client can express its needs directly to the full
design team, including the building services engineer.

BSRIA BG 1/2006

Assembling the project team
The selection of the project team is based on many factors
beyond the scope of this publication. Project team selection has
been adequately covered in many other publications (for
example in Chapter 17 of the Handbook of Project Management,
Gower, 2000).
The selection and internal and external management of the
project team (and the contractual conditions under which that
management must operate) is vital to the success of a project.
Specifically, the early appointment of the building services
engineer can add real value to a project, particularly the
orientation of the building on its site, the layout of the building
and the space planning, and their effects on the operating
efficiency and energy use of the building.
In very general terms, conditions of engagement attempt to
limit exposure to litigation by imposing boundaries of
responsibility rather than fostering a spirit of co-operation that is
essential for a successful project. Great care must be taken by the
client and the clients advisors to ensure that the responsibilities
defined for different members of the team do not leave areas
unaddressed or create areas of overlapping responsibilities.
Before accepting any terms of engagement it is essential that the
client and the specialist and professional team members fully
understand and agree the contents, limitations and respective
responsibilities of all participants. The latest BSRIA publication,
BG 6/2006: A Design Framework for Building Services Design
Activities and Drawing Definitions, provides project teams with a
set of comprehensive pro-formas, completion of which will
determine which member of the project team is taking the lead
on particular aspects of design.
More detail of the appointment of the building services
engineer is discussed on page 3.
Briefing
Briefing can be defined as:
The process by which a client informs others of needs,

aspirations and desires, either formally or informally
the process by which a clients requirements are
investigated, developed and communicated to the
construction industry.
Briefing is an essential and important part of the project process.

It sets the cost and value parameters for the project and defines a
clients requirements and needs. Good briefing is essential for
good design. It will ensure the project team delivers a product
that meets the needs of the client and end user, and delivers a
building that will benefit the client's business interests.
In many projects, the client (who appoints the building team)
will not be the same as the end-user of the completed building.
For example, a university that is building a new teaching block
may delegate the role of client to its internal facility managers,
although the end-users will be the lecturers and students.

BSRIA BG 1/2006
In situations like this, the buildings success will be determined

by the degree to which it meets the needs of the end-users.
The project team, including the services engineer, should find
out as much as they can about the end-users needs. As users
can change, so can the requirements. It is better to have a
consolidated brief from the main client that states the end users
requirements.
Buildings are a major financial expenditure for most clients.
Some poorly performing buildings have been reported in the
post-occupancy evaluation project, PROBE (Post-occupancy
Review Of Buildings and their Engineering). These studies
showed that buildings with poorly performing architecture and
engineering services could create unsatisfactory working
environments. This can have serious consequences for a clients
business.
Briefing, in the context of the building process, is thought of as
solely referring to a client brief. In practice, the briefing process
extends throughout the design stages of a building project. It is
an iterative process involving regular feedback from clients,
advisers, the design team and end users. The brief ideally
should provide everything the design team need to know about
the building the client requires, the site being used and its links
to the local environment.
Good briefing is essential to ensure that the clients needs are
met and that best value for money is obtained. The brief
usually starts as a statement of needs from the client and then
evolves into a consolidated brief for the project.
The statement of needs will usually contain the following data:
The clients business function

the clients business objectives
the structure of the client organisation
the clients perceived need for the project
any relevant historical background
the triggers that have necessitated change
the perceived consequences of failure/risk analysis
the nature of advice needed to progress the project.
The statement of needs is entirely in the hands of the client and

has a profound effect throughout the project. It is important
that all consultants and contractors involved in the project have
seen the statement of needs and understand it.
The consolidated brief would include all of the basic
information contained in the client brief and strategic brief, as
listed above. It would also include specific details of the project
team and proposed building design solutions (in so far as these
have been decided), client requirements regarding issues such as
the attitude to be adopted towards health and safety, the
procurement method to be adopted, and the quality criteria to
be applied throughout the project.

Typical contents include:
Details of the project team

project description
description of the proposed building functions
site location and access details
details of constraints arising from legislation or other factors
total floor areas of proposed buildings
building layouts
proposed number of occupants
details of any special equipment or processes to be housed in
buildings
space requirements for people and equipment
internal and external environmental design conditions
design solutions to be adopted
the required life span of the proposed building and of
individual components
the agreed construction procurement strategy
cost budgets
design and construction programmes.
The consolidated brief develops alongside the proposals from

the project team, including contractors and specialists.
Appointment and duties of the building services

engineer
Appointment
An enquiry for design services might come to the building
services engineer from the client, the architect, the main
contractor or the m&e subcontractor, depending on the nature
of the project and the procurement route. However, the
approach to selection is likely to be based on one of the
following methods.
Appointment on merit, whereby appointment is based purely
on the clients previous experience of working with the
building services engineer. Fees may be calculated according to
a partnering arrangement between the client and the engineer,
or by negotiation.
Competitive interview, whereby some form of specified
presentation must be given. This might be appropriate where
the client has an outline project description and wishes to hear
the designers views before making an appointment. The scope
of services and fee would be negotiated afterwards with the
preferred firm.
Fee tender and qualifications, whereby a designer is selected

based on an assessment of proven technical qualifications and
ability, as well as the quoted fee.
One to one negotiation, whereby appointment is based on
one or more interviews. This is a useful method for getting a
designer on board at a very early stage in order to help the
client consider, develop and define requirements.
Qualifications-based selection, whereby a designer is
selected on quality, such as technical qualifications, previous
relevant experience, and general suitability. Having short-listed,
typically, three companies on this basis, the finalists are
interviewed and a selection made. The scope of services and
fee is negotiated afterwards.
Design duties
Most m&e designers are appointed using Agreement B2,
published by the Association of Consulting Engineers. The
duties within Agreement B2 can be aligned with the RIBA work
stages, as shown in the example in Table 2.
The duties of the building services engineer also reflect the fact
that building services are dynamic systems. The selection of
components, their installation and their commissioning all
influence the system performance. More so than other design
disciplines, building services design is an iterative process, where
initial assumptions about materials and construction methods
may be shown to be incapable of achieving the functional
specification. In these cases, changes to the materials or
components may mean a re-design.
One of the major causes of conflict between building services
engineers and other members of the project team is a lack of
clear understanding regarding the division of responsibilities,
particularly at the interfaces of work done by designer and
installer. One example of this is responsibility for preparation of
the co-ordination drawings. Other areas of conflict include the
degree of detail provided on drawings.
This can cover:
Precise services routes

responsibility for design re-evaluation due to alternative
plant selection and the implication of changes
responsibility for the specification of requirements for
systems commissioning
the preparation of handover material.
Design ideas competition, whereby a designer is chosen

based on design ideas. The design fee is stated in the
competition conditions.
Design submission, whereby a designer is chosen based on a
design submission and the quoted fee. The client would usually
pay for the design submission.
Fee tender, whereby a designer is selected based solely on the
fee quoted for a given project brief and description of service
required.

BSRIA BG 1/2006

Table 2: RIBA work stages and detailed design duties from ACE agreement B2, Normal Services.
C1
Obtain information regarding utility services to the site

Comment on any physical site restrictions
Make initial recommendations regarding technical viability of the works
C2
Visit the site as necessary and gather relevant data and information
Advise the client of the need for any surveys or special investigations, such as occupancy
survey or drainage survey
Consult with utilities and the relevant authorities
Consider alternative outline solutions
Prepare outline reports and sketches in order to develop the brief
Provide an approximate cost plan and advice based on unit area rates
C3
Develop the design of the detailed proposals in collaboration with other consultants
Prepare sketch drawings showing spatial/structural requirements for plant rooms, major
items of plant, major ducts and service routes
Assess preliminary loads for power, heating and cooling
Assess the thermal performance of the building envelope and examine details of solar
control. Prepare initial sizing of heating/cooling plant
Negotiate with utility authorities regarding incoming services
C4
Develop the design and prepare sufficient schematic drawings, schedules and specifications to
allow consultants to finalise their proposals
Assist the lead consultant in co-ordinating the m&e services into the overall design
Prepare a revised cost plan based on unit area rates
C5
F - Production information
Prepare detailed design drawings

Prepare specifications
C6
G & H - Tender
documentation and tender
action
Assemble documents for tender

Comment on tender returns
C7
J, K & L - Mobilisation/
project planning,
construction, practical
completion
Advise the client on the need for the appointment of site staff
Comment on installation drawings and builders work drawings submitted by the contractor
Attend relevant site meetings and make other periodic visits to site
Provide technical advice regarding payment to contractors
Examine testing and commissioning procedures
Examine records of commissioning results
Comment on record drawings and operation and maintenance manuals prepared by the
contractor
Inspect the works on completion and record any defects
C8
C - Outline proposals
D - Detailed proposals
E - Final proposals
Part M - Feedback
Activities not defined by ACE, but important to performance

Fine tuning
User education
Pre-design
Obtain and inform the clients brief

Discuss roles and responsibilities of the project team
Discuss the likely requirement for site staff, such as Clerk of Works, Facilities manager
Design
A - Appraisal stage
B - Strategic brief
ACE
Construction
ACE Agreement B2 Detailed design service*
Handover
RIBA work stage
* Summarised by Fulcrum Consulting
The new BSRIA guide BG 6/2006: A Design Framework for

Building Services provides detailed proformas for clients, design
teams and contractors to agree an allocation of design activities
and deliverables among themselves. In this way the potential
for conflict arising from duplication or omission of design
activities can be minimised.
There are potential areas of conflict between members of a
project team that are particularly relevant to building services
engineers.

BSRIA BG 1/2006
Conflicts can arise when building services designs are based on

superseded versions of the architects layout drawings. This can
occur when the architect continues a design right up to issue of
tender information. If significant changes are made to the
architectural drawings, the building services engineer will have
to modify design during or after the tender process. This gives
the potential for delays and additional cost.
Tender returns from building services contractors may exceed
the clients budget. This usually requires some re-design by the
building services engineers. Responsibility for paying for this
re-design will need to be established.

The design process
Overview
Design is a complex process which involves the activity of
translating ideas, proposals and statements of needs and
requirements into precise descriptions of a specific product.
Design problems are often ill-defined, and their solutions are
often not obvious. There is also rarely one correct answer to a
design problem. Different designers will arrive at different but
possibly equally satisfactory solutions.
Two major features characterise the design process. First, design
tends to evolve through a series of stages at which the solution is
increasingly refined to greater levels of detail, moving from
broad outline through to fine detail. Second, design tends to
contain iterative cycles of activities during which designs and
design components are tested, evaluated and refined. Feedback
loops are therefore an essential component of design. Most
models of the design process involve many feedback and
iteration loops.
In turn, this may lead to a series of subsequent revisions. Such

revisions will have cost implications, which should also be
considered as the overall process is managed and controlled.
The calculations used in this document are taken from the
BSRIA guide BG 30/2003: Practical Guide to HVAC Building
Services Calculations. The use of standard calculations helps
designers to document their design process, which then makes it
easier to make revisions at a later date.
Figure 1: Simple example of a building services design process.
There are many instances where the expertise of the building

services engineer can influence the form of a building,
including:
Suggesting thermal mass for use in passive heating and

cooling systems
optimising fenestration and roof lights to maximise daylight
without compromising thermal performance
suggesting a narrow footprint to allow natural ventilation
modification to floor heights to accommodate sufficient
underfloor or ceiling voids for services distribution
suggested orientation of the building to optimise solar gain
either to minimise to prevent overheating in summer or to
maximise to encourage thermal gain in winter
suggested orientation to use prevailing winds to enhance
natural ventilation
suggested layout of spaces within the building to simplify
services distribution
contributing to structural design options to accommodate
services distribution.
The building services design process

Figure 1 shows an example of the building services design
process, based on the model developed by BSRIA and
published in AG 1/2002: Design Checks for HVAC. This gives a
simple design sequence from a statement of need, through
problem analysis, synthesis, and evaluation to a final solution.
Only some of the feedback loops are shown, but in practice
there are often feedback loops between all tasks and even within
specific tasks.
Based on the model developed in the Design Checks HVAC A Quality

Control Framework for Building Services Engineers, AG 1/2002.
This sequence of design tasks has been developed into a design

map showing the breadth of design choices and considerations
for building services design, (Figure 2). This provides an
overview of the design process to both inform the designer and
to enable design elements to be seen in context. However, the
real design process usually involves a number of iterations with
overlap from one design stage to another. It may be necessary to
revise calculations or modify assumptions at almost any stage.

BSRIA BG 1/2006

Figure 2: HVAC design map.

BSRIA BG 1/2006
2 THE DEMONSTRATION PROJECT

Introduction
Clients functional requirements
The remainder of this report demonstrates the application of the

calculations published in BSRIAs Guide BG 30/2003: A
Practical Guide to HVAC Building Services Calculations, as a
construction project progresses through the two principal stages
of HVAC design: outline and detailed proposals, followed by
final proposals and production information.
The demonstration project has been based on a real building to
make the calculation process as realistic as possible.
Office layout
The office space of the building should generally be open-plan
with the facility to incorporate 3 m-wide by 6 m-deep cellular
offices around the perimeter, as and when required. The
planning grid is 1500 mm, within a 75 m structural grid. The
planning grid will affect how building services components can
be modularised for easiest fit and the spacings to be used
between components. Both structural and planning grids are
shown on the arrangement drawings in Appendix A.
The project starts with a building specification, which is

summarised in the rest of this section. Section 3 then presents
the calculations made during outline and detailed proposals
stages (ACE Stages C3 and C4). Section 4 presents the
calculations made during final proposals stage (ACE Stages C5).
Lighting
The client is keen to optimise the amount of natural daylight in
the office space, but appreciates that the size of the office and
the likelihood of partitions being installed for separate perimeter
offices will reduce daylight effectiveness.
The specification details given below are of the type usually

provided by an experienced client. In many cases, the initial
brief may contain much less detail. In these situations the
building services engineer should meet with the client and other
members of the design team to understand the clients needs for
the building and to discuss how the different design disciplines
can work together to produce the optimum design.
Design occupancy
The client requires an occupancy density of 1 person per 15 m2
of offices. This is within the current guidance of 12 m2 to
17 m2 per person published by the British Council for Offices
(BCO). This allows for approximately 255 occupants for the
client and the same for the tenant.
It is assumed that the appointment of the consulting engineer is

as detailed in ACE Agreement B2 Schedule I Detailed Design
Normal Duties.
The building
The development is on an existing estate, purpose-built for
business use, and is located in a previously undeveloped part of
the estate. The original estate was developed in the 1940s and
has changed ownership three times with various tenants on
short term and long-term leases. The estate is five miles from
the M3 in southern England, and is surrounded by controlled
forestry land.
The development consists of three main areas, shown in
drawing 70206/01 in Appendix A:
Two blocks of offices (each including an internal atrium):

10 220 m2 (110 000 ft2)
laboratories and workshops at the rear of the building:
3345 m2 (36 000 ft2)
reception area and internal two-storey circulation space that
links the office blocks and the laboratories.
The client will occupy one of the two office blocks, plus the
laboratory and workshops. The remaining office block will be let
to a local business. The client has provided the following details
of what it wants and what it needs according to its business
requirements. Where appropriate, comments and references
explain the criteria.
Source of equipment
Systems and components are to be obtained from reliable
sources able to provide matching spares and replacements.
Duty and standby provision
The term duty/standby describes a plant arrangement whereby
duplicate or standby plant is provided to maintain continuity of
service in the event of failure of the main plant or duty plant
items.
This should not be confused with spare capacity, which is an
additional plant capacity over and above the design value. Spare
capacity is typically used to provide a boost in power at start-up
of the system, or to lessen the effects of losing an item of duty
plant.
There are no business-critical activities planned for the office
building (such as data centres, or dealing rooms), so standby
plant is not required. However, as office work would be
compromised by failure of cooling in summer or heating in
winter, the systems should be easily accessible for maintenance
and repair.
Security systems
The building is to incorporate a closed-circuit television system
around the perimeter of the building and at entrances and exits
of the building. A door access system is required for all
entrances and exists to the building.

BSRIA BG 1/2006

Design criteria for building services
Specific design criteria are generated in two ways.
First the client may specify the criteria based on previous
experience of construction. This requires the client to have a
detailed knowledge of the design and construction processes and
will typically arise if the client is a property developer or
frequently involved in procuring buildings.
Second the design team, usually led by an architect or a design
and build contractor, will meet with the client to ascertain the
buildings requirements and translate these into technical design
criteria. Depending on the timing of design team
appointments, these meetings may not always include building
services engineers.
However the criteria are generated, the building services
engineer will need to check that these comply with the
appropriate regulations.
Using the information provided, the architect can design a
suitable building and the building services engineer can start the
preliminary calculations to design the services.
For the demonstration project, many of the criteria specified by
or agreed with the client are based on recommendations from
CIBSE or the British Council for Offices (BCO). For brevity,
the criteria included here focus on the office building that is the
subject of the design process in later sections of this guide.
Occupancy heat load (office)
At 22C (Winter)
Sensible: 90 W/person.
Latent: 50 W/person.
Based on CIBSE Guide A 1999 table 6.1
At 24C (Summer)
Sensible: 80 W/person.
Latent: 60 W/person.
Based on CIBSE Guide A 1999 table 6.1
Small power loads (office)
20 W/m of net office area. The BCO s recommended range
is 15 to 25 W/m2 which allows for future expansion. A higher
figure may be appropriate if the clients business requires lots of
office equipment.
Lighting loads (office)
12 W/m of net office area. This is at the lower end of the
BCOs recommended range of 10 to 25 W/m2.
Fresh air allowance (office)
12 l/s per person based on 1 person/15 m2. This is in line with
the BCOs recommended range of 8 l/s to 12 l/s.

BSRIA BG 1/2006
Infiltration rate
The specification is for the building to achieve the good
practice guidelines for air tightness, with an air permeability
index of 5 m3/(h.m2)
Indoor design conditions
Winter
Offices: 220C20C (This is the BCOs recommendation).
Summer
Offices: 240C20C (This is the BCOs recommendation).
Indoor design conditions should reflect the average condition in
the space and not the temperature at the thermostat. The
figures for this demonstration project were selected by the client
in the knowledge that these will require mechanical
refrigeration. A different client who wishes to use natural
ventilation or thermal mass to regulate summer temperatures
will need to agree different criteria.
Outdoor design conditions
Winter design
-4C db; saturated.
Summer design
29C db; 20C wb.
The chillers must operate in conditions up to 40C in order to
provide cooling capacity in the event of ambient conditions
above design.
Humidity
The client has not provided any specific criteria for humidity
control in this case, but 50% has been used for the purposes of
these calculations as a typical figure for office environments.
Acoustics
The standard for the open plan office spaces is NR38, as
recommended by BCO Guide (2000). In the perimeter zone
where cellular offices may be installed at a later date, cross-talk
must be limited to maintain privacy between adjacent offices.
The BCO recommended noise level standard for these cellular
offices is NR35.
Relevant building construction details

Building grid and floor capacity
Primary grid (generally): 75 m2
floor slab loading capacity: 350 kN/m2
additional capacity for light partitions 10 kN/m2.
(Allowance of 10% of floor area to withstand 75 kN/m2 in a
location defined by the developer.)
Roof drainage
The roofs of the offices, and the circulation space are to
incorporate uPVC rainwater outlets, connecting to rainwater
drainage collection system, to down pipes within service risers.
Roof drainage design to incorporate overflow system to provide
a safety warning.

External wall elevations
External solid wall
Outer brick skin: 105 mm
insulation: 75 mm
air gap: 50 mm
lightweight block: 100 mm
plaster: 13 mm.
Glazed elevations
Polyester powder coated aluminium thermally broken
curtain wall system.
full height (slab to soffit) glazingouter pane: 6 mm clear
air gap:12 mm
toughened inner pane: 6 mm.
(Clear high performance thermal coating to the outside surface
of the inner pane.)
Internal partitions
General internal solid partitions comprise: plasterboard, an air
gap and plasterboard.
A glazed wall to the atrium at the centre of each office block
comprises floor to ceiling double-glazed units.
Insulation: 25 mm
oversite/blinding: 250 mm
felt/bitumen: 5 mm
insulation: 100 mm
cast concrete: 210 mm.
Ceilings (offices)
The perimeter margin is formed with a British Gypsum
mineral-fibre ceiling, to a 1500 mm planning grid,
incorporating 500 500 mm white perforated Tegular-metal
tiles in an exposed fineline grid. The ceiling provides a
minimum of 25 dB(A) sound reduction.
Building dimensions
The dimensions for a building are normally taken from the
approved drawings by the architect. The main dimensions
concerning the office space in this hypothetical project are
given below and shown in Figure 3.
Internal office height: 28 m

raised floor zone (depth): 015 m
ceiling void (height): 075 m nominal height, but allow
02 m beam depth below slab soffit on grid lines,
therefore 055
distance between columns: 75 m
maximum internal distance between walls: 375 m
reinforced concrete floor thickness: 02 m
planning grid: 15 m.
A glazed wall between the offices and the two-storey circulation

space comprises floor to ceiling double-glazed units as per the
external faade.
Ground floor and roof construction

Carpet tiles
raised flooring system (150 mm void)
concrete slab 100 mm.
For calculating heat gain and loss through the external walls,
designers should use the slab to soffit dimension not the internal
office height, as heat will also be transmitted to and from the
floor and ceiling voids.
Figure 3: General office layout and vertical section.

BSRIA BG 1/2006

Calculation topics
The calculation topics covered by in BSRIA Guide BG 30/2003: Practical Guide to HVAC Building Services Calculations and this
model demonstration project are listed in Table 3, below. The references (for example H1) are to the calculation sheets in
BG 30/2003. Table 3 also indicates which topics are covered in which of the two stages of the design process that are demonstrated
in this report.
Table 3: Schedule of calculation topics.
Calculation topic
Outline and detailed

proposals
(section 3)
Final proposals and

production information
(section 4)
Stack effect (H1)

Infiltration (H2)
U values (H3)
Condensation risk (H4)
Heat loss (H5)
Plant heating load (H6)
Radiator sizing (H7)

Boiler sizing (H8)
Flue sizing (H9)
Pipe sizing general (W1)
Pipe sizing straight lengths (W2)
Pipe sizing pressure drop across fittings (W3)
System resistance for pipework index run (W4)
Pump sizing (W5)
Water system pressurisation (W6)

Internal heat gains (C1)
External gains (C2)
Cooling plant loads (C3)
Ventilation fresh air requirements (C4)
Supply air quantity and condition (C5)
Heating/cooling coil sizing (C6)
Humidifier sizing (C7)

Duct sizing general (A1)
Duct sizing selecting a circular duct size (A2)
Duct sizing circular to rectangular ducts (A3)
Ductwork - pressure loss through fittings (A4)
Duct sizing index run (A5)
Fan sizing (A6)
Grille and diffuser sizing (A7)
Air density correction (A8)

Pressurisation of spaces (A9)
Acoustics for building services (new)
Dehumidification (new)
Control valve selection/sizing (new)
Effect of return air temperature on coil duty (new)
Heating plant configuration and load matching (new)
The calculations marked new will be included in a revision of the Practical Guide to calculations.
10

BSRIA BG 1/2006
3 OUTLINE AND DETAILED PROPOSALS STAGE

Introduction
At outline and detailed proposals stage, the building services
need to be developed to a sufficient level of detail to allow
important strategic decisions to be made regarding plant space,
services void depths and types of system.
The concept for the building and the layout of the floor-plate is
as described in Section 2.
For the purposes of this project, only the offices have been
studied and only services for these areas have been considered.
As this demonstration shows the application of the building
services calculations, it is not necessary to produce designs for all
parts of the office, therefore only one floor of one of the office
blocks will be studied in detail. Loads for other floors will be
approximated to give full plant loads, although in practice the
engineer would repeat the calculations for each floor.
The services being covered at this stage are determined by the
calculation topics. The main services being addressed are the
ventilation and air conditioning and comfort cooling systems to
the office areas. This will include central plant as well as
services through risers and distribution throughout the floors.
For the purposes of this project, it has been assumed that toilet
cores will be served with separate supply and extract ventilation
systems. As such they will have no interaction with the office
area air conditioning and ventilation systems.
Work plan and methodology

To execute this stage of the project, the following steps should
be followed in the order shown. The order of some steps can
be changed depending on the nature of a project. The
approach followed by the demonstration project is similar to the
scheme-design phase of most projects.
1. Determine the design criteria. Design criteria would
normally be developed between the design team and the
client. However, some clients who are well versed in the
building process, or have very particular requirements, may
provide the complete design criteria. (For this project the
design criteria have been developed and agreed by the
steering group based on criteria specified by the client of
the actual development on which this demonstration has
been based).
2. Refer to architectural and structural drawings and
specifications of construction elements. Drawings and
specifications will form the basis of all the work covered
here, so it is imperative that they represent the current
information. Drawings must include plans, elevations and
sections for all areas under consideration to provide a
detailed understanding of the building form and
arrangement, and to enable the accurate take-off of
information. Drawings will also provide an understanding
of the construction elements, along with written
information from the architect and other design team
members.
The input drawings used for this work are shown in
Appendix A. They have been based on those provided to
the building services consultant for the development that
was used as the basis of this demonstration.
3. Analyse space, and layout requirements Spatial

requirements are needed for ceiling voids, offices, and
partitions. It is important to understand the internal layout
of the building to assess the viability of likely building
services systems. This may have been done already as part
of the work detailed in Section 2 Building services. For
instance, if there is only a 150 mm ceiling void, it is unlikely
that conventional fan-coil units could be used. Further
discussions may need to be held with the rest of the design
team to provide adequate space for plant and systems.
4. Determine zoning/general arrangements The
arrangement of the floor plan, together with any zoning
required for solar gain patterns, needs to be determined in
order to locate local plant items and lay out central
ductwork and pipework distribution runs.
5. Calculate U values Once the information described in the
two previous sections has been assembled, the first
calculations to be performed should be for the U values.
These are then used as the basis to calculate heat gains and
losses.
6. Carry out condensation analysis The construction
elements need to be assessed to make sure that condensation
will not occur through the structure. If a potential problem
is highlighted at this stage, discussions can be held with the
design team on the best way to overcome it.
7. Calculate fresh air supply rates The initial data should
be included in the design criteria, based on either a quantity
of fresh-air per person, or the volume of the space allowing
a set number of air changes per hour. For this project a
fresh-air rate per person will be used.
8. Calculate infiltration rates through the external
envelope This figure is needed to feed into the heat gain
and loss calculations as warm outside air will infiltrate into
the building in summer and increase the heat gains, and
warm internal air will leak out of the building in winter and
increase the heat loss.
9. Calculate heat gains and cooling loads Using the data
from previous sections, the heat gains will be calculated
using a computer-based calculation package. The majority
of heat gain and loss calculations are performed using such
tools. Figures will be required for each room, along with
the maximum simultaneous load. This will be done in this
case using Hevacomp software although other packages such
as TAS, Cymap and IES can also be used for these
calculations.
10. Calculate heat losses and heating loads As the note
above. These are derived from the worst-case scenario,
taking into account fabric and infiltration losses.
11. Select systems Taking into account the figures produced
in the foregoing calculations, and the physical arrangement
of the building, the next step is to select suitable air
conditioning and/or ventilation systems.

BSRIA BG 1/2006
11
Se See Guide to HVAC Building Services Calculations, BG 30/2003

12. Select/arrange room units Once system has been
selected and the zoning/general arrangements have been
determined, the local plant or room units can be selected.
1. Design criteria
13. Arrange room ductwork Ductwork distribution can now

be run to the various local plant or room units. Sizing will
often be on rules of thumb figures at this stage. In reality,
this may be coordinated simultaneously with the pipework
as both may need to be accommodated close together.
14. Arrange room pipework Pipework distribution can now

be determined for the local plant or room units. As with
the ductwork above, sizing at this stage may be based on
rules of thumb data.
15. Arrange risers In many cases this step may go before 13
and 14, with the ductwork and pipework then being
arranged away from the risers. The earlier calculations will
give indications of the likely sizes of ductwork and
pipework to be accommodated. It may be necessary to
have further discussions with the design team if there is not
sufficient space available. Such issues should be raised at this
stage rather than once detail design is under way, and the
building form largely fixed. This is part of the function of
scheme design. For this project, any assumptions made will
be stated and discussed with the steering group.
16. Arrange services to plant areas Distribution between the
risers and the central plant area can be developed. Again,
this could be done earlier in the process.
17. Select central plant Preliminary plant selections can be
made, based on the data from the earlier calculations. In
turn this will help the designer to compare early plant area
layouts against space availability. This information is also
required by the structural engineer as the weight of the
plant will have a bearing on their structural load
calculations.
Design margins
Depending on the particular case, it may be considered prudent
to add design margins to the calculated building and plant loads.
These margins may be added for a variety of reasons, and may
vary in magnitude. Any such margin should be considered
carefully to understand the potential effects it may have on plant
and system operation, capital costs and energy consumption.
Research work carried out by organisations such as BSRIA and
CIBSE has looked at the reasons for adding margins, as well as
the values adopted and has reported the findings in a report.
Engineering Design Calculations and the Use of Margins (CIBSE,
1998) This report identifies nine different types of margin. The
executive summary of this publication is included in Appendix G.
Only very specific margins have been included in the calculations
carried out in this model demonstration project.
12

BSRIA BG 1/2006
The design criteria are as detailed in Section 2. Apart from the

details of the building fabric and layout, the criteria needed by
the building services engineer to allow outline design are:
Outdoor design conditions (these depend on the geographic

location of the building)
indoor design conditions (these depend on the use to which
the building will be put)
infiltration rate (a figure for infiltration air changes per hour
will calculated based on the specified level of air-tightness)
a fresh air allowance
occupancy level (this will be determined by the client or
based on recognised standards if the information is not
available from the client)
occupancy heat load (this will be determined by the type of
work being done in the building there are standard figures
for different types of occupation if this is not provided by
the client)
small power loads (this will be determined by the client or
available from the client)
lighting loads (this will be determined by the client or based
on recognised standards if the information is not available
from the client)
acoustic standards (this will be determined by the client or
available from the client).
This information may often be produced in the form of room

data sheets. These can be particularly useful on projects with
spaces of many different uses.
2. Building drawings and construction elements

The general nature of the building is described in the following
drawings, together with others illustrating the exact areas. The
original drawing on which these drawings have been based
were provided by the steering group.
Drawing list for scheme design stage
70206/01: Building plan ground floor
70206/02: Building plan first floor
70206/03: Building plan second floor
70206/04: Building elevations 2
70206/05: Detailed work area ground floor.
Samples of these drawings can be found in Appendix A.
The nature of the construction elements are as detailed in
Section 2.
3. Space and layout

The physical arrangement of the space needs to be appreciated
before the system selection process can be started.
The shape or physical characteristics of all areas need to be
carefully considered in determining the heating, ventilation and
air conditioning possibilities, including ceiling height, sizes of
rooms or spaces, depth of any floor or ceiling void, plan area
and the position of risers.
H3

In the case of the demonstration building, the office floor plan is
a large, square, single space, with a smaller square removed from
near the centre. In essence, the space is a square-sided
doughnut. The distance across the room from the external wall
to the internal atrium wall is approximately 153 m at the
widest point.
The building has a large proportion of floor-to-ceiling glazing
on the external faade, with little in the way of solar shading.
This may result in significant heat gains around the perimeter,
depending on the type of glazing used. In this case, there are
no opening windows in the main office area, so fresh air must
be provided mechanically.
From the sketch drawings of the building (Figure 4), it can be
seen that a deep ceiling void of 750 mm has been allowed for
routing of ductwork and pipework services across the office
area. However, this dimension cannot be taken at face value
and the following checks should be made to establish vital
details which can reduce this dimension if:
The dimension is shown from underside of the structural

slab to the suspended ceiling, then the depth of the
structural beams will also need to be allowed for in this
case 200 mm
the structural beams are steel (rather than concrete) then an
additional allowance for fire insulation must be made
the dimension is shown to the underside of the suspended
ceiling, then the thickness of the ceiling system should also
be deducted in this case 50 mm
an allowance must also be made for the depth of any
recessed luminaries in this case another 50 mm above the
ceiling.
These checks can easily be missed and can turn out to be vital
when laying out services, particularly when trying to coordinate cross-over points, and working around immovable
items such as drainage pipework laid to fall.
Figure 4 shows the detail of the ceiling zone, and also shows the
zones within which ductwork, pipework and lighting will be
designed.
Figure 4: Detail of ceiling void.
In other areas some services can use the full 700 mm available
from the ceiling to the underside of the structural slab.
There is also a floor void, much shallower at 150 mm. As with
the ceiling void, it should be established whether the 150 mm
dimension includes the depth of the finished floor construction
or is a clear dimension. While this space will be used for small
power, data and telephone cabling, these services are not
covered in this guide.
Each floor plate has been provided with two vertical risers
linking all floors to the roof where the main plant is to be
situated. The risers are sized to accommodate ductwork and
pipework services. However, all air-conditioning systems will
require more space to house the larger ducts than fresh air only
ventilation systems, with the heating and cooling functions
being carried out by local plant such as fan-coil units.
System selection will also be affected by the relationship of the
building to its surroundings. Nearby roads can create problems
of noise and pollution if the building services strategy requires
openable windows. In this case, the building is near to a road
test track, although otherwise in a rural location. This may
have been a significant factor in deciding to design a fully airconditioned building with no opening windows.
4. Zoning/general arrangements
The demonstration building lends itself to zoning on a floor-byfloor basis, and permit partial occupancy and different timings of
occupation, such as normal office hours for some departments
and weekend working for others. Furthermore, the two risers
allow services to be arranged in two zones per floor, with each
riser serving half the floor.
Zoning considerations to deal with solar gains are not a strong
consideration as they might be with some other systems as the
fan-coil units have their own local control.
5. U values H3
Computer program details
Once the architect has provided all the necessary details and the
figures are confirmed and agreed with the design engineer, then
the information can be used in the computer program. It is
important to spend the time to get these details right as the
calculations that follow and subsequent service design will rely
on them.
Designers should check also that any assumptions made by the
program are correct and relevant to the building details.
The dimensions shown for the pipework and ductwork zones

are the maximum available at the points where the services cross
each other and the structural beams.
Thermal bridges
A layer that consists of more than one fabric element will have
bridges, for example a layer of insulation with timber studs, or
concrete blocks with mortar joints as shown in Figure 5. The
diagram shows concrete blocks and mortar in a schematic
arrangement to allow the easy calculation of the area of the
joints around four blocks. In practice the blocks would be laid
in a running bond, with alternate rows of blocks offset from one
another.

BSRIA BG 1/2006
13
H3

Figure 5: A Thermally bridged fabric element.
Table 4: Data for the concrete roof construction.

Fabric element
Thickness
d (m)
Felt/bitumen
Glass fibre quilt
Cast concrete
0005
0100
0210
Thermal
conductivity
(W/mK)
0500
0040
113
The first step is to calculate the resistances for each fabric

element. This is done by dividing the thickness of the fabric by
the associated thermal conductivity:
R=
Figure 6 shows two bridged layers of a construction. As heat

that passes through the two layers can do so through four
different layer combinations, the overall U value of the
construction will depend on the proportions of each different
fabric in each layer.
When using a computer package to calculate U values of
constructions with bridged layers the designer should take care
to check that the data is correct for each bridge and layer and
that it is entered correctly.
d
D
The resistances for the roof construction are shown below. For
simplicity the fabric element initials have been used for the
subscript of each Resistance (R):
Table 5: Elemental resistances for roof construction.
Fabric element (from
outside to inside)
Reference
symbol
Resistance
R (m2K/W)
External surface
Res
0040
Felt/bitumen
Rf
0010
Glass fibre quilt
Rg
250
Cast concrete
Rc
0186
Internal surface
Ris
0117
Figure 6: A Two-bridged fabric element.
The basic formula for calculating the U value of an element or

structure can be calculated from:
1
U =
R
total
Rtotal is the sum of all the resistances for each layer (not
forgetting the internal and external surface resistances). This
roof construction has no bridged layers so there is no need to
deal with proportions of different elements in the same layer
and the different paths though which heat can travel.
Rtotal = Res + Rf + Rg + Rc + Ris
Rtotal = 004 + 001 + 25 + 0186 + 0117
Rtotal = 2853 m2K/W
Roof example
For this model project the architect has provided most of the
fabric details but the U values still need to be determined. The
following data has been supplied for the construction of the
concrete roof:
So the U value for this roof construction is:
All other U values for the building have been determined using
a computer package with the following results:
14
Internal surface resistance: 0117 m2K/W

external surface resistance: 004 m2K/W.

BSRIA BG 1/2006
U roof =
1
2 853
= 0 351 W/m K
H4
C4
H2

Table 6: U-values for building fabric.
Construction
7. Fresh air supply rates C4

2
U Value (W/m K)
Ground floor
0415
Standard wall
0396
Double glazing
2801
External window clerestorey
2967
Reinforced-concrete ceiling
3268
Single glazing
4270
Double-glazed roof lights
2967
Roof
0351
The design criteria state that a fresh air allowance of 12

litres/second per person l/s/p should be used. To obtain the
total fresh air supply rate, the occupation density must be taken
into account.
Total floor area: 1400 m2 3 floors = 4200 m2
Occupancy density: 1 person/15 m2
Therefore, occupancy:
420015 = 280 people
Total fresh air supply:
Note: recent changes in Part L of the Building Regulations mean

that a building with these U-values would no longer gain
approval.
6. Condensation analysis H4
Summary of condensation calculation
Condensation calculations are important to identify where there
is risk of moisture forming within the structure of the building
(interstitial condensation). Where risk of condensation is
identified then steps need to be taken to alter the composition
of the building fabric.
Condensation calculations are complicated, especially for multilayer constructions. The outputs of the calculation are the
vapour pressures at each node through the construction, based
on the internal and external conditions, the thickness of each
layer and the thermal and vapour resistivities/conductivities of
each material used in the construction. These are compared
with saturated vapour pressures derived from CIBSE tables.
Where the calculated vapour pressure is greater than the
saturated vapour pressure, then condensation will occur.
If the calculation indicates that condensation will be present,
then the design engineer needs to feed this information back to
the design team and to the client. Two courses of action can
then be taken:
The team can decide to alter the building fabric, for

example by introducing more insulation, or a material with
a higher moisture resistivity, to prevent condensation
forming
the team can introduce a vapour barrier to contain the
condensation and alter the design of the building fabric to
remove the moisture that condenses.
A detailed example of the condensation risk calculation for the

roof structure of the demonstration project is included in
Appendix F. However, in practice, these calculations are
usually done by a computer package, and responsibility for
analysing condensation risk normally lies with the architect.
280 12 litres = 3360 l/s

+5% for duct leakage
3360 105 = 3528 l/s
Fresh air supply/floor:
35283 = 1176 l/s
For scheme design purposes, a notional 5% has been added to
the fresh air supply system to allow for air leakage from the
ductwork. More accurate figures can be obtained from
DW 133 and DW 144.
A typical extract rate for such a scheme would be 90% of the
supply rate. The excess supply air would provide a slight
positive pressure within the office areas, as well as some makeup air for toilet extract systems.
The ventilation system does not form part of the fire strategy
and would be automatically turned off by the fire alarm system.
8. Infiltration rate H2
The calculation for infiltration rate is used to arrive at the
number of air changes per hour through the external envelope
of the building. It is also used in the heat gain and loss
calculations.
The specification to achieve is the good practice guide for an air
leakage index of 5 m3(h.m2) (for a balanced, mechanically
ventilated office building. See extract from CIBSE TM23
below).
The equation to estimate the infiltration rate is:
I=
1
20
air leakage index (air changes per hour)
Where:
I = the resultant infiltration rate
S = the surface area of the external envelope of the room,
floor or building
V = the internal volume of the room floor or building.
For the purposes of the ground floor outline design, the floor
plan can be assumed to have a perimeter zone of cellular offices,
which will attract all the infiltration. These offices can be used
to calculate the relevant values of S and V.
BSRIA BG 1/2006
15
H2
C1
C2
C3

There are 23 perimeter offices of 3 m 6 m, and three offices of
45 m 6 m. These numbers are arrived at by comparing the
area of perimeter infiltration on drawing 70206/05 with the
hypothetical layout of cellular offices on drawing 70206/06 in
Appendix D.
In this calculation S is the faade area and floor area is not
included. If the infiltration rate was being calculated for the
topmost floor, then the roof area would form part of this. But for
the ground floor, the length of faade of the perimeter offices is:
15 15 + 25 15 + 15 15 + 10 15 = 975 m
Table 7: Air leakage good practice and best practice figures.

Building type
Air leakage index m3/(h.m2)

Good practice
Best practice
Dwellings
150
80
Offices (with
balanced mechanical
ventilation)
50
25
Superstores
50
20
Building type
Air permeability m3/(h.m2) at 50 Pa

Good practice
Good practice
The slab to soffit height of the ground floor is 37 m. So the

exposed facade area is 361 m2.
Dwellings
100
100
V is the volume of the 26 perimeter offices into which infiltration

is occurring. Their floor area is 495 m2. With a slab to soffit
height of 37 m this gives a volume of 18315 m2.
Offices (with
balanced mechanical
ventilation)
35
35
Superstores
30
30
The equation for infiltration now gives:

I=
1
20
361
1831 5
I = 005 air changes per hour

Note that this calculation is sensitive to the arrangement of the
perimeter cellular offices. If, instead of being 3 m wide by 6 m
deep, the perimeter offices had been assumed to be 6 m wide by
3 m deep, then the perimeter volume would have been
approximately halved and the infiltration rate approximately
doubled to 01 air changes per hour.
Furthermore, the design criteria should allow for the fact that
infiltration will tend to increase with the age of the building. In
some cases it may be considered appropriate to allow some
capacity in the plant to allow for this, or to use a greater
infiltration figure for the design calculations.
For the purposes of this demonstration project, the design process
has assumed an air change rate of 02 air changes per hour. This
is in line with the following statement from CIBSE TM 23: The
above relationship relates to average standards of construction and average
weather conditions. For typical office buildings, an air leakage index of
10 m3/(h.m2 ) at 50 Pa implies an average air infiltration rate of about
02 air changes per hour for average wind speeds.
For comparison, the good practice and best practice figures
quoted in CIBSE TM23 are in Table 7.
9. Heat gains and cooling loads C1
C2 C3
The heat gain calculations were carried out using a proprietary

software package. Resultant air temperature was used in the
calculations rather than air temperature. This is the value
generally used in CIBSE guidance.
For the purposes of these calculations, each floor was
considered as a single room, without any individual offices
being created. This approach is generally sufficient for scheme
design as the layout is often subject to change at this stage of a
project. Detailed heat gain calculations are therefore usually
carried out at the detailed design stage to avoid wasting time.
The data is included in Appendix B, but the results are
summarised in Table 8 and Table 9, rounded to one decimal
place.
The heat gains to each room are shown in the first table. The
load from the fresh air will be dealt with at the central plant,
and not by the room units. While the values shown below
therefore exclude fresh air supply they include infiltration.
Table 8: Floor-by-floor heat gains.
Room/area
G01 ground floor
F01 first floor
S01 second floor
Sensible
gain kW
669
754
1035
Latent
gain kW
56
56
58
Total gain
kW
725
810
1093
However, the total heat gain to the building cannot be

calculated by simply adding together the room totals. The
results must be checked to see when the maximum
simultaneous load occurs. For the three rooms under
consideration, the results are summarised in Table 9.
16

BSRIA BG 1/2006
H5

Table 9: Summary of simultaneous heat gains across floors.
11. System selection
Time
Room
Sensible
gain kW
Latent
gain kW
Total
gain kW
1400
G01 ground floor
631
56
687
F01 first floor
712
56
768
S01 second floor
1022
56
1078
Simultaneous
total
2365
168
2533
G01 ground floor
661
58
719
F01 first floor
746
58
803
S01 second floor
1035
58
1093
Simultaneous
total
2442
173
2615
G01 ground floor
669
56
725
F01 first floor
754
56
810
S01 second floor
1001
56
1057
Simultaneous
total
2423
168
2591
1500
1600
Table 9 shows that the peak simultaneous load for the building
occurs at 15.00 h, and these are the values that should be used as
the starting point for plant load selections.
10. Heat losses and heating loads H5

A similar approach to that detailed above under heat gains and
loads should be adopted for the heat loss calculations.
The heat loss figures are made up of fabric and infiltration
components, and the results for the three rooms are shown in the
following table. As with the heat gains, the fresh air load has been
dealt with at the central plant and not added to the room loads.
As the heat losses are normally calculated ignoring solar gain, the
loads can be assumed to be simultaneous and simply added
together to get the building total. A full printout of the heat loss
results is contained in Appendix C.
Table 10: Floor-by-floor heat losses.
Room/area
Fabric loss
kW
Infiltration
loss kW
Total loss
kW
G01 ground floor
387
176
563
F01 first floor
238
175
413
S01 second floor
624
183
807
Building total
1249
534
1783
In terms of this demonstration building, selecting the system

is not an issue as a fan-coil system was specified by the client.
This sometimes happens when the client is a developer and
wishes to maintain consistency across a portfolio of
properties. However, in many instances the building services
engineer, in consultation with the rest of the design team,
will recommend a system for the building and use this in his
scheme and detailed design reports.
To select a system, it is usual to draw up a rating and
weighting matrix against key selection criteria. These criteria
would usually cover the cost of the system (preferably wholelife cost rather than just purchase cost), the comfort provided
by the system, the type and degree of control available to the
users of the building, the space requirements for plant,
distribution systems or intrusion into occupied areas, and the
energy considerations required by Building Regulations.
Table 11 on page 18 is based on BSRIAs publication
FMS 1/97 Guidance and the Standard Specification for Ventilation
Hygiene, and it indicates most of the contemporary systems
available in the air conditioning market and assesses them
against a number of selection criteria. Guidance on the
flexibility of different systems has also been included as this is
an important aspect of system selection
There are other references to comparisons between different
systems, including:
AG 11/98 Guide to Project Management for Building

Services
GPG290 Ventilation and Cooling Option Appraisal (A
Clients Guide)
GPG291 A Designers Guide to the Options for Ventilation
and Cooling
GIR85 New Ways of Cooling
GIR30 Energy Efficient Office of The Future.
The last four publications are available through Action

Energy (www.actionenergy.org.uk).
The effectiveness of chilled beams will vary depending on
whether they are active (such as incorporating a fan to
enhance air flow) or passive (purely reliant on natural
convection).
In terms of the physical characteristics of the building, fancoil units are a good solution as they can cope well with both
open plan and cellular offices. Although the floor plan for
the demonstration project is entirely open plan, the client has
stipulated that the space is able to be converted into
perimeter offices. This functionality has to be allowed for in
the system selection. The fan-coil units can be located in the
ceiling void and arranged to accommodate any future
partitioning with the minimum of changes.

BSRIA BG 1/2006
17

Table 11: Summary of HVAC systems.
Constant
volume (CV)
Variable air
volume (VAV)
Good
but
limited
Good
but
complex
Flexibility
Duct
Occupied area
Plant room
Maintenance costs
C02 emission kgm2/y
Energy efficiency
Noise level
Control
System type
Space requirements
Air distribution
System performance
Low
Very
good
Good to
average
No data
Low to
average
High
None
High3
Low5
Low
Very
good
Very
good1
401
Average
to high
High
None
High
Medium6
Fan-coil units
Good
Can be
high
Fair to
good
Average
50
High
Low
None or
moderate
Moderate
High7
Chilled beams
Good
None
See note4
Very
good
No data
Low to
average
Low
None
None2
High7
Chilled ceilings
Good
None
See note4
Very
good
No data
Low
Low
None
None2
See
note8
Displacement
ventilation2
Good
Very low
or none
Good
Very
good
No data
Average
Low
None or
moderate
Moderate
See
note9
Room-based
heat pumps
Very
good
Can be
high
Good
Very
good
No data
Average
to high
Low
None or
moderate
None2
High7
Split systems
Local
only
High
Poor
Poor
75
Average
to high
Low
None or
moderate
None
Low10
Variable
refrigerant flow
(VRF)
Good
Can be
high
Fair
Good to
average
50
Average
to high
Low
None or
moderate
None
Medium1
1
Notes:
1
System fitted with variable speed fan.
2
No ductwork is required although there is likely to be a separate ducted ventilation system.
3
Space requirement for ductwork is high when used as a single zone air-conditioning system as opposed to a ventilation system for
partially centralised air/water systems.
4
The Quality of air distribution is difficult to categorise as it will be influenced by the type of ventilation system installed.
5
Would need terminal reheat resulting in poor energy consumption
6
Additional terminal units can be added without disruption elsewhere in the building; however, it may be difficult to fit extra units
due to depth of ceiling void required for distribution ductwork.
7
Provided distribution pipes are run locally with valved connections installed, additional units can be added to suit room layouts
without disruption.
8
Limitations on cellularisation due to limited cooling load.
9
Limits on cellularisation, and cooling loads limited (unless combined with chilled beams using swirl-type floor supply ventilation
terminal units)
10
Would need additional space externally for extra condensers and routing of pipes to outside. Extra power supplies would also be
required.
11
Extra terminal units would require the system to be re-charged with refrigerant.
18

BSRIA BG 1/2006
A1
A2
A3

With a fan-coil system, there is a choice to be made between
having all the cooling done at the fan-coil unit (comprising
both sensible and latent cooling), or having central cooling done
at the air handling unit with sensible cooling only at the fan-coil
units. There are advantages and disadvantages to both these
strategies. The pros and cons of each are outlined in Table 12.
Table 12: Local versus central cooling.
Advantages
Cooling at fan-coil
units only
Cooling at AHU
and fan-coil units
Simpler cooling
Little or no
regime
Lower overall
cooling load for
the building
Lower energy
costs for cooling
Disadvantages
Greater
consequences
from fan-coil unit
failure (particularly
in cellular offices)
Condensate
drainage has
installation and
maintenance costs
condensation at
fan-coils
Smaller cooling
loads for
individual fan-coil
units
Cooling capacity
greatly affected
by cooling coil
failure within the
Air Handling
Units
Higher energy
costs for cooling
The option of having all the cooling at the fan-coil units gives a
much simpler regime by which air is heated and cooled and
provided at the desired room condition.
For the purposes of this demonstration project, the design team
has concluded that the economies in running costs from having
all the cooling at the fan-coils will outweigh the installation and
maintenance costs of condensate drainage, and it has been
decided to proceed with this option. Drainage runs from all the
fan-coil units to the risers will need to be designed, but this
calculation is not included in this guide.
Fan-coil units are very flexible in that a very wide range of sizes
and duties is available to suit almost any heat gain load likely to
be encountered in a normal office environment. The units are
generally viewed as reliable, and a degree of redundancy is
provided by the number of units installed. In most cases, if one
unit should fail, the loss in cooling capacity is small compared to
the remaining capacity available.
Similarly, with so many units installed, it is normal to carry one
or two units as spares so that any unit that fails can be quickly
replaced. This strategy causes minimum down-time and robust
reliability for reasonable costs.
Fan-coil units are controlled in one of two ways:
Altering the water flow through the heating and cooling

coils with control valves
altering the air flow though the duct with actuated
dampers.
Waterside control is usually thought to be more energy

efficient, but can introduce problems of maintaining the valves
and can be more difficult to commission.
12. Room unit selection

Fan-coil units are available to suit most typical applications. In
the case of the building being studied, the fan-coils are to be
used in two ways to serve potential perimeter offices and for
general open-plan internal areas.
With the variety of sizes and duties available, it is a safe
assumption that appropriate units can be found to meet the
individual heating and cooling loads for this building, but as
larger fan-coil units can be noisy, this aspect should be checked
against the noise specification as part of the detailed design.
When checking the sourcing of fan-coil units, the voltage rating
of imported models should be checked to make sure they are
properly rated for 230 V.
Similarly, the supply and return grilles for the supply air from
the fan-coil units and the extract air into the ceiling plenum are
usually selected at the final proposal stage. The principal of
using such equipment is well proven in many installations, and
it is a safe assumption to make at this stage that they can be used
successfully on this scheme. However, the client may wish to
see examples of grilles at outline or detailed stages to help
visualise the finished building. At this stage a number of
possible suitable examples could be provided.
Where steel beams are used, and ductwork and other services
are required to pass through the pre-formed holes, coordination of the grilles with the beams is very important, and
should be considered at this stage of the project (see BSRIAs
Services Co-Ordination with Structural Beams Guidance for a
defection-free interface, IEP2/2003).
13. Room ductwork ground floor A1 A2 A3

With the system selected, fresh air is supplied around the floor
plan, based on the occupancy requirements detailed above. The
distribution arrangement is shown on drawing number
70206/06 in Appendix D. The schematic for the ventilation
system is shown in drawing 70206K/03 in Appendix D.
Approximate ductwork sizing is required at this stage for a
number of purposes:
To ensure that services can be accommodated in the space

available
to size and select plant
for discussions with local authorities and the fire officer.
For the floor plan being studied, the air flow calculated above is
1176 litres or 1176 m3/s, for the whole floor. As there are two
risers, and the ductwork has been arranged to serve
approximately half the floor from each riser, the air flow into
one riser at ground floor level will be 0588 m3/s.
Taking an average pressure drop of 1 Pa/m (System Features,
Table 1, Rules of Thumb, UK 4th edition, BSRIA BG 14/2003),
together with volume flow, an approximate size for ductwork at
the entry point to the floor can be determined.
BSRIA BG 1/2006
19
A1
A2
A3
W1

From Figure 4.2, CIBSE Guide B3 2002, a volume flow rate of
0588 m3/s, at a pressure drop of 1 Pa/m, results in a circular
duct size of approximately 370 mm diameter, with air velocity
of approximately 55 m/s. As a cross check, this velocity is
within the 3 6 m/s recommended for a low velocity system.
With typical chilled water temperatures for the system of 60C

flow and 120C return, the temperature difference across the
circuit is 60C. This temperature difference is one factor that
determines the rate at which heat can be extracted from the
space when cooling is required.
Although circular ducts are more efficient from the point of

view of air-flow, and may be cheaper to purchase, a rectangular
duct may need to be chosen to fit within the ceiling void. If
this is the case, then a conversion needs to be made from an
equivalent circular duct to a rectangular duct. As circular and
rectangular ducts have different flow characteristics (rectangular
ducts give rise to more turbulent air flow), a direct arithmetic
conversion of cross-sectional area cannot be done.
The formula Q = m
& Cp T links Q (heat gain 873 kW) with
m
& : (mass flow rate of chilled water) and T (temperature
difference of flow and return 6OC) via Cp (specific heat capacity
of water 42 kJ/kgK). This formula can be used to determine
the mass flow rate of chilled water.
Calculation sheet A3 explains the method for converting

circular to rectangular ducts, using the CIBSE Guide C
Table 4.40. From this table, the diameter of 370 mm is not
shown, so the nearest quoted diameters should be used (366
mm, 381 mm or 385 mm). These give rectangular duct
dimensions of 450 mm 250 mm, 400 mm 300 mm, and 350
mm 350 mm respectively.
To provide more efficient air-flow, designers should choose an
aspect ratio nearest to one, in other words, the duct that is
closest to a square. In this case this would be 350 mm
350 mm. That said, dimensional constraints may force the
designer to choose a flatter, wider duct.
In addition to circular and rectangular ducts, flat-oval ducts are
also available, but are not considered here.
The return air path is intended to be via stub ducts with
bellmouth openings within the ceiling space at each floor,
adjacent to the riser ducts. These will need to be carefully
placed to integrate with any fire curtains, and all other aspects of
the fire strategy developed for the building.
14. Room pipework ground floor W1

General
A similar approach to that used for ductwork sizing can be used
for sizing the pipework throughout the ceiling void, and for the
runs back to the chiller and boiler plant on the roof.
Both risers will be used to distribute the chilled water and low
temperature hot water pipework, with each serving
approximately half the ground floor. Similar procedures would
be used to size pipework for the other floors. This layout is also
shown on drawing 70206/06 in Appendix D.
Chilled water
For the ground floor, the approximate maximum heat gain has
been calculated as 725 kW (see table of room heat gains in
Table 8).
However, as the fresh air cooling load is being dealt with at the
fan-coil units, this must be added for pipe sizing. Therefore,
the total chilled water load for the ground floor is 725 kW +
148 kW, the 148 kW being one third of the fresh air load for
the building, 4442 kW (see section on chillers, page 22).
20

BSRIA BG 1/2006
873 = m
& 42 6
87 3
426
&
= m
m
& = 346kg/s
There are two risers feeding the ground floor, so the chilled
water flow per riser to the ground floor is therefore 173 kg/s.
Assuming an average pressure drop of 200 Pa/m (System
Features Table 1, Rules of Thumb, UK 4th edition, BSRIA BG
14/2003), and by using the flow rate calculated above, the size
of pipework required to meet the chilled water flow to half of
the ground floor can be found.
From Table 4.16, CIBSE Guide C 2001, a volume flow rate of
173 kg/s at an approximate resistance of 200 Pa/m requires a
pipe size of 50 mm. The actual pressure drop for this pipe at
this volume flow rate is 170 Pa/m, and the velocity of water
through the pipe is approximately 08 m/s. Both this pressure
drop and velocity are comfortably within the typical ranges of
100 300 Pa and 075 15 m/s. However, selecting the next
smaller size pipe 40 mm gives a pressure drop of 580 Pa/m,
well above the preferred range, and so would not be suitable.
The schematic for chilled water services is shown in drawing
70206K/01 in Appendix D.
Other considerations to take into account as the design develops
include:
Avoiding frequent changes in pipe diameter to simplify

procurement and installation
offsetting the reduced cost of a smaller pipe with the
increased pump size and ongoing energy cost arising from a
higher pressure drop along the pipe.
Low temperature hot water

The approximate sizes of low temperature hot water (lthw)
pipework can be determined in the same way as the chilled
water services.
The approximate maximum heat loss for the ground floor has
been calculated as 563 kW (see page 17).
With typical low temperature hot water temperatures of 80C
flow and 70C return, the temperature difference across the
circuit is 10C. Engineers often use temperatures of 82C flow
W1
They have no special relevance. However, if exposed heat

emitters are being used, such as radiators, then the surface
temperature of the emitters may be a limiting factor for the
safety of the occupants.
Chilled water
A similar exercise to that described above can be done for the
risers. From the figures in the section heat gains and cooling
loads on pages 16, the maximum simultaneous load for all floors
is 30587 kW from 26145 kW heat gain + 4442 kW fresh air
load. This equates to a flow rate of 121 kg/s using a flow and
return temperature difference of 6OC, or 605 kg/s at each riser.
A lower return temperature could be used, say 60OC, which

would reduce the mass flow rate required through the
pipework. However, flow rates that are too low can cause
problems during commissioning.
From Table 4.16 in CIBSE Guide C, as above, an approximate

pipe size for the riser would be 80 mm. At the design flow rate,
this pipe has a resistance of just over 200 Pa/m and a flow
velocity of 12 m/s.
and 71C return, but these are just Celsius conversions of the
old Fahrenheit temperatures of 180F flow and 160F return.
& Cp T is used to calculate the flow rate

The formula Q = m
for the low temperature hot water. As noted above, Q is 563
kW.
563 = m
& 42 10
563
4 2 10
&
= m
m
& = 134kg/s
Dividing this Figure by 2 gives a flow rate to the ground floor

from each riser of 067 kg/s.
Assuming an average pressure drop of 200 Pa/m, and the flow
rate detailed above, the size of pipework can be found from
Table 4.16 in CIBSE Guide C, 2001. The most suitable size is
32 mm. The schematic for heating services is shown in drawing
70206K/02 in Appendix D.
Similarly, the pipework size for the total flow to both risers in
the block would be 100 mm, giving a resistance of 200 Pa/m
and a flow velocity of 15 m/s. As mentioned earlier, the sizes
of all these runs can be reviewed at detail design stage for
energy or cost saving benefits.
Low temperature hot water
The total flow required into each riser, to serve half of all three
floors, can be found by summing the heat losses for each floor,
and dividing by two.
The total heat loss for all floors, as detailed in Table 10 is 1783
kW. Using the formula Q = m
& Cp T, the flow rate will be
386 kg/s. Therefore, the flow to each riser will be 193 kg/s.
From CIBSE Guide C, the pipe size for the whole block will be
65 mm (giving a resistance of 160 Pa/m and flow velocity of
10 m/s), with a 50 mm service to each riser (giving a resistance
of 160 Pa/m and a flow velocity of 10 m/s).
15. Vertical risers

General
As detailed above, two risers running the full height of the
block have been provided. These have been positioned so as to
allow the floors to be split into two halves. This greatly helps
the distribution of both ductwork and pipework services.
Ductwork
For the ductwork in the risers, the calculation starts with a flow
rate for all three floors, for example, 0588 m3/s 3, giving
1764 m3/s. From Figure 4.2 of CIBSE Guide B3 2002, this
flow rate at 1 Pa/m pressure drop requires a duct of
approximately 530 mm diameter.
Again, if a rectangular duct is required, the conversion method
must be used. Table 4.40 in CIBSE Guide C does not quote a
diameter of 530 mm, so the nearest equivalents are used instead.
These are 537 mm and 522 mm, giving rectangular duct
dimensions of 600 mm 400 mm, and 500 mm 450 mm
respectively. A similar sized duct will emerge from the
calculations of each riser with extract air.
Ductwork from the air-handling units will be sized in a similar
way, based on the total volume to both risers (3528 m3/s). The
circular duct diameter is 720 mm, with a rectangular equivalent
of 650 mm 650 mm, or 700 600 mm. The size to be
chosen will depend on the size and shape of discharge point
from the air handling unit. An allowance of 50 mm should be
made on ductwork dimensions for thermal insulation.
16. Services to plant areas

Plant location
Many aspects of this model demonstration project were
determined by the client in the client brief. For example, the
type of air conditioning system to be used was stipulated by the
client, reflecting their experience in the commercial property
market.
The location of the central plant areas were also determined by
the client. The engineers were responsible for laying out their
plant to best advantage, but within the designated areas. Many
factors should be considered such as where to place plant.
Air movement Items such as air-cooled packaged chillers and
dry air coolers must be positioned to allow adequate air
movement around them for cooling as otherwise the duty may
be reduced.
Maintenance Adequate space must be provided for safe
maintenance of the equipment and eventual replacement.
Weight The equipment must be located where its weight can
be supported. The weight in use must also be considered, and
not just the weight of the equipment itself.
Services Allowance must also be made for connecting services
in and out of the plant. For example, large air handling units
will normally have large ductwork connected to it, and this may
require large turning radii.

BSRIA BG 1/2006
21
H6
H8

Visual Large plant may have a visual aspect, and warrant
discussion with the client and architect. The positioning may
also be subject to the planning process.
Locations of fresh air intakes and exhaust outlets to avoid recirculating exhaust air.
Acoustics Sound levels from plant and equipment.
For a fuller discussion on the factors affecting space allowance
for plant, see BSRIAs, AG 1/2002: Design Checks for HVAC,
pages 16 and 17.
17. Central plant selection H6 H8

Fresh air handling unit
There is one fresh air handling unit for the block distributing air
via two vertical risers.
3
The fresh air supply air handling unit has a duty of 3528 m /s.
From earlier calculations, the ductwork has been sized using a
pressure drop of 1 Pa/m. From drawing 70206G/21, the
approximate length of ductwork (assuming that the ground
floor far riser is the index run) is 90 m. This gives a pressure
drop through the straight ducts of 90 Pa.
There are also approximately 20 and fittings in the ductwork
run (such as bends, control dampers, fire dampers, offtakes and
changes in duct section) which will provide resistance to the
airflow. At this stage of the design, when detailed layouts are
not known, it is usual to take a rule of thumb figure of 10 Pa
pressure drop through each fitting. This adds a further 200 Pa
resistance.
The allowance for pressure drop through the air handling unit is
estimated on the basis of the installed components. Typically
this might add up to 600 Pa. However, at scheme design stage,
it is normal practice to quote the external static pressure to the
manufacturer, for example 300 Pa along with the required
volume flow rate. This is sufficient for plant selection at this
stage for allocation of space, but must be calculated properly at
detailed design stage.
The selection criteria to be used are:
Volume flow rate of 3.528 m3/s
External static resistance of 300 Pa
The components are:
An inlet louvre
a frost coil/preheater
a panel filter
a bag filter
a heating coil
a fan.
Note that no cooling coil is specified as all the cooling is being

done at the fan-coil units. No heat exchanger is specified to
simplify the calculation of mixed air temperature at the fan-coil
units.
22

BSRIA BG 1/2006
Using manufacturers selection data, a typical unit can be

selected. A little extra space can be added for flexibility, as
commitments made to the rest of the design team for space at
this stage can be difficult to reassess later.
Chillers
Two dual-circuit packaged air-cooled chillers have been
provided for each office block. This gives a good degree of
flexibility while keeping capital and maintenance costs to
acceptable levels.
The total load for the chillers is calculated by adding the
building cooling load to the fresh air cooling load. The
building cooling load was calculated on page 16.
The fresh air load has been calculated using a simple rule of
thumb, which should be sufficient at this stage. The method
used is:
& h
Qfresh air= m
Where m
&
= v& (density of air at 20C, kg/m3)
= 3528 1.2
= 423 kg/s
h is the difference in enthalpy between the outside air
condition and the supply condition. This supply condition can
be taken as the room condition minus 20C for heat gain from
the fan motor. This is illustrated on the psychrometric chart in
Figure 7.
So:
Qfresh air = 423 10.5
4442 kW
Figure 7: Psychrometric chart for sizing chiller plant.
H6
H8
H9

Chilled water plant
Duty = Building cooling load (from page 16) 26145 kW
Fresh air load 4442 kW
Chilled water plant total 30587 kW
Chiller plant using refrigerant (R134a) has been selected, with
multiple compressors. The advantage of multiple compressor
plant is to provide some continuity of cooling in the event that
one compressor fails. Each chiller has been selected to provide
approximately 60% of the scheme design load, giving
approximately 90% design capacity in the event of a single
chiller circuit failure.
good degree of control with high operating efficiencies in all

conditions. If a boiler size of 112 kW is selected, then three
boilers can supply all the heat loss load, including the pre-heat
factor. Because the maximum heating load will only be
required on a few days of the year, if one boiler does fail there is
only a slim chance that the heating load on any particular day
could not be met.
The use of atmospheric burner units ensures that the available
gas pressure will be sufficient, and avoid the need for gas booster
equipment for the higher pressures typically required onto
forced-draught burners.
The chillers selected are two Carrier 30GTN 050 packaged air
cooled chillers. Details of equipment selected are contained in
Appendix D.
The boilers selected are three 112 kW atmospheric gas-fired

boilers. Details of equipment selected are contained in
Appendix D.
Boilers H6 H8
For this exercise, it has been assumed that the boilers are dealing
with the heating load only, and not any domestic hot water
requirements as domestic water services have not been a part of
this project.
Flue sizing H9
Flue or chimney sizing is important at scheme design stage as it
may determine whether planning permission is granted by the
local authority
Also, as with the air handling and chilled water plant earlier,
boilers have been provided for each block, and not on a total
building basis.
Heating plant
The duty required is based on the heat loss from the building
plus the fresh air load:
Heat loss = 1783 kW
Fresh air load = 1014kW
Where:
Q= m
& Cp T (in this case Cp is the specific heat capacity of
air 1026 kJ/kgC)
= 3528 1026 (24 - (-4)) (T is the difference
between room temperature and winter outdoor
temperature)
= 4339 28 = 1014 kW
Heating plant total load = 2797 kW
Say:
280 kW
A pre-heat load is applied to the calculated plant load. This
additional capacity ensures the building can be heated from a
low temperature in a timely fashion, ready for occupancy. This
is based on factors such as the thermal weight of the building
and heating periods. A minimum pre-heat factor of 12 is
recommended by CIBSE Guide A and this will be used in this
instance:
Heating load including pre-heat factor = 280 kW 12
= 336 kW
Boilers have been selected which can be arranged in a modular
format to meet the design heating load. Each boiler has high
and low fire settings, and the number of boilers selected gives a
From a technical perspective, flue heights and diameters need to

be calculated to disperse the exhaust gases from the boilers. The
boilers are to run on natural gas, which is considered to have a
very low sulphur content, so the calculation to be followed is
that for non-sulphur bearing fuels. It is expected that a single
chimney will be provided on the roof for each set of boilers.
Step 1. Boiler heat input is 336 kW.
Step 2. From CIBSE Guide B1, figure A2.3, this level of heat
input gives a height of 09 m for a chimney through or adjacent
to a building. This is the height of chimney to be added to the
overall building height.
Completion of scheme design stage

The recommendations made during the outline proposals stage
are put to the client in the lead consultants report for approval
before work starts on final proposals and production
information. This report includes schematic drawings showing
the principles of the heating, cooling and ventilation systems,
and the engineering arrangement of plant, ductwork and
pipework associated with these systems. In the case of this
project, primary and secondary pumping circuits have been
suggested for hot and chilled water. These drawings are
included within Appendix D.
Many of the subjects considered during the scheme design will
need to be revisited during detailed design, as the level of
precision needed in the design will increase. Some of the more
detailed calculations may even require a fundamental rethink of
the tentative decisions made during the scheme design stage.
Obviously, if many of the early decisions are found to be
impractical, then the efficiency of the design process can be
severely affected. For this reason experienced designers will
need to bring their accumulated skill and knowledge to bear
during the scheme design (or even earlier), to make sure that
the project does not head down too many dead ends.

BSRIA BG 1/2006
23
4 FINAL PROPOSALS/PRODUCTION INFORMATION STAGE

Introduction
This section of the guide works through typical calculations and
considerations of the final proposal and production information
stages of design for building services. This follows on from the
project definition and outline proposal stages covered in earlier
sections of the guide.
The main features of the model building that are relevant to the
detailed design stage are:
New build
three storeys
open-plan offices
concrete floors and frame
full-height glazing on external facades and most internal
walls (including the enclosed atrium)
potential for future internal partitioning
roof-top plant area
ducted fresh air supply from the roof. The air will be heated
but not cooled
fan-coil units in the office.
Figure 8 shows a plan layout for the ground floor office area
which is the main focus of this guide. The floor area has been
sub-divided into a number of internal zones (edged in red), any
of which may form a future office space. Each zone has its own
fan-coil unit.
The calculation methodology adopted for each part of the design
follows that in BSRIA Guide 30/2003: A Practical Guide to
HVAC Building Services Calculations. The relevant calculations are
listed in Table 3. These calculations also build on those produced
in the outline and detailed proposals stages in the previous
section.
The HVAC services need to be developed to a sufficient level
of detail to allow contractors to price their tenders for procuring
and installing the services. The detailed heat gain and heat loss
calculations should be calculated to enable individual fan-coil
ratings to be determined. The sizing calculations for pipework
and ductwork are illustrated for the index pipe circuit and duct
run (the circuit and run that has the largest pressure drop). As
the central plant will be housed on the roof, the index
circuit/run will be on the ground floor. Calculations are also
included to show Kv values for two-port valves in a variable
flow version of the ground floor index circuit. The sizes of the
main pump and fan are calculated on the basis of the pressure
losses in the index circuit and run. Finally, the slot diffusers will
be installed around the outside perimeter of the office floor will
be sized.
Work plan and methodology

To execute this stage of the project, the following steps are
proposed, in the order shown. The order of some steps can be
changed depending on the nature of the project.
1. Detailed heat gains and losses Sub-divide the
previously calculated overall heat gain and heat loss values
for the ground floor of the demonstration project in order
to achieve individual heating and cooling kW ratings for
each fan-coil unit.
24

BSRIA BG 1/2006
2. Supply air quantity, condition and heating and

cooling coil sizing Sub-divide the previously calculated
fresh-air supply volume for the ground floor of the
demonstration project to achieve supply air volumes for
outlets. Based on the calculated kilowatt ratings for each
fan-coil unit, calculate the appropriate heating and cooling
design flow rates required to achieve these outputs.
3. Detailed pipework layout and sizing For the ground
floor pipework system, prepare a detailed design schematic
drawing showing the layout of pipes, valves and other
components (a constant volume system will be assumed).
For the ground floor index circuit, size all of the pipework,
and calculate the pressure losses through associated
equipment and fittings in the circuit.
4. Valve sizing In a constant flow system calculate the
balancing pressures for the regulating valves. For a variable
flow version of the ground floor index circuit, calculate the
required Kv values for two-port valves to achieve an
adequate valve authority.
5. Pump sizing from overall pressure losses For the
heating and cooling pipework risers, calculate the overall
pressure losses. Together with the ground floor index circuit
pressure loss, use these losses as the basis for sizing the main
pumps.
6. Detailed ductwork layout and sizing For the ground
floor supply ductwork system, prepare a detailed design
schematic-drawing showing the layout of ducts, dampers
and other components. For the supply ductwork riser,
calculate the overall pressure losses. Together with the
ground floor index circuit duct pressure loss, use these losses
as the basis for sizing the main fan.
7. Select diffusers For the fan-coil unit supply ductwork
system, select slot diffusers for the perimeter of the office
space.
1. Room heat gains and losses

The scheme design stage involved calculating the overall
heating and cooling loads for each floor of the demonstration
project, and for the building as a whole. These loads need to be
broken down into individual room loads in order to size
individual fan coil units.
Using proprietary design software, the heat losses and sensible
and latent heat gains can be determined for each room. The
main design input values for the detailed design stage are listed
in Appendix E. Table 13 shows the resulting heat losses and
gains for a group of ground floor rooms.
C5

Figure 8: Ground floor plan with zones, fan coil locations and ductwork/pipework layouts.
Table 13: Summary of room heat losses and gains.

Zone
Properties
G01-G04
18 m
Two people
360 W small power load
02 air changes per hour
West facing window 30x28 m
G05
G06
G07-G12
G13-G14
G15-G16
Heat losses (W)

Room Losses
Fresh air loss
Heat gains (W)

Sensible gain
Latent gain
876
749
1520
123
1291
749
2210
124
1846
749
2635
125
876
749
1402
123
675 m
Five people
773
1872
2509
300
54 m2
Four people
619
1498
1989
240
27 m
Two people
2
18 m
Two people
South facing window 30x28 m
2
18 m
Two people
South facing window 30x28
2

BSRIA BG 1/2006
25
C5

2. Determining supply air quantity and condition
C5
Following the methodology laid out in BSRIA Guide 30/2003,
(section C5), an appropriate quantity and condition can be
determined for the supply air leaving the fan-coil units.
Step 1. Select a reasonable supply temperature differential for
the cooling situation (the difference between room air
temperature and supply air temperature).
The room design air temperature for the demonstration project
is given as 24C. The CIBSE Guide B2 Section 4.2.3.4
recommends a maximum cooling temperature differential of
10C for low ceiling fan-coil unit applications. Hence a
minimum supply temperature of 14C (24C-10C) is
acceptable within the guidance.
Step 2. Calculate the required mass flow rate for cooling from
the equation:
m
& = Qs /(Cp t)
Where:
m
& = mass flow rate of supply air (kg/s)
Qs = sensible heat gain, or loss (kW)
Cp = specific heat capacity of air (=1026 kJ/kgK)
t = temperature difference between supply air and room air
(10 K)
Hence, for rooms G01-G04, with a sensible gain of 152 kW,
the required mass flow rate to each room will be:
m
& = 1520 (1026 10) = 0148 kg/s
This value must be cross-checked to ensure that it is sufficient to

cover the fresh air requirement for these rooms. To make a
comparison with the fresh air requirement, the mass flow rate of
air must be converted to volume flow rate. From psychrometric
data it can be established that the specific volume (v) of air at
room condition (24C, 50%RH) is 0854 m3/kg. Hence the
required volume flow rate of air to each room will be:
m
& v = 0148 0854 = 0126 m3/s
The fresh air requirement for these rooms is:

Two people times 12 l/s = 24 l/s = 0024 m3/s.
This shows that the fresh air requirement can easily be
accommodated within the overall supply air volume.
The overall room air change rate should also be checked against
CIBSE norms. The room volume is approximately 18 m2 area
by 28 m high. Therefore, a volume flow rate of 0126 m3/s
gives a room air change rate (ac/h) of:
(0126 60 60) (18 28) = 90 ac/h
This value is within the normally quoted rule of thumb of
10 ac/h.
26

BSRIA BG 1/2006
Step 3. For the calculated mass flow rate of air, calculate the
likely air temperature to be supplied to the space in heating
mode and assess whether this is acceptable. This calculation
only applies to the room heat-loss, as, for a fan-coil application,
it would be normal to heat ventilation air at a central air
handling unit.
t = Qs (Cp m
& )
Hence, for rooms G01-G04, the supply air temperature
differential will be:
t = 0876 (102 0148) = 58 K
The upper limit for supply air temperature in heating mode is
normally 12 K above the design room temperature, and 58 K is
within this limit.
Step 4. Repeat the calculation with a different supply
temperature differential if any of the results are unsatisfactory.
If it had been found that the fresh air volume could not be
accommodated within the total supply air volume, or the value
for air changes per hour was too high, then it would be
necessary to go back to step 1 and try a different supply
temperature differential.
Step 5. Calculate the moisture content differential between
supply air and room air at the calculated supply air mass flow
rate from the equation:
g = Ql ( m
& hfg)
Where:
g = moisture content differential between supply air and
room air (kg/kg)
Ql = latent heat gain (kW)
m
& = mass flow rate of supply air (kg/s)
hfg = latent heat of evaporation (2450 kJ/kg)
Hence, for rooms G01-G04 which have a latent heat gain of
0123 kW, the resulting moisture content differential will be:
g = 0123 (0148 2450) = 000034 kg/kg
From a psychrometric chart, moisture content at the
summertime design condition (24C, 50% saturation) is 00095
kg/kg.
Hence the summer time supply condition is therefore 14C
(24C - 10C) at a moisture content of 000916 kg/kg (00095
000034).
Furthermore the winter time supply condition is 278C
(22C+ 58C).
C5
C6

Heating/cooling coil sizing C6
section C6, an appropriate method for sizing the heating and
cooling coils on the fan-coil units is as follows:
Heating coils
Step 1. From Calculation C5 for rooms G01-G04, the
required supply air mass flow rate for winter cases is 0148 kg/s.
The winter time supply condition is 278C.
Step 2. Calculate the required heating coil duty:
& s cp t
Qh = m
Where:
Qh = heating coil duty (kW)
m
& s = mass flow rate of supply air (kg/s)
T = temperature difference between supply air and return
air to fan-coil unit (K)
For the heating situation, the air entering each fan-coil unit will
be a mixture of room air at 22C and pre-heated fresh air, also
at 22C. To offset the room heat losses, this air must be raised
to 278C as previously calculated. These figures give a heating
coil duty of:
Qh = 0148 102 58 = 0876 kW
This result is no surprise, as it equals the sensible heat loss
shown in Table 13.
Step 3. For water to air heating coils, calculate the mass flow
rate for the heating water:
m
& w = Qh (Cp T)
Where:
m
& w = mass flow rate of heating water (kg/s)
Cp = specific heat capacity of water (42 kJ/kgK)
T = temperature difference between flow and return
water (K)
In order to complete this calculation, a design value for T
must be selected. In the past it was customary to select heating
flow and return temperatures of 82C-71C as these represent a
direct conversion from 180F-160F in the old Fahrenheit scale.
However, this temperature differential should not be seen as
fixed and should be re-considered for each application.
By increasing the temperature differential, fan-coil design flowrates will be reduced, resulting in the need to pump less water
and consequently save on pipe installation costs and pump
energy. In many European countries, a 20C temperature
differential is common for this reason.
However, one consequence of increasing the design

temperature drop may be that the resulting design flow rates
end up too low to be measured by commonly available flow
measurement devices (0012 kg/s tends to be the minimum
measurable value through a 15 mm device). This will mean
that the system cannot be commissioned following CIBSE and
BSRIA guidance.
For this reason, systems with large temperature drops are usually
designed with some means of self-balancing so that flow
measurement devices can be omitted. This may be achieved
either by:
Creating a reverse-return layout

sizing mains pipework with very low resistance relative to
branch pipes
grouping identical fan-coil units in pairs and only measuring
their combined flow rates.
With the introduction of condensing boilers, another option is

to reduce the flow temperature from the boiler from 82C to
60-70C. This will achieve the best possible efficiency from the
boiler by ensuring that the return water temperature is below
55C for as much of the year as possible. It will also provide
better control of room supply air temperatures, minimising the
risk of over-heating. If the supply air is over-heated, there is a
risk that its increased buoyancy will cause it to stay at high level
once it is discharged into the room. It may then be extracted
through a return air grille before it can mix enough with the
room air. This problem can be avoided by reducing the
heating-coil design temperature.
For the purposes of the demonstration project, a design
temperature differential of 20C has been selected (70C-50C)
with the intention of pairing up identical low-flow fan-coil
units to achieve measurable flow values. The pipe layout
dictated by these heating considerations will also be applied to
the chilled water system.
Where the calculations indicate adjacent fan-coil units are not
quite identical (as for FC G05 and FCG06), the two fan-coil
units have been treated as identical units to facilitate pipe sizing.
For the heating coil serving rooms G01-G04, the heating water
mass flow rate will be:
m
& w = 0876 (42 20) = 00104 kg/s
Cooling coils
Step 1. From Calculation C5 for rooms G01-G04 the
required supply air mass flow rate for winter cases is 0148 kg/s.
The summer-time supply condition is 14C at a moisture
content of 000916 kg/kg.
Step 2. Calculate the required cooling coil duty:
For a fan-coil application, the air entering the coil will be a
mixture of re-circulated room air and fresh air supplied via a
central air-handling unit.

BSRIA BG 1/2006
27
C5
C6

In the demonstration project, if the coil were to be sized for
sensible cooling only, its duty could be calculated using the
same equation as for the heating coil. However, as the coil will
be de-humidifying the air, the change in enthalpy must be
taken into account. The required coil cooling duty can be
calculated from:
Room ratio line = Sensible heat gain=152 kW

(Sensible + latent heat gain)
(152 + 0123) kW = 0925
Figure 9: Psychrometric process for a fan-coil unit.
Qc = m
& s h
Where:
Qc = cooling coil duty (kW)
m
& s = mass flow rate of supply air (kg/s)
h = the specific enthalpy difference between the on-coil
and off-coil conditions (kJ/kg).
The specific enthalpy of air at the summer-time off-coil supply
condition can be determined as 37 kJ/kg by reference to the
psychrometric chart in Figure 9. In order to establish the
specific enthalpy coming on to the coil, the mixture condition
must be established. The mix temperature can be found from:
& rc trc) + ( m
& ao tao)} m
&m
tm = {( m
Where:
tm = temperature of mixed air (K)
m
& rc = mass flow rate of re-circulated room air (kg/s)
trc = temperature of re-circulated room air (24C minus the
room condition)
m
& ao = mass flow rate of entering fresh air (kg/s)
tao = temperature of entering fresh air (29C summer
outside temperature)
m
& m = mass flow rate of mixed air (kg/s).
The mass flow-rate of re-circulated room air can be determined
as the total mixed air flow-rate minus the fresh air flow-rate.
The volume of fresh air required is 0024 m3/s which, at specific
volume 0847 m3/kg, is equivalent to 0028 kg/s. The mass
flow rate of re-circulated air to rooms G01-G04 therefore is
0148 0028 = 012 kg/s.
Hence:
tm = {(012 24) + (0028 29)} 0148 = 249C
Therefore, for the mixed air condition, it can be determined
from a psychrometric chart that the specific enthalpy of the air
will be 50 kJ/kg and the total cooling coil load will be:
Qc = m
& s h = 0148 (50 37) = 1924 kW
Having reached this point, the entire process can be plotted on
a psychrometric chart. Figure 9 shows the psychrometric
process for the fan-coil units serving rooms G01-G04, based on
the preceding design decisions. The room ratio line for this
psychrometric chart is calculated using the sensible and latent
heat gains for rooms G01-G04 from Table 13.
Step 3. For water-to-air cooling coils, calculate the mass flowrate for the chilled water:
m
& w = Qc (Cp T)
Where:
m
& w = mass flow rate of cooling water (kg/s)
Cp = specific heat capacity of water (= 42 kJ/kgK)
T= temperature difference between flow and return
water (K).
For cooling applications, it is customary to select chilled-water
flow and return temperatures of 6C and 12C respectively.
However, as for heating applications, these values should be reconsidered for each project.
For the demonstration project, the room ratio-line in Figure 9
can be extended so that the 100% saturation curve is crossed at
approximately 12C db. This is the required average coil
temperature, suggesting that 9C flow and 15C return is
feasible for this application. The opportunity to increase chilled
water temperatures from the norm is largely due to the higher
than usual room design temperature selected for the
demonstration project (24C). The benefit of this selection is
an overall energy saving at the chiller, and chilled water
pipework that is less prone to condensation.
For the cooling coils serving rooms G01-G04, the chilled water
mass-flow rate will be:
m
& w = 1924 (42 6) = 0076 kg/s
28

BSRIA BG 1/2006
C5
C6

Conclusion
The calculations involved in determining supply air conditions
and selecting heating and cooling coils can be automated on a
spreadsheet to create a full schedule of fan-coil. Such
spreadsheets help the designer to vary supply air temperature
differentials to arrive at the best relationship between supply air
condition and the room air change rate.
Table 14 shows a schedule of fan-coil data for rooms G01
G16 (as identified in Figure 9) incorporating the preceding
design decisions.
This information can be forwarded to a fan-coil manufacturer

who will select units that comply with the general requirements
of the schedule. In addition to the heat transfer requirements of
the fan-coil units it is also appropriate to specify any other
limiting criteria such as:
Acceptable room noise criteria

the maximum chilled and heating water pressure drops
across the coils
the maximum resistance of external ductwork.
Table 14: Fan-coil schedule.

Reference
Room
served
Room heat gains

load (kW)
Sensible
Latent
Cooling
load (kW)
Chilled
water
(kg/s)
Heating
load
(kW)
Heating
water
(kg/s)
Air volume (l/s)

Fresh air
Total
Air change
rate (ac/h)
FC G01
G01
1520
0123
1926
0076
0876
00104
240
1255
90
FC G02
G02
1520
0123
1926
0076
0876
00104
240
1255
90
FC G03
G03
1520
0123
1926
0076
0876
00104
240
1255
90
FC G04
G04
1520
0123
1926
0076
0876
00104
240
1255
90
FC G05
G05
2210
0124
3339
0132
1846
00220
240
1824
87
FC G06
G06
2635
0125
3339
0132
1846
00220
240
2175
155
FC G07
G07
1402
0123
1776
0070
0876
00104
240
1157
83
FC G08
G08
1402
0123
1776
0070
0876
00104
240
1157
83
FC G09
G09
1402
0123
1776
0070
0876
00104
240
1157
83
FC G10
G10
1402
0123
1776
0070
0876
00104
240
1157
83
FC G11
G11
1402
0123
1776
0070
0876
00104
240
1157
83
FC G12
G12
1402
0123
1776
0070
0876
00104
240
1157
83
FC G13
G13
2509
0300
3179
0126
0773
00092
600
2071
39
FC G14
G14
2509
0300
3179
0126
0773
00092
600
2071
39
FC G15
G15
1989
0240
2520
0100
0619
00074
480
1642
39
FC G16
G16
1989
0240
2520
0100
0619
00074
480
1642
39
Chilled water temperatures: Flow: 9C, Return: 15C

Heating water temperatures: Flow: 70C, Return: 50C
Winter room air condition: 22C db
Summer room air condition: 24C db, 50% saturation
Minimum supply air temperature: 14C
Design supply air temperature drop across coil: 10C
Summertime fresh air supply condition: 29C db, 20C wb
Heat gains from fan-coil unit fans excluded from total cooling loads. Values in bold have been arbitrarily adjusted to facilitate pipe sizing

BSRIA BG 1/2006
29
C5
C6

Figure 10: Schematic layout for heating and chilled water pipework.
30

BSRIA BG 1/2006
W2

3. Heating and cooling pipe sizing straight
lengths W2
section W2, the main pipe sizes for heating and chilled water
pipework can be determined.
Step 1. Sketch the system. Figure 10 shows a schematic layout
of the demonstration project heating system. To avoid
confusion, the chilled water system will have the same layout.
In reality this is not a very effective way of dealing with the

problem of heat losses from pipes, as fan-coil unit flow rates
would need to be increased significantly to compensate for
reduced inlet temperatures. This is because the heat transfer
across a heating coil is not particularly sensitive to changes in
flow rate.
The pipework layout has been decided taking into account the
decisions made when sizing the heating coils (calculation C6 see
page 27). In order to achieve branch flow-rates that are high
enough to be measured using commonly available flowmeasurement devices (in excess of 0012 l/s), terminal units
have been paired. A single regulating valve and flow
measurement device (orifice plate) serves each pair.
In the case of the demonstration project, where the design flow

temperature is to be set at 70C, an alternative approach has
been decided. First, pipework will be insulated to minimise
pipe heat emissions. Then, instead of increasing flow rates to
compensate for heat losses, boiler flow temperatures will be
increased slightly. Fixing a boiler flow temperature of 73C
should ensure a temperature of 70C at the inlets to the fan-coil
units, with no need to increase flow rates. Furthermore, in
making this allowance, there is no need to increase boiler size as
almost all pipe heat-loss will be inside the building.
To make this solution work, the units forming each pair have
been deliberately selected (or sized) as identical units with
identical flow rates and pressure losses. It can therefore be
assumed that water entering each branch will divide evenly
between the two units.
If there had been significant amounts of external pipework, a

more detailed assessment would be required, possibly resulting
in additional boiler loads. In the context of this project, the
amount of pipework on the roof of the building is not deemed
to be significant.
The schematic in Figure 10 also shows flushing by-passes and

flushing drains for flushing and chemical cleaning purposes, in
compliance with BSRIA Application Guide AG 1/2001.1: PreCommission Cleaning of Pipework Systems.
Although heat emissions from pipes are generally not relevant to

the chilled water system, some allowance for temperature gains
from the chilled water pump may be appropriate. For example,
for the demonstration project, it is appropriate to allow for a
05C temperature rise between the chiller and the fan-coil
units due to heat gains on pipework and from pumps. For a
system flow temperature of 9C the chiller flow temperature
should therefore be set at 85C.
Pipe materials must also be considered at this point. Heavy

grade mild-steel pipes tend to be the most economic for heating
and chilled water applications. While mild-steel would not be
considered for potable supplies due to its ability to corrode
when in contact with oxygen and water, it is acceptable for
closed heating and chilled water circuits. In these systems, the
ingress of air should be minimal and corrosion can be controlled
by inhibitor chemicals.
For these reasons heavy grade mild-steel is the preferred choice
for pipe mains, although run-outs to terminal branches are
better in copper. Copper is generally easier to bend at small
diameters making it easier to connect to fixed terminals.
Step 2. Estimate the pipe heat emission for each pipe section.
If heat is lost from the pipework before the heating water
reaches the fan-coil unit, the entering water temperature will be
lower than its design value, resulting in less heat transfer than
expected.
The normal solution to this problem is to make an allowance
for heat emissions from pipework, typically 25 W/m for
insulated pipes and 100 W/m for un-insulated pipes. This
allowance is added to the fan-coil unit heating loads to give an
overall load from which heating flow rates can be calculated.
The result of this approach is that, although the water reaching
the fan-coil units may be at a lower temperature than when it
left the boiler, the temperature drop across the fan-coil units
will be less than the design value (as the flow rate is higher).
Step 3. Select an appropriate temperature drop across the

system.
For the heating system, an appropriate design temperature drop
had to be considered when determining the heating load under
calculation C6 (see page 27). For the reasons previously
explained, a temperature drop of 20C has been selected
(70C-50C).
Step 4. Select an acceptable design value for either pressure
drop per unit length or velocity.
Most engineers tend to size-pipework within a maximum
pressure drop per unit length. High pressure-drops per metre
result in smaller, cheaper pipes, but higher pump energy
consumption. As a trade-off between installed cost and pump
energy costs, pipe pressure losses in the range 250 360 Pa/m
as quoted in CIBSE Guide C4 Flow of Fluids in Pipe and Ducts
are acceptable. Flows within these ranges are usually well
below acceptable velocity limits for pipes. Commonly-quoted
velocity limits for pipes are usually indicative of the point at
which fluid flow may become noisy or pipe surface erosion may
occur. These limits are usually well above the normal range of
limiting pressure losses per metre.
In view of the relatively low cost of energy, a decision has been
made on the demonstration project to size pipes within a limit
of 350 Pa/m.

BSRIA BG 1/2006
31
W2
W3

Step 5. Size pipes using tables 4.9-4.33 of the CIBSE Guide
C4 Flow of Fluids in Pipe and Ducts.
The pipe sizes for individual branches, together with their flow
rate, flow velocities and anticipated Pa/m pressure losses are
shown in Table 15. For the layout illustrated in Figure 10,
Table 16 shows the same calculations carried out for the chilledwater schematic.
Heating and cooling pipe sizing fittings W3

Step 1. Find the appropriate velocity pressure loss factor (
zeta) value for the fitting in Tables 4.47-4.59 of CIBSE
Guide C4. For example, for a 15 mm smooth radiused copper
elbow, the appropriate value is 093.
The precise number of elbows used in the branch will
ultimately be determined by the pipe fitter installing the system.
To allow for a worst case situation, four elbows have been
allowed for in the demonstration project.
Step 2. Find the equivalent length for the type/size of pipe to
which the fitting will be connected. For example, the hot
water pipe serving fan-coil units FC G15 and G16 has a flow
rate of 00148 l/s (Table 14 shows 00074 l/s for each unit), and
an approximate temperature of 75C. CIBSE table 4.13 gives a
pressure loss of 16 Pa/m and equivalent length of 03 for 15 mm
copper pipe. Note that the CIBSE table quotes flow rates in
kilograms per second, but for this design the density of water
can be taken as 1 kilogram per litre, so flow rates in kg/s are
equivalent to flow rates in litres per second.
This pressure loss is low compared to the range of losses
included in the table, but 15 mm is typically taken as the
smallest practical size. This is because smaller diameter copper
pipes are prone to kinking when bent site bending is often
required if pipe routing has to be adjusted during installation.
Note that to obtain the pressure loss figures, a pipe sizing chart
designed for water at 75C is being used whereas for the
demonstration project the average flow temperature will be
around 60C. Probably the only way to gain a truly accurate
value for pressure losses is to use a computer program which
will correct density and viscosity values depending on
temperature. This is why a relatively low pressure loss per
metre is the result for such a low flow.
Step 3. Multiply the value by the equivalent length to give
the actual equivalent length of pipework which the fitting
represents for example 093 03 = 0279 m.
Step 4. Multiply the calculated equivalent length by the
pressure per metre value to determine the total pressure drop
through the fitting, for example 0279 m 16 Pa/m = 45 Pa
and as there are four elbows per branch the total pressure drop
is 18 Pa.
Although pressure losses through standard fittings such as bends,
enlargements and restrictions can be calculated in this way,
pressure losses for some components must be determined using
manufacturers data. For example, pressure losses through fan
coil units will usually be determined by the manufacturer.
32

BSRIA BG 1/2006
Some designers specify maximum pressure losses across fan-coil

units which the manufacturer must not exceed, typically 5 kPa
for heating fan-coil units and 12 kPa for chilled water fan coil
units. This approach has been taken for the demonstration
project.
It is important to clarify that these values account for the
pressure drops across the four-port valve attached to each fancoil unit and are not just for the fan-coil unit on its own. Since
most four-port valves are supplied already attached to the fancoil unit, fan-coil unit manufacturers are able to ensure that the
total pressure drop is within the specified limit.
4. Valve sizing
Double regulating valves
Another common item which must be sized based on
manufacturers data is the double regulating valve. Valve
manufacturers issue Kv values rather than values for their
products. The Kv value represents the flow rate in cubic metres
per hour required to induce a pressure drop of 1 bar (100 kPa).
However, the Kv value can be used in the following equation to
calculate pressure loss at a given flow rate:
P = (36Q Kv)2
Where:
P = the Pressure loss across the valve (kPa) and
Q = the flow rate through the valve (l/s).
In order to calculate the pressure drop across the regulating
valve, the valve must first be selected so that its Kv value can be
determined. Valve manufacturers have deliberately designed
their valves such that most valves will be line size in other
words the same size as the connecting pipework. For 15 mm
regulating valves the choice may be between alternatives for
low, medium or standard flow rate.
In order to ensure that regulating valves do not need to be
closed below 25% open, thereby running the risk of blockages
due from any circulating debris, the valve manufacturer should
also be informed of the pressure to be taken out due to valve
closure. This value can be calculated as the difference between
the pressure drop around the index circuit and the circuit in
question, as illustrated in Figure 11. Balancing pressures
(sometimes referred to as residual pressures) have been
calculated for the demonstration project circuit and are
indicated in Table 15 and Table 16 for heating and chilled
water branches.
An example calculation for the regulating valve for the heating
circuit G15/G16 would be as follows:
G15/G16 residual pressure = Pindex - PG15/G16, where Pindex
- PG15/G16 have been calculated (see the results given in Steps
4 and 5 on page 33).
G15/G16 residual pressure = 29989 20843 = 9146. The
figure given at the bottom of Table 15, 9145, does not
include rounding errors. Balancing, or residual pressures, are
given for all the circuits in the pipework layout.
W4

Figure 11: Calculation of regulating valve residual pressures.
For the demonstration project, the ground floor circuit shown

in Figure 10 is deemed to be part of the index circuit from the
pump. As the pump is located at roof level, the circuit serving
fan-coil units G11 and G12 is most likely to be the index
circuit.
In order to calculate the maximum pressure drop across the
pipework serving the ground floor, the following steps must be
followed:
Step 1. Identify each section of pipe and the fittings (branches,
connectors, valves, strainer and terminals). For the calculations
summarised in Table 15 and Table 16, bends have been
included with the pipe sections.
Note that this diagram illustrates the principal of residual

pressure and does not directly represent the schematic pipework
layout shown in Figure 10.
Two-port and four-port control valves
The demonstration project has been designed as a constant flow
system with four-port valves at fan-coil units. As the authority
of a four-port valve relies on it being sized to have a full-open
pressure loss approximately equivalent to the fan-coil unit it
controls, it can be selected by the fan-coil unit manufacturer.
Hence there is no need for designers to consider the sizing of
these control valves.
However, due to the advent of low cost variable-speed pumps,
variable volume systems with two-port control valves are now
more common. For these valves, achieving good control
authority requires the valve to be selected to achieve a high
pressure loss relative to the complete circuit in which it is
located. If the two-port valve is to be shut off against the pump
pressure, it must be selected such that in its full open position,
its pressure drop is at least 23% of the pump pressure.
Alternatively, if a differential-pressure control valve is installed
somewhere between the pump and the two-port valve, then the
valve must be sized such that its pressure drop is at least 23% of
the differential pressure controlled by the differential-pressure
control valve.
As for regulating valves, pressure losses across fully open twoport control valves are calculated using manufacturers published
Kv values.
5. System resistance for pipework index run W4

Having sized pipework and calculated individual pressure losses
through pipes, fittings, terminal units and valves, the designer
must determine the maximum design pressure loss through the
system. This will be required in order to size the pump.
By definition, the maximum pressure drop through the system
will occur in the index circuit This is the circuit which has the
greatest resistance and hence pressure drop. It is normally the
circuit extending from the pump to the furthest extremity of
the system, as this circuit will have the greatest pipe pressureloss. However, if pressure losses in the terminal units are large
compared to the pipe losses, then the index circuit may not be
the most remote.
Step 2. Identify each circuit by the pipe sections and fittings it

comprises. For example, referring to Figure 10, circuit
G15/G16 comprises:
A flow pipe from the riser to the main branch

the strainer in the above section of pipe
the diverging tee between the pipe from the riser and the
main branch
the flow pipe along the main branch
the straight through tee and contraction from 20 mm to 15
mm between the main branch and the terminal branch
the flow pipe along the terminal branch
the diverging tee between the two fan-coil units
flexible connections to and from the fan-coil units
the return pipe along the terminal branch
the converging tee in the terminal branch
the regulating valve in the terminal return branch
the converging tee with the main branch and expansion
from 15 mm to 20 mm
the return pipe along the main branch
the converging tee with pipe back to the riser
the return pipe back to the riser
the regulating valve in the above section of the pipe.
Step 3. Calculate the pressure losses across the fitting and the
pipework in each section. For the demonstration project this
exercise has been completed and the results shown in Table 15
and Table 16.
Step 4. Add up the total pressures losses from each section to
determine the overall circuit design pressure-drop. For example,
for the circuit serving units G16/G15 the total design pressure
loss will be the sum of the pressure losses for the pipe sections
identified at Step 2. Reading the values from Table 15 the
calculation is:
(2074 + 1243 + 6202 + 5559 + 5765) = 20 843 Pa
Step 5. Repeat this exercise for each of the circuits in the
system starting each time from the riser. Identify the circuit
with the largest pressure drop. This is the index circuit and its
pressure drop is the maximum design pressure drop for the
system. For the heating circuit in the demonstration project,
the index circuit is the end branch with a maximum pressure
drop of 29 989 Pa. Following the same principles, the
maximum pressure loss for the chilled water system can be
calculated as 40 932 Pa.

BSRIA BG 1/2006
33
Table 15: Heating system pipe sizing results.
34

BSRIA BG 1/2006
Table 16: Chilled water system pipe sizing results.

BSRIA BG 1/2006
35
W5

Pump sizing W5
Step 2. Convert mass flow rate to volume flow rate in l/s.
Step 1. Calculate the index run pressure-drop and the total

system mass flow rate.
As the density of water is close to 1000 Kg/m3 at typical

operating temperatures for heating and chilled water systems,
each litre of water has a mass of almost exactly 1 kg. Hence,
litres per second values are effectively the same as kilograms per
second values. Only when the fluid density differs (as for glycol
mixtures) is there a difference between mass and volume flow
rates. In these cases the equation linking volume flow rate, mass
flow rate and density must be used:
For the demonstration project, the pumps distributing flow to

the fan-coil units are secondary dual pump-sets distributing flow
from a low pressure-loss header to the fan-coil units. The
arrangement is shown on drawings 70206 K/01 and /02 in
Appendix D. When sizing the heating and chilled water
secondary pumps, no account therefore needs to be taken of
pressure losses through boilers, chillers and associated
equipment. These losses will be dealt with by a separate set of
primary pumps.
To size the secondary pumps the pressure drop in the entire
index circuit must be calculated. The pressure drops in both
heating and chilled water pipework from the riser to the index
terminal has already been calculated on page 33 using
calculation W4. The results are summarised in Table 15 and
Table 16. The pressure losses in pipes between the rooftop
secondary pumps and the ground floor must be added to the
aforementioned pressure drops.
If the maximum pipe pressure drop per metre is 350 Pa/m, then
this implies an average pressure loss of around 300 Pa/m.
Assuming a total flow and return pipe distance of say, 15 m
(approximately 28 m per floor, times six for the flow and return
for each of the three floors) then the total pressure loss in these
pipes is 4500 Pa. A further 25% should be added for fittings and
components giving 5625 Pa. This figure is added to the ground
floor pipework pressure losses from Table 15 and Table 16 to
give overall heating and chilled system pressure losses:
For heating: 29 989 + 5625 = 35 614 Pa (say 35 600 Pa)
For chilled water: 40 932 + 5625 = 46 557 Pa (say
46 600 Pa).
The total system mass flow rate will be the sum of flow rates to
all parts of the building. From the outline/detail design stage
calculation in the section on vertical risers on page 21, the
overall heating and chilled water flow rates were estimated as
386 kg/s and 121 kg/s respectively. While the chilled water
flow rate is still valid, the overall heating water flow rate will
change due to the decision to adopt a 20C temperature drop
rather than the 10C considered at scheme design stage.
Repeating the calculation on page 21 for low temperature hot
water, and using the formula Q = m Cp T with the total
building heat loss of 1783 kW, Cp = 42 kJ/kgK and T of
20C, the revised flow rate can be calculated as 212 kg/s.
To allow the system to be commissioned at a flow rate of up to
110% of design flow rate at the pump, a 10% margin should be
added to the design flow rates. The square law relationship
between flow rate and pressure loss means that system pressures
must be increased by 21%. The final duties against which
pumps can be selected will therefore be:
For heating: 233 kg/s against 43 000 Pa
For chilled water: 133 kg/s against 56 400 Pa
36

BSRIA BG 1/2006
Q=m
Where:
Q = volume flow rate (m3/s)
m = mass flow rate (kg/s)
= fluid density (kg/m3)
If this equation is used and a volume flow rate in m3/s is
calculated, then this is converted to litres per second by
multiplying by 1000.
For the remaining steps, only the pumps in the heating circuit
have been sized.
Step 3. Determine system equation constant R. This is done
by substituting the required P (pressure) and Q (volume flow
rate) values into the equation P = RQ2 and then solving for
R. For example, for the heating system:
R = P Q2 = 43 000 2332 = 7921
Step 4. Select a pump that will operate within the required
parameters and plot the system and pump characteristics on the
same graph.
Having calculated R at the preceding step, values of P can be
calculated at different values of Q and the resulting system
characteristic plotted on a graph. This can be overlaid with the
selected pump curve taken from a pump manufacturers
literature.
The result for the demonstration project heating system is
shown in Figure 12.
Figure 12: Pump sizing chart.
A2

Step 5. Determine the operating point. Identify the operating
pressure and flow rate.
This is the point at which the two curves intersect. It can be
seen from Figure 12 that this occurs at a flow rate of
approximately 245l/s and 48 000Pa.
Step 6. Calculate the pump speed to achieve the required
values, or select another pump.
It can be seen from Figure 12 that the operating point gives a
flow rate and pressure loss higher than that required for the
system. If the excess pressure is large, it can be removed by
changing pump pulleys (on belt-driven pumps) or by throttling
a regulating valve on exit from the pump. Incorporating speed
control in pump motors enables the required flow rate to be
achieved enabling their speed to be reduced. This solution is
more energy efficient than changing pump pulleys or throttling
regulating valves, and makes commissioning much simpler.
6. Duct sizing selecting a circular duct size A2

A sketch layout of the fresh-air ductwork serving ground floor
fan-coil units for the demonstration project is shown in
Figure 13. Calculated fresh air loads for each zone, taken from
Table 14 are also shown. Individual sections of ductwork are
identified and the flow rates are based on the theoretical
occupancies of the rooms and zones shown in Table 13.
Figure 13: Schematic layout for fresh air ductwork.
from using flat oval or rectangular ductwork. For more

guidance on this subject see BSRIAs Services Co-Ordination with
Structural Beams Guidance for a Defection-Free Interface,
IEP2/2003.
For the demonstration project it has been assumed that there is
sufficient space for circular ductwork. However, once circular
duct sizes have been calculated, these can be checked against the
space allowed in the ceiling void.
A limiting velocity or pressure drop per metre must also be
selected as the basis for sizing the ducts. From the CIBSE
Guide C4 Table 4.34, 1 Pa/m is usually chosen as the maximum
pressure drop per metre in low velocity systems, with velocities
in the range 3-6 m/s.
For final run-outs to supply duct openings, noise is a concern
and for this reason a velocity limit of 25 m/s has been decided
based on the recommendation in CIBSE Guide B5 Table B5.3.
For the demonstration project, sizing within these limits and
sizing the ductwork manually, using Figure 4.2 from CIBSE
Guide C, the ducts forming the demonstration project circuit
can be sized. The results are shown in Table 17 for the fresh-air
supply ductwork from the riser to the ground floor fan-coil
units.
The ductwork diameters can now be compared with the size of
the ductwork zone shown in Figure 4. The ductwork zone is
given as 330 mm deep so obviously this creates a problem with
fitting 350 mm diameter ductwork, especially as allowance also
needs to be made for duct insulation. However, it will only be
crucial for services not to stray out of their zones at points
where they cross, and particularly where they pass under the
structural beams. Detailed discussion of the co-ordination of
pipework and ductwork with the structure is outside the scope
of this guide. However BSRIAs Services Co-ordination with
Structural Beams Guidance for a Defect-free Interface, IEP2/2003
explains these issues in great detail.
If clashes are detected, then a further iteration of design is
required, with consideration required of different services routes
and/or more complex cross-overs involving more fittings, such
changes are likely to affect the pressure losses along the pipes
and ducts.
For the purposes of this demonstration of the design process, it
is assumed that any potential clashes can be resolved without
affecting the routing of the services, so that the performance
requirements for pumps and fans do not need to be reexamined.
Before ducts can be sized, it must be decided whether the

ductwork will be circular or rectangular. Circular ductwork has
a lower pressure drop per metre than rectangular ductwork and
so requires less fan energy to distribute the air. The materials
are also cheaper. As a general rule, circular ductwork should
always be preferred unless there are spatial or other benefits
BSRIA BG 1/2006
37
A2

Table 17: Results of the duct sizing calculation.
P/l
(Pa/m)
Velocity
(m/s)
Diameter
(mm)
qs/qc
(or A2/A1)
Duct
loss (Pa)
Fittings
loss (Pa)
Total
loss (Pa)
0504
54
350
09
0456
09
51
350
09
90 branch tee
0004
1000
007
107
90deg branch tee
0004
1800
006
0432
08
48
350
186
09
10
90 branch tee
Concentric reducer
0004
005
1600
006
166
0408
07
44
0360
055
39
350
09
90 branch tee
0004
0700
005
075
350
09
07
90 branch tee
Concentric reducer
0004
005
0550
004
059
0336
0312
09
49
300
09
90 branch tee
0004
3000
006
306
46
300
08
90 branch tee
0016
2700
020
290
0252
0228
06
37
300
09
90 branch tee
0004
1800
003
183
05
34
300
08
07
Diverging Y piece 45
Concentric reducer
029
005
1500
201
351
10
11
0180
08
39
250
09
90 branch tee
0004
1600
004
164
0156
05
29
250
08
90 branch tee
0016
1500
008
12
158
0132
04
27
250
08
06
90 branch tee
Concentric reducer
0016
005
1200
007
127
13
0108
09
36
200
08
90 branch tee
0016
2700
012
282
14
0084
055
28
200
07
90 branch tee
0036
1650
017
182
15
0060
03
19
200
04
Diverging Y piece 45
90 bend
Duct outlet
054
024
100
1200
386
506
16
0048
02
17
200
90 bend
024
0800
042
122
17
0024
0023
15
150
0092
18
0024
0023
15
150
0092
19
0024
0023
15
150
20
0024
0023
15
150
0092
21
0024
0023
15
150
0092
22
0060
03
19
200
23
0024
0023
15
150
0092
24
0048
02
17
200
0800
08
25
0024
0023
15
150
0092
009
26
0024
0023
15
150
0092
009
27
0024
0023
15
150
0092
009
28
0024
0023
15
150
0092
009
29
0024
0023
15
150
0092
009
30
0024
0023
15
150
0092
009
Index circuit (sum of sections 1-15):
3141
Section
Length
(m)
Q
3
(m /s)
38

BSRIA BG 1/2006
Fittings
90 bend
90 bend
024
024
0092
1200
009
009
032
042
009
009
052
172
009
A4
A5
A6

Ductwork pressure loss through fittings A4
Step 1. Work out the ratios between flow rates or cross
sectional areas for each branch tee, reducer, and diverging tee.
These will be needed in order to select pressure loss coefficients
from CIBSE Guide C4. For the demonstration project,
appropriate ratios have been calculated and are included in
Table 17.
Step 2. Check the value for the density of air. For the
demonstration project, fresh air is to be supplied at the room
design temperature 22-24C for which air density is
approximately 12 kg/m3.
Step 3. Select the appropriate velocity pressure loss factors for
the fittings. Table 17 shows pressure loss coefficients ()
selected from CIBSE Guide C4 for the system used in the
demonstration project.
Step 4. Calculate velocities for each duct section. Table 17
indicates velocities for the system used in the demonstration
project, previously determined when sizing the ducts (A2).
Step 5. Calculate pressure losses through fittings from the
equation:
P = 0.5 v
Where:
P = pressure loss across fitting (Pa)
= pressure loss coefficient for fitting
= density of air (kg/m3)
v = air velocity (m/s)
For example, section 1 of ductwork includes a 90 branch tee.
The qs/qc ratio is 09, which from CIBSE Guide C4 gives a
pressure-loss coefficient of 0004. The density of air is 12
kg/m3 and the air velocity is 54 m/s. The above formula gives
the following pressure loss across the fitting:
P = 05 0004 12 542 = 007 Pa.
For the demonstration project, the results of these calculations
are shown in Table 17.
Duct sizing index run A5

Following the same principles as for the pipework system, the
index run of ductwork can be determined.
Step 1. Identify and assign a reference to each duct section.
Figure 13 shows the different duct sections numbered 1-30 for
the demonstration project.
Step 2. Identify for each duct run its combination of duct
sections. For example, from Figure 13 the shortest duct branch
comprises duct sections 1 and 16 whereas the longest comprises
duct sections 1 to 15 inclusive.
Step 3. Calculate all direct pressure-losses across fittings and

ductwork in each section. For the demonstration project, the
results of this calculation are shown in Table 17.
Step 4. Add up the total pressure losses from each section
within a run to give the run pressure drop. For example, for
the shortest run comprising duct sections 1 and 16 the total
pressure losses will be 107 + 122 = 229 Pa. For the longest
run comprising duct sections1-15, the total pressure loss will be
3141 Pa, say 32 Pa.
Step 5. Identify the index run as the run with the highest
pressure drop. For the demonstration project this will obviously
be the longest run with a pressure drop of 32 Pa. As for a pipe
system, the index run is usually the run of ductwork to the
furthest system extremity. Unlike a pipework system, where
high terminal-unit pressure losses can sometimes result in an
index appearing mid-way down a circuit, ductwork index
circuits are almost certain to be the run that feeds the end
outlet. The preceding calculation is therefore often unnecessary
but should be completed where similar adjacent branches occur
and where the longest run is not immediately obvious.
Supply fan sizing A6

Step 1. Calculate the ductwork index run pressure loss and
total system flow rate.
To size the supply fan the pressure drop in the entire index run
must be calculated. There are four components to this total
pressure drop:
1. Pressure drop on the ground floor from the riser to the
index terminal
2. pressure drop in the riser ductwork
3. pressure drop in bends from rise to air handling unit
4. pressure drop within the air handling unit.
These components are calculated separately and added together
to give the index run pressure-drop for sizing the fan.
1. The pressure drop in the ground floor ductwork from the
riser to the index terminal has already been calculated on pages
37 to 39, and summarised in Table 17, as 32 Pa.
2. The pressure drop in the riser can be estimated as follows. If
the maximum duct velocity is 6 m/s with a maximum pressure
drop per metre of 1 Pa/m then this implies an average pressure
loss of around 08 Pa/m inclusive of losses across fittings. The
length of supply duct in the riser from the air handling unit on
the roof to the ground floor ceiling void is approximately 9 m
(39 m for each of first and second floors plus 12 m from the air
handling unit to the roof slab). This gives a pressure loss in this
length of ductwork of 7 Pa.
3. Pressure drop in bends in the riser ductwork can be
estimated as follows. Assuming velocity of 6 m/s and a pressure
loss coefficient () of 04 for a bend, then with a density () of
12 kg/m3 then the pressure drop per bend is:
P = 05 v2 = 05 04 12 62 = 86 Pa.

BSRIA BG 1/2006
39
A6
A7

Allowing for three bends between the air handling unit and the
riser supply duct gives a pressure loss of 25 Pa.
4. Finally, allowance must be made for pressure losses across
components within the air handling unit. These will depend on
the precise components installed, an inventory of which will be
available from the manufacturer. For the purposes of this
demonstration project the following components and values
have been assumed:
from the fan. Alternatively a different fan can be selected which

gives a better match. Final selection should also ensure that the
fan motor can cope with changes in pressure loss due to
differences between clean and dirty filters.
Figure 14: Fan sizing chart.
An inlet louvre: 20 Pa
a frost coil/pre-heater: 40 Pa
a bag filter (dirty): 150 Pa
a fan: 40 Pa
an attenuator: 70 Pa.
The total of 320 Pa makes up the largest part of the overall

supply system pressure losses.
The pressure loss within the supply system is calculated by
adding together these four elements:
7. Grille and diffuser sizing A7

Ptotal = 32 + 7 +25 + 320 = 384 Pa (say 380 Pa)
The total air flow rate will be the sum of fresh air flow rates to
all parts of the building plus an allowance for air leakage from
the ductwork. From the scheme design report, the overall fresh
air flow-rate for the building is 3528 m3/s, say 353 m3/s.
Step 2. Determine the system equation constant R. This is
done by substituting the required P and Q values into the
equation P = RQ2 and then solving for R. For example, for
the fresh air ductwork system:
R = P Q2 = 380 3532 = 30
Step 3. Select a fan that will operate within the required
parameters and plot the system and fan characteristics on the
same graph.
Having calculated R at the preceding step, values of P can be
calculated at different values of Q and the resulting system
characteristic plotted on a graph. This can be overlaid with the
selected fan curve taken from a fan manufacturers literature.
The result for the demonstration project heating system will
appear as in Figure 14.
Step 4. Determine the operating point. Identify the operating
pressure and flow rate.
This is the point at which the two curves intersect. It can be
seen from Figure 14 that this occurs at a flow rate of 3.7 m3/s
and 400 Pa.
Step 5. Check for a mismatch between fan design duty and
operating point.
It can be seen from Figure 14 that the operating point gives a
flow rate and pressure loss that are higher than are required for
the system. If the excess pressure is large, it can be removed by
either reducing the fan speed or by throttling a damper on exit
40

BSRIA BG 1/2006
For the demonstration project, fan-coil units will supply air to

diffusers located in each office area. From Table 14 it can be
seen that, for the majority of rooms, the required air change rate
is less than 10 ac/h apart from room G06, (the exposed corner
room) which requires 155 ac/h.
For the majority of rooms with less than 10 ac/h, diffusers can
be selected which will encourage a coanda effect whereby the
air stream sticks to the ceiling on leaving the diffuser. For room
G06 which requires a high air volume, consideration may need
to be given to the use of a swirl diffuser which will discharge
supply air in a highly turbulent swirl.
The feasibility of these different options can be assessed by
reference to diffuser manufacturers selection data. This is
commonly in the form of a selection nomogram. The main
steps to be followed are described in A Practical Guide to HVAC
Building Services Calculations, BG 30/2003 section A7.
The design information required for selecting a diffuser
comprises:
Room dimensions
volume flow rate of air through the diffuser
required throw from the diffuser, based on the position of
the diffuser in the room and whether or not there are other
grilles or diffusers in the room
the limiting noise level for the room.
Using this information, and a nomogram from a diffuser

manufacturer, an appropriate diffuser can be selected. In this
case, a diffuser for room G05 will be selected.
The dimensions of the room are 45 m wide, 6 m deep, 28 m
high. The volume flow rate is 182 l/s (from the fan-coil
schedule in Table 14). The required throw for a single terminal
device in a room is 75% of the distance to the opposite wall,
45 m. The limiting noise level for the room is NR35.
A7

For this example, the selection nomogram of a particular
diffuser manufacturer (Air Diffusion) is being used (see Figure
15). This particular nomogram requires that the volume flow
rate be expressed per metre length of diffuser. In this case 4 m
of diffuser will be installed in this 45 m-wide room giving 45
l/s/m. In addition, the required throw must be corrected to
allow for the actual length of each section of the diffuser. The
length depends on the size of the plenum boxes above the
diffuser. In this case the 4 m long diffuser will have two 2 m
plenum boxes. A throw multiplier of 125 is taken from the
table attached to the nomogram (see Figure 15). This gives a
nominal throw of 45 m 125 = 36 m for use on the
nomogram.
Figure 15: A linear slot diffusers selection nomogram for horizontal

projection with ceiling effect.
On the nomogram, a straight line is drawn through the points

on the left hand scales of volume flow rate (45 l/s/m) and
throw (36 m) across to the other scales. From these scales it
can be seen that a one-slot diffuser will be appropriate. This will
have a sound power level of 28 dB and a pressure drop of
30 Pa.
The sound rating is within the limits specified, so this choice of
diffuser is suitable for this room. The nomogram should then
be used on all combinations of room size and flow rate to select
the range of diffusers for the building.
Closing statement
This report has shown one iteration through each of two stages
of the design process: for outline/detailed design and for final
proposals/production information. In practice, further
iterations may be necessary at each stage, particularly where the
design is concurrent with structural, architectural and other
buildings services design and therefore has to keep up to date
with how these designs evolve.
Furthermore, situations may arise where the results of detailed
calculations show that some of the assumptions made at earlier
stages are no longer valid. In these cases, the iterative process
will require that the earlier design stages are also re-examined.
In addition, if contractors are not involved in the early part of
the design process then further suggestions for changes to items
of plant, or design philosophies can arise when the m&e
contractor is engaged.
Finally, there is much repetition of detail in the design for a
whole building which has not been included here as not
necessary to demonstrate the application of the calculation
procedures published in A Practical Guide to HVAC Building
Services Calculations, BG 30/2003. For example, this guide only
looks at one floor of one office building. For a complete
design, all floors of both blocks, plus the other parts of the
building (laboratories, reception area, internal circulation space)
will also have to be modelled.

BSRIA BG 1/2006
41
A7

Closing statement
References
This report has shown one iteration through each of two stages
of the design process: for outline/detailed design and for final
proposals/production information. In practice, further
iterations may be necessary at each stage, particularly where the
design is concurrent with structural, architectural and other
buildings services design and therefore has to keep up to date
with how these designs evolve.
Association of Consulting Engineers, Conditions of Engagement for

Mechanical and Electrical Services Engineering B(2), ACE 2002.
Furthermore, situations may arise where the results of detailed

calculations show that some of the assumptions made at earlier
stages are no longer valid. In these cases, the iterative process
will require that the earlier design stages are also re-examined.
In addition, if contractors are not involved in the early part of
the design process then further suggestions for changes to items
of plant, or design philosophies can arise when the m&e
contractor is engaged.
Finally, there is much repetition of detail in the design for a
whole building which has not been included here as this is not
necessary to demonstrate the application of the calculation
procedures published in BSRIA Guide BG 30/2003. For
example, this guide only looks at one floor of one office
building. For a complete design, all floors of both blocks, plus
the other parts of the building (laboratories, reception area,
internal circulation space) will also have to be modelled.
British Council for Offices, Best Practice in the Specification for

Offices, BCO Guide 2000, ISBN 0 95241 312 4.
Building Regulations Part L of the Building Regulations, 2002.
Churcher D, A Design Framework for Building Services Design
Activities and Drawing Definitions, BSRIA, BG 6/2006,
ISBN 086022 656 5.
CIBSE Guide A, Environmental Design, 1999,
ISBN 0 900953 969.
CIBSE Guide B3, 2002
CIBSE Guide B5.
CIBSE Guide C, 2001
CIBSE Guide C4, Flow of Fluids in Pipe and Ducts.
CIBSE, Engineering Design Calculations and the Use of Margins,
1998.
CIBSE, Testing Buildings for Air Leakage, TM23, 2000, ISBN 1
903287 103.
Cross N, Design: Principles & Practice Product Planning and the
Design Brief, Open University 1995, ISBN 07492 71892.
Griffiths & Armour, PII scheme reports 1995 99. 1999.
Heywood M, Hawkins G, Mitchell S, Services Co-Ordination
with Structural Beams Guidance for a Defection-Free Interface,
BSRIA, IEP 2/2003, ISBN 086022 634 4.
Lawrence Race G and Mitchell S, A Practical Guide to HVAC
Building Services Calculations, BSRIA, BG 30/2003,
ISBN 086022 618 2.
Lawrence Race G, Design Checks for HVAC A Quality Control
Framework for Building Services Engineers, BSRIA, AG 1/2002,
ISBN 0 86022 589 5.
Loyd S, Guidance and the Standard Specification for Ventilation
Hygiene, FMS 1/97, BSRIA, ISBN 0 86022 454 6.
Parsloe C J and Wild L, Project Management Handbook for Building
Services, BSRIA, AG 11/98, ISBN 086022 502 X.
Parsloe C J, Pre-Commission Cleaning of Pipework Systems. BSRIA
AG 1/2001.1, ISBN 0 86022 569 0.
Parsloe C J, The Allocation of Design Responsibilities for Building
Engineering Services A Code of Conduct to Avoid Conflict,
BSRIA, TN 21/97, ISBN 086022 474 0.
Pennycook K, Rules of Thumb, UK 4th edition, BSRIA, BG
14/2003, ISBN 086022 626 3.
RIBA, Plan of Work Stages, Royal Institute of British Architects,
1999.
Turner and Simister (eds), 2000, Gower Handbook of Project
Management, Third edition, Gower Training, Aldershot.
42

BSRIA BG 1/2006
APPENDIX A ARRANGEMENT DRAWINGS FOR OUTLINE HVAC DESIGN

BSRIA BG 1/2006
43
44

BSRIA BG 1/2006

BSRIA BG 1/2006
45
46

BSRIA BG 1/2006

BSRIA BG 1/2006
47
APPENDIX B OUTLINE DESIGN HEAT GAIN CALCULATIONS

A summary of data calculated using a computer package. Room G01 refers to the ground floor, F01 to the first floor and S01 to the
second floor.
BSRIA
Old Bracknell Lane West, Bracknell
Berkshire RG12 7AH
Page 1
Date:
Project: Model demonstration
Project no: 70206
Engineer: Sally
Checked by:
Date checked:
File: P:\(N) Construction Practice\70206

Design basis
Location: London
Design day: 15 July
No external shading is considered

Resultant room design temperature used, no room heat losses are added to project totals
Supply temperature for air flow calculation: 14C
Fresh air: None
No fixed temperature air is included

Room inside temperatures are not allowed to rise
Rooms included: Room G01 selected

Project total results
Sun
time
(h)
Outside
temperature
(C)
100
2.00
3.00
4.00
5.00
6.00
7.00
8.00
9.00
10.00
11.00
12.00
13.00
14.00
15.00
16.00
17.00
18.00
19.00
20.00
21.00
22.00
23.00
24.00
145
138
135
138
145
158
174
192
212
233
251
267
280
287
290
287
280
267
251
233
213
192
174
158
Building loads
Fresh air loads
Sensible
(kW)
Latent
(kW)
Sensible
(kW)
Latent
(kW)
Sensible
(kW)
Latent
(kW)
Total
(kW)
000
000
000
000
000
590
1393
1679
5461
5821
5997
5992
5824
6312
6611
6685
6509
6211
2556
1665
220
000
000
000
000
000
000
000
000
000
000
000
222
299
379
455
518
560
575
560
5.18
454
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
590
1393
1679
5461
5821
5997
5992
5824
6312
6611
6685
6509
6211
2556
1665
220
000
000
000
000
000
000
000
000
000
000
000
222
299
379
455
518
560
575
560
518
454
000
000
000
000
000
000
000
000
000
000
000
590
1393
1679
5683
6120
6375
6447
6342
6871
7186
7245*
7027
6665
2556
1665
220
000
000
000
*Peak coincident plant load 7245 kW 16.00 h (518 W/m2 and 185 W/m3)
Supply air flow rate 55712 m3/s. Fresh air flow rate 00000 m3/s (Total 933 occupants)
Hevacomp Design Database, CIBSE gains Version 16.04
48

BSRIA BG 1/2006
Plant total loads

second floor.
BSRIA
Berkshire RG12 7AH
Page 2
Date:
Project no: 70206
Engineer: Sally
Checked by:
Date checked:

Design basis
Location: London
Design day: 15 July

Fresh air: None

Rooms included: Room F01 selected

Sun
time
Outside
temperature
(h)
(C)
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
9.00
10.00
11.00
12.00
13.00
14.00
15.00
16.00
17.00
18.00
19.00
20.00
21.00
22.00
23.00
24.00
145
138
135
138
145
158
174
192
212
233
251
267
280
287
290
287
280
267
251
233
213
192
174
158
Building loads
Sensible
(kW)
000
000
000
000
000
837
1737
2045
6218
6610
6793
6773
6570
7119
7457
7544
7350
7022
2991
1993
365
048
000
000
Latent
(kW)
000
000'
000
000
000
000
000
000
222
299
379
455
518
560
575
560
518
454
000
000
000
000
000
000
Fresh air loads

Sensible
(kW)
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
Latent
(kW)
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
Plant total loads

Sensible
(kW)
000
000
000
000
000
837
1737
2045
6218
6610
6793
6773
6570
7119
7457
7544
7350
7022
2991
1993
365
048
000
000
Latent
(kW)
000
000
000
000
000
000
000
000
222
299
379
455
518
560
575
560
518
454
000
000
000
000
000
000
Total
(kW)
000
000
000
000
000
837
1737
2045
6440
6909
7171
7228
7088
7678
8032
8104*
7868
7476
2991
1993
365
048
000
000

BSRIA BG 1/2006
49

second floor.
BSRIA
Berkshire RG12 7AH
Page 3
Date:
Project no: 70206
Engineer: Sally
Checked by:
Date checked:

Design basis
Location: London
Design day: 15 July
Fresh air: None
Rooms included: Room S01 selected

Sun
time
Outside
temperature
(h)
(C)
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
9.00
10.00
11.00
12.00
13.00
14.00
15.00
16.00
17.00
18.00
19.00
20.00
21.00
22.00
23.00
24.00
145
138
135
138
145
158
174
192
212
233
251
267
280
287
290
287
280
267
251
233
213
192
174
158
Building loads
Sensible
(kW)
000
000
000
000
000
000
563
1780
6688
7966
8907
9474
9635
10223
10353
10005
9175
8081
3385
1587
000
000
000
000
Latent
(kW)
000
000
000
000
000
000
000
000
222
299
379
455
518
560
575
560
518
454
000
000
000
000
000
000
Fresh air loads

Sensible
(kW)
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
Latent
(kW)
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
3
3
Supply air flow rate 86272 m /s. Fresh air flow rate 00000 m /s (Total 933 occupants)
50

BSRIA BG 1/2006
Plant total loads

Sensible
(kW)
000
000
000
000
000
000
563
1780
6688
7966
8907
9474
9635
10223
10353
10005
9175
8081
3385
1587
000
000
000
000
Latent
(kW)
000
000
000
000
000
000
000
000
222
299
379
455
518
560
575
560
518
454
000
000
000
000
000
000
Total
(kW)
000
000
000
000
000
000
563
1780
6911
8265
9286
9928
10153
10783
10927*
10565
9693
8535
3385
1587
000
000
000
000

second floor.
BSRIA
Berkshire RG12 7AH
Page 4
Date:
Project no: 70206
Engineer: Sally
Checked by:
Date checked:

Design basis
Location: London
Design day: 15 July
Fresh air: None
Rooms included: All rooms selected

Room ref
F01
G01
S01
No. off
1
1
1
Peak room loads (Watts)

Sensible
Latent
Total
75 444
66 855
103 526
5 598
5 598
5 745
81 042
72 453
109 272
Time of
peak
Maximum
temperature
16.00 h
16.00 h
15.00 h
24.0C
24.0C
24.0C
Air flow
(m 3/s)
62870
55712
86272

BSRIA BG 1/2006
51

second floor.
BSRIA
Berkshire RG12 7AH
Page 5
Date:
Project no: 70206
Engineer: Sally
Checked by:
Date checked:

Design basis
Location: London
Design day: 15 July
Fresh air: None

Sun
time
(h)
Outside
temperature
(C)
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
9.00
10.00
11.00
12.00
13.00
14.00
15.00
16.00
17.00
18.00
19.00
20.00
21.00
22.00
23.00
24.00
145
138
135
138
145
158
174
192
212
233
251
267
280
287
290
287
280
267
251
233
213
192
174
158
Building loads
Fresh air loads
Sensible
(kW)
Latent
(kW)
Sensible
(kW)
Latent
(kW)
Sensible
(kW)
Latent
(kW)
Total
(kW)
000
000
000
000
000
1428
3694
5504
18367
20398
21697
22239
22029
23653
24421
24235
23035
21313
8932
5245
585
048
000
000
000
000
000
000
000
000
000
000
667
896
1136
1364
1554
1680
1724
1679
1553
1363
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
1428
3694
5504
18367
20398
21697
22239
22029
23653
24421
24235
23035
21313
8932
5245
585
048
000
000
000
000
000
000
000
000
000
000
667
896
1136
1364
1554
1680
1724
1679
1553
1363
000
000
000
000
000
000
000
000
000
000
000
1428
3694
5504
19034
21294
22833
23603
23582
25333
26145*
25914
24588
22677
8932
5245
585
048
000
000
52

BSRIA BG 1/2006
Plant total loads
APPENDIX C OUTLINE DESIGN HEAT LOSS CALCULATIONS

second floor.
BSRIA
Berkshire RG12 7AH
Date:
Project no: 70206
Engineer: Sally
Page 1
Checked by:
Date checked:

Outside design temperature: -4C
Main system of heating: Forced warm air down from high level
Project total heat losses
Total exposed wall loss
Total exposed roof loss
Total exposed floor loss
Total internal floor loss
Total internal ceiling loss
Total internal partition loss
Total window loss
Total rooflight loss
Total infiltration loss
Total fixed air loss
Total heat loss
Total floor area
Total volume
Pre-heat period (h)
0
1
2
3
4
5
6
7
8
4200 m2
11 760 m3
5 019
39 013
14 929
0
0
14 311
51 651
0
53 367
0
-178 290
4245
1516
W
W
W
W
W
W
W
W
W
W
W
W/m2
W/m3
Plant ratio
Plant boosted output

(kW)
154
149
144
140
136
132
128
124
120
27535
26622
25761
24948
24179
23450
22758
22101
21476
Hevacomp Design Database, Heat loss Version 16.04

BSRIA BG 1/2006
53

second floor.
BSRIA
Berkshire RG12 7AH
Page 2
Date:
Project no: 70206
Engineer: Sally
Checked by:
Date checked:

Room reference: G01
No. off: 1
Room name: Ground floor open plan office area

Temperatures:
Resultant: 220
Environmental: 217
Air: 229
Mean radiant: 211
System: Double/triple panel radiators
Total room heat loss: 56 270 W (402 W/m2 144 W/m3)
Surface
Area
U Value
Temperature
difference
Heat Loss (W)
22 surfaces
Total results shown
38 686 W
Total fabric loss

Infiltration loss
38 686 W
17 583 W
Total heat loss
56 270 W
54

BSRIA BG 1/2006

second floor.
BSRIA
Berkshire RG12 7AH
Page 3
Date:
Project no: 70206
Engineer: Sally
Checked by:
Date checked:

Room reference: F01
No. off: 1
Room name: First floor open plan office area

Temperatures:
Resultant: 220
Environmental: 217
Air: 228
Mean radiant: 212
Total room heat loss
Surface
41 291 W
Area
(295 W/m2
U Value
105 W/m3)
Temperature
difference
Heat Loss (W)
21 surfaces
Total results shown
23 801 W
Total fabric loss

Infiltration loss
23 801 W
17 490 W
Total heat loss
41 291 W

BSRIA BG 1/2006
55

second floor.
BSRIA
Berkshire RG12 7AH
Page 4
Date:
Project no: 70206
Engineer: Sally
Checked by:
Date checked:

Room reference: S01
No. off: 1
Room name: Second floor open plan office area

Temperatures:
Resultant: 220
Environmental: 213
Air: 240
Mean radiant: 200
Surface
Area
U Value
Temperature
difference
Heat Loss (W)
21 surfaces
Total results shown
62 435 W
Total fabric loss

Infiltration loss
62 435 W
18 294 W
Total heat loss
80 729 W
56

BSRIA BG 1/2006
APPENDIX D OUTLINE DESIGN

BSRIA BG 1/2006
57
58

BSRIA BG 1/2006

BSRIA BG 1/2006
59
60

BSRIA BG 1/2006
Hoval AtmoGas AG
Technical and performance details.
Models AG55 to AG118.
Model
AG55
AG64
AG71
AG82
AG91
AG100
AG109
AG118
Output
(kW)
27-54
36-63
36-70
45-81
45-90
54-99
54-108
54-117
863
946
1113
1113
1280
1280
1447
1447
952
952
1007
1007
1007
1007
1007
1007
102
102
124
124
124
124
124
124
75
75
159
75
159
75
159
75
452
494
536
578
619
661
703
745
Flue
O/D
180
180
180
200
200
200
220
220
Dry weight
(kg)
230
257
283
305
334
357
386
408
Water content
(litres)
29
33
36
40
44
47
51
54
Water flow rates (l/s)

Hydraulic resistance (mbar)
No. of
Fuel consumption
3
sections
Nat gas (m /hr)
T = 11k
T = 20k
T = 11k
T = 20k
AG55
7
117
064
19
55
625
AG64
8
137
075
46
14
729
AG71
9
152
084
80
24
810
AG82
10
176
097
99
30
935
AG91
11
195
107
132
40
1038
AG100
12
215
118
179
54
1141
AG109
13
235
129
214
65
1243
AG118
14
254
140
264
80
1346
Note: Above figures are based on maximum boiler output. All models are supplied for high/low operation, natural gas (with propane option)
Model
Electrical supply
Boiler operating data
Gas data
Electrical supply 230 V 50 Hz 6 A
Maximum working pressure 60 bar

Test pressure
90 bar
Max. operating temperature 90C
Min. water flow rate is boiler rating l/s
90
Minimum gas supply pressure at inlet:

Natural gas 20 mbar, Propane 37 mbar
(A propane conversion kit is available for
all models).
Pump overrun
A pump overrun of 5 minutes is
recommended on boiler shutdown.
See separate AtmoGas AG leaflet for boilers 120-380 kW.

BSRIA BG 1/2006
61

30GTN
Air-Cooled Chillers
With ComfortLink Controls
40 to 400 Nominal Tons
Size
040
045
050
060
070
080
090
100
110
130
150
170
190
210
230
245
255
270
290
315
330
360
360
390
420
Cooling capacity
60 Hz
50 Hz
kW
Tons
kW
Tons
126
36
123
35
146
42
147
42
177
50
180
51
223
63
212
60
255
73
247
70
288
82
285
81
307
87
319
91
350
100
352
100
378
108
390
111
426
121
434
124
493
140
509
145
553
157
557
158
606
172
632
180
707
201
703
200
781
222
805
229
802
228
787
224
844
240
866
246
900
256
909
259
988
281
1022
291
1087
309
1097
312
1106
314
1113
317
1213
345
1188
338
1313
373
1335
380
1414
402
1407
400
Length
mm
ft
2515
83
2515
83
2515
83
3124
103
3124
103
3429
113
3429
113
4267
140
4267
140
5944
195
5944
195
5944
195
6858
225
6858
225
9354
307
9354
307
11887
390
10221
335
12527
411
11906
391
11906
391
13716
450
12815
420
13716
450
13716
450
Dimensions
Width
mm
ft
2261
74
2261
74
2261
74
2261
74
2261
74
2337
77
2337
77
2337
77
2337
77
2337
77
2337
77
2337
77
2337
77
2337
77
2337
77
2337
77
2337
77
2337
77
2337
77
2337
77
2337
77
2337
77
2337
77
2337
77
2337
77
Height
mm
ft
2261
74
2261
74
2261
74
2261
74
2261
74
2438
80
2438
80
2438
80
2438
80
2438
80
2438
80
2438
80
2438
80
2438
80
2438
80
2438
80
2438
80
2438
80
2438
80
2438
80
2438
80
2438
80
2438
80
2438
80
2438
80
Carrier Corporation A member of the United Technologies Corporation Family Stock Symbol: UTX Copyright 2005
Legal Notice Privacy Policy
62

BSRIA BG 1/2006
APPENDIX E FINAL PROPOSALS/PRODUCTION INFORMATION

DATA FOR HEAT GAIN AND HEAT LOSS CALCULATIONS
Heat gain data
Item
Latitude
Height above sea level

Design month
Design day
Sky clarity
Ground reflection
Plant start time
Plant stop time
Occupancy start time
Occupancy finish time
Air/dry resultant temperature
Outside maximum dry bulb temperature (C)
Outside minimum dry bulb temperature (C)
Outside maximum wet bulb temperature (C)
Outside minimum wet bulb temperature (C)
Sensible heat (W/person)
Latent heat (W/person)
Room design temperature (C)
Infiltration (Air changes per hour)
Percentage saturation
Occupancy (m2 per person)
Small power W/m2
Fresh air (l/s/person)
Room height (m)
Window height (m)
Walls U value
Windows U value
Floor U value
Value
Source
517
0
7
15
095
02
0600
2000
0900
1800
Air
29
13
20
11
80
60
24
02
50
15
20
12
28
28
039
28
041
Heat loss data

Item
Outside air temperature (C)
Room design temperature (C)
Infiltration (Air changes per hour)
Fresh air ventilation (l/s/person)
Occupancy (m2 per person)
Room height (m)
Window height (m)
Walls U value
Windows U value
Floor U value
Value
Source
-4
22
02
12
15
28
28
039
28
041

BSRIA BG 1/2006
63
H4
APPENDIX F CONDENSATION RISK H4

This appendix gives an example of the condensation risk
calculation for the roof structure of the demonstration project
building.
As discussed on page 13 (calculating U values), the architect
should supply the fabric details of the building. These details (for
each element) required for condensation calculations include:
Description of element
fabric thickness (d)
conductivity ()
vapour resistivities (r)
internal/external surface resistances (R)
internal/external conditions (C and %RH).
Step 6. With the details given the thermal resistance (R) and
vapour resistance (G) can be determined for each element using
R=
d
and G = d r
D
Step 7. The nodes of the roof construction (the point where

two elements meet) are shown in the figure below, with a
description of the two elements that meet in the table
following. It is at each of these points where the temperature
and vapour pressure needs to be calculated.
Figure 17: The nodes of the roof construction.
The table below gives this data for the roof.

Table 18: Roof element conductivities and vapour resistivities.
Fabric
element
Thickness
(m)
Felt/
bitumen
Glass fibre
quilt
Cast concrete
Conductivity
(W/mK)
Vapour
resistivities
(GNs/kgm)
CIBSE
Guide A
ref.
0005
05
15000
Table 3.48
01
004
Table 3.49
021
113
115
Table 3.50
Internal surface resistance (mK/W)

External surface resistance (m2K/W)
Internal temperatures/humidity
Rsi = 01171
Rse = 004
22C, 50%rh
External temperatures/humidity
-4C, 100%rh
Figure 16: Roof element conductivities and vapour resistivities.
Table 19: Description of the nodes of the roof construction.

Node
Description
Inside/cast concrete
Cast concrete/glass fibre quilt
Glass fibre quilt/felt bitumen
Felt bitumen/outside
Step 2. (continued on next page)
(See Table 20.)

Table 20: Thermal and vapour resistances of roof elements.
Fabric element
Thickness (m)
Conductivity
(W/mK)
0005
05
Glass fibre Quilt
01
004
25
05
Cast concrete
021
113
0186
115
2415
Internal surface
01171
N/A
Elemental sums for thermal and vapour resistance
2853
External surface
Felt/ Bitumen
64

BSRIA BG 1/2006
Thermal resistance
2
(m K/W)
Vapour resistivity
(GNs/kgm)
Vapour resistance
(GNs/kg)
004
N/A
001
15000
75
997
H4

Step 2. (continued)
Now the temperature and vapour pressure from inside to
outside through each node can be calculated. The equations
used are:
t n = t si
(t si
t se
) R
n
si
se
R
si
and
P vn = P vsi
Similar calculations are made to arrive at the temperatures and

vapour pressures at all other nodes in the roof construction, and
these are summarised in Table 21. Temperatures have been
quoted to one decimal place and pressures to two decimal
places. More accuracy can be maintained during calculation,
but there is no reason to quote final figures to three or four
decimal places.
(P vsi
P vse
G
si
se
G
si
Where:
tn = Temperature at node n
tsi = Internal surface temperature = 22C
tse = External surface temperature = -4C
Table 21: Summary of temperatures and vapour pressures (Pv) at

nodes within roof construction.
Node
Inside
Node 1
Node 2
Node 3
Node 4
Outside
Description
t (C)
Pv (kPa)
22
Cast concrete/glass fibre
quilt
Glass fibre quilt/felt bitumen
Felt bitumen/outside
209
192
134
134
112
-35
-36
-4
112
043
043
R = Sum of thermal resistances from (and including) the

si
internal surface to node n
se
R = Sum of thermal resistances from the internal surface to

si
the external surface = 2853 m2K/W
Pvn = Vapour pressure at node n
Pvsi = Internal vapour pressure at 22C, 50%rh = 1339 kPa
Pvse = External vapour pressure at -4C, 100%rh = 0437 kPa
Properties of humid air are given in various sources such as

section 1 of CIBSE Guide C, and on the CIBSE
psychrometric charts.
n
G = Sum of vapour resistance from (and including) the

si
internal surface to node n
se
G = Sum of vapour resistance from internal to external

si
surface = 997 GNs/kg
Using the temperature equation above, the temperature at

node 1 is as follows:
t 1 = 22
(((22 -4 ) 0 1171 ) 2.853 ) = 20 85C
Where:
1
R = 0 1171m K/W
si
The vapour pressure at node 1 is the same as the internal vapour

pressure as there is no surface effect. The vapour pressure at
node 2 is:
Pv2 = 1 339 (((1 339 0 4371 ) 24 15 ) 99.65 )
= 1 120kPa
Step 8. The saturated vapour pressure at each node can now

be determined. If the tables are used to identify the saturated
vapour pressure, it is likely that some interpolation will be
required. This example uses the tables in section 1 of CIBSE
Guide C, the tables give properties of humid air from -10C to
60C in 05C increments.
The following values are used in determining the saturated

vapour pressure at the temperature of each node.
Table 22: Saturated vapour pressures (Ps) from CIBSE Guide C
suitable for interpolating values for the roof construction.
T (C)
21
20
19
18
17
-3
-4
Ps (kPa)
2486
2337
2196
2063
1936
04756
04371
The saturated vapour pressure (Ps ) for the inside temperature

can be read straight from the property tables: 22C, 100% rh
gives Ps = 2643 kPa.
To interpolate the Ps values from psychrometric tables, use the
nearest values above and below the node temperature and the
corresponding saturated vapour pressures at those temperatures.
There are different ways to interpolate values; in general it is the
approach that changes but the principles stay the same.
Where:
2
G = 0 + 24 15 = 24 15 GNs/kg
si

BSRIA BG 1/2006
65
H4

Step 3. (continued)
One way to interpolate the saturated vapour pressures for each

node is to use the equation below with the nearest temperature
(and corresponding Ps ) above the node temperature (upper
temperature and upper Ps ) and the nearest temperature (and
corresponding Ps ) below the node temperature (lower
temperature and lower Ps )
Psn = saturated vapour pressure of node
Psn = lower Ps + (((temp. of node lower temp) (upper
temp. lower temp.)) (upper Ps lower Ps))
Node 1
t1 = 20932C interpolate between 21C and 20C
21C has a Ps of 2486 kPa
Therefore:
Ps1 = 2337 + (((20932 20) (21 20)) (2486 2337))
= 2476 kPa
Node 2
t2 = 19239 C interpolate between 20C and 19C
Therefore:
Ps2 = 2196 + (((19239 19) (20 19)) (2337 2196))
= 2230 kPa
Table 23: Temperature, vapour pressure and saturated vapour

pressure at roof construction nodes.
Node
Description
Inside
t (C)
Pv (kPa)
Ps (kPa)
22
134
264
Node 1
209
134
248
Node 2
Cast concrete/
glass fibre quilt
192
112
223
Node 3
Glass fibre quilt/

felt bitumen
-35
112
046
Node 4
Felt
bitumen/outside
-36
044
045
-4
044
044
Outside
The highlighted figures in the table show where the calculated

vapour pressure exceeds the saturated vapour pressure,
indicating condensation.
Step 9. The calculated values of Pv and Ps for each node are
shown in Figure 18. When the calculated value of Pv exceeds Ps
then condensation will occur. In reality, the saturated vapour
pressure Ps cannot be exceeded and the actual vapour pressure at
this point will equal the saturated vapour pressure. At this
point, the calculations need to be repeated in more detail by
sub-dividing the construction at this point.
Figure 18: Graph of vapour and saturated vapour pressure at roof

construction nodes.
Node 3
t3 = -3544C interpolate between -3C and -4C
-3C has a Ps of 04756 kPa
Therefore:
Ps3 = 04371 + (((-3544 -4) (-3 -4)) (04756
04371)) = 0455 kPa
Node 4
t4 = -3635C interpolate between -3C and -4C
Therefore:
Ps4 = 04371 + (((-3635 -4) (-3 -4)) (04756
04371)) = 0451 kPa
Table 23 gives the calculated temperature, vapour pressure and
the saturated vapour pressure at each node, again quoted to one
and two decimal places.
66

BSRIA BG 1/2006
For the roof construction condensation occurs at node 3, which

is between the glass fibre quilt and the felt/bitumen layers. The
layers will therefore be separated into two sub-constructions,
the first being from the inside to node 3, the second being from
node 3 to the outside, see Figure 19.
H4

Step 4. (continued)
Sub-construction 2
Figure 19: Sub-constructions within roof fabric.
The process is repeated again using the same equation. Note

that node 3 is the inside of the sub-construction, so where si is
used within the equation this refers to node 3. Likewise the
vapour pressure at node 3 is saturated so Pvsi=Ps3.
In this case all the vapour pressures are already known. Pv3 is
the saturated vapour pressure. Pv4 and Pvse are the external
vapour pressure.
The Pv values through the sub-divisions are collated in Table 24
and shown in Figure 20 (node 3 is not repeated). This
construction has few layers, and it can be seen from the table
that condensation will only occur at node 3 (and outside, but
this is not a problem). If a construction has many layers and the
vapour pressures are shown graphically then it may be useful to
expand the potential problem areas. This has been done in
Figure 21 which is an expansion of Figure 20 from node 3 to
outside.
It is only the vapour pressure at each node that needs to be

calculated, the temperatures and saturated vapour pressures will
remain the same as those already calculated. The vapour
pressure calculations are now repeated but each sub-division is
considered separately.
Table 24: Vapour pressure and saturated vapour pressure

distribution through sub-divisions.
Node
Description
t (C)
Pv (kPa)
Ps (kPa)
22
1339
2643
Sub-construction 1
Inside
Sub-construction 1
Node 1
20932
1339
24797
Using:
Node 2
Cast concrete/
glass fibre quilt
19239
04725
2220
Node 3
Glass fibre quilt/

felt bitumen
-3544
04546
04546
P vn = P vsi
(P vsi
P vse
G
si
se
G
si
Node 3 is the external node of the sub-construction so where

se is used within the equation this refers to node 3. Also as the
vapour pressure at node 3 is saturated Pvse= Ps3.
The vapour pressure at node 3, node 1 and inside are all known
so only the vapour pressure at node 2 needs to be determined.
Pv at Node 2
Sub-construction 2
Node 3
Glass fibre quilt/

felt bitumen
-3544
04546
04546
Node 4
Felt
bitumen/outside
-3635
04371
04511
-4
04371
04371
Outside
Figure 20: Graph of vapour pressure and saturated vapour pressure

distribution through sub-divisions.
Pv2 = 1 339 (((1 339 0 4546) 24 15) 24 65)

= 0 4725 kPa
Where:
n
si
si
se
si
si
G = G = 24 15
and
G = G = 24 65GNs / kg

BSRIA BG 1/2006
67
H4

Figure 21: Graph of vapour pressure and saturated vapour pressure
from node 3 to outside.
Rate of condensation
The rate of condensation at the saturated node can now be
calculated. This is the difference between the moisture flowing
through sub-construction 1 and the moisture flowing through
sub-construction 2.
The following equation for mass flow rate per unit area is used:
q m = (Pv Pvn ) G
se
si
For sub-construction 1:
qm1 = (1339 04546) 2465 = 00359 kPakg/GNs
= 359 10-8kg/m2s
For sub-construction 2:
qm2 = (04546 04371) 75 = 00002 kPakg/GNs
= 2 10-10kg/m2s
qm1 qm2 = 357 10-8kg/m2s = 013g/m2h
With the amount of condensation calculated there are two
options. First, the designer can calculate the rate of evaporation
that occurs in warmer weather to see if this is sufficient to offset
the condensation. Alternatively, the fabric construction can be
altered in order to prevent condensation occurring. Normally
this would be a vapour barrier at the node where the
condensation would occur.
68

BSRIA BG 1/2006
APPENDIX G EXECUTIVE SUMMARY FROM ENGINEERING DESIGN

Calculations and the use of margins
The executive summary from a CIBSE Research Report,
published in 1998, is reproduced below, in an edited form with
the permission of the CIBSE.
Executive summary
This report discusses the current use of design margins in
building services engineering design calculation based on
primary research of current industry practice. It covers both the
magnitude of margins used in practice and the justifications
given for their use. The research also looks in detail at the
various issues connected with over-engineering in the building
services industry, and in construction in general, such as overdesign, over-specification and over sizing. Defines these and
derives a model showing these various issues in the context of
the overall building process.
It is suggested that additive design margins can contribute to

over-engineering. The research also identifies the specific areas
to which design margins could contribute. It also takes an
overview of other industry initiatives to combat the problem
such as value engineering and feedback.
Detailed definitions of various kinds of margins are given,
together with a review of the range of margins used in practice.
Nine different types of design margins are identified and
recommendations made for action to reduce the use of
unnecessary margins, with specific action to be taken by both
client and designer.
A margin can be defined as an amount allowed beyond what is
needed or an allowance for contingencies. For the purposes of
this report the term design margin is used in the broadest sense
to mean any percentage increase added to a design value,
parameter or calculations result whether a deliberate and valid
design decision, a contingency or safety factor, or an inadvertent
addition caused by, for example, selecting the next size up of a
plant item.
The research report is intended to make designers, clients,
project managers and others involved in the design and
specification of building services for new or refurbished
buildings aware of the potential contribution of excessive design
margins to the problem of over-engineering and over-sized
plant, and to provide guidance and procedures that can be used
to review the design margins. This should aid communication
between client and designer and assist the client in the
development of both risk management and value engineering
strategies for a project.
Conclusions
In conclusion, design margins can be seen as part of the wider
problem of over-engineering, coupled with design deficiencies
and a lack of feedback to design.
There would seem to be substantial evidence of over-design

and consequent over-sizing and inefficient operation.
Buildings are still not delivering their expected performance.
The type of client bespoke or speculative, has a substantial

impact on the approach to design and the avoidance of
over-engineering.
Good briefing is the key to good design, closely followed by

good design feedback.
An absence of complaints cannot be taken as evidence of

good design.
Fear of litigation or liability mitigates against effective design

certainly against it being published or more widely
disseminated.
Over-design and over-engineering have often been hidden

in the past by poorly constructed buildings with higher than
recommended air-leakage rates.
There is over-reliance on complex control systems to

deliver efficient system operation at low loads and
compensate for over-sizing.
There is a lack of appreciation of system performance in the

normal operating range such as low load operation. This is
exacerbated by the current fee structure which does not
allow designers sufficient time to fully analyse expected
system performance.
There is a lack of appreciation of the interaction between

use and system in the occupied space.
There is a lack of appreciation of the different between the

system design concept and actual operating information.
There is no effective method to easily evaluate system

performance in use to identify over-sizing. Deficiencies are
usually only obvious when systems under-perform.
The following list shows the different types of margins that have
been identified during the course of this research.
Use of margins
To allow for uncertainties in the initial design assumptions.
These vary from uncertainties linked to the client brief to
uncertainties in the selection of design temperatures to
variations in allowances for internal heat gains.
To allow for actual uncertainties in building performance.

These are added to allow for variations in U values due to
moisture content, variation in material and the effect of
actual construction method or assembly. They are also used
to allow for variations in the actual infiltration rate to the
building, degree of airtightness and the integrity and quality
of the fabric construction.
To allow for uncertainties in the calculation methods.
To allow for uncertainties in equipment performance, such

as in expectation of a variation in actual installed
performance from that quoted from manufacturers test data.
To allow for uncertainties in system performance,

anticipation of variation in actual system performance from
design caused by minor changes due to practicalities of
installation and the combined performance of the
component parts and controls.

BSRIA BG 1/2006
69
APPENDIX G EXECUTIVE SUMMARY FROM ENGINEERING DESIGN

Calculations and the use of margins
As genuine design or safety margins to enable the plant and

system to operate as designed, to allow for pre-heat or to
enable a ventilation system to be commissioned to current
standards.
To meet client requirements for plant redundancy or

future-proofing and spare extra capacity to allow for future
changes or expansion.
To allow for deterioration of system performance over time

such as fouling factors, allowance for dirty filters, reduced
heat output and reduced light output.
From habit or rote, custom and practice or rule of thumb.
Recommendations
Margins should not be added to design unless there is a valid
and justified design reason for their use.
Where margins are deemed necessary, they should always be

clearly identified within calculations and the justification for
their use noted. This should be approved by the client and
appropriate steps taken to reduce the use of margins where
possible by validation and testing exercises. Agreement to
this should form part of the briefing process.
Steps should be taken to avoid the occurrence of cumulative

margins. The margins used throughout a calculation should
be reviewed at the end to identify possible duplication and
any excess margins should be removed.
Designers should clearly specify the building performance

assumptions made as part of the design. It should then be
the clients responsibility to ensure, using performance
testing as appropriate, that the building meets its
performance criteria, thus allowing systems to deliver their
stated performance.
Designers should set out clearly for the client the operating
limits of the design and ensure the client is aware of, and
satisfied with, the anticipated real performance of the
system. The risk of under-performance needs to be made
explicit and set in context.
Designers should have accurate information available on

product performance, this could be achieved by product
performance certification such as Eurovent or ARI, but not
CE marking.
System operation at part-load conditions should be

investigated as part of the control design strategy and
measures taken to improve efficiency and reduce poor
performance at low load conditions. Ideally, design analysis
against real weather data should show expected proportions
of annual operation at different load percentages. Plant
selection should be optimised to the year-round operating
pattern of loads.
There should be an effective technical design quality

assurance system in design organisations to check the
appropriateness of initial design assumptions, parameters and
methodologies. The should include a review of the
appropriateness of the design margins used.
70

BSRIA BG 1/2006
There should be adequate technical benchmarking

information derived from actual system operation available
to assist technical quality assurance procedures and in the
initial design process.
There needs to be effective quality control of the actual

technical design process and procedures. As a minimum
requirement there should be checks on input data and on
staged calculation outputs and comparisons with
benchmarking data where available.
Organisations should hold in-house post-project reviews as

part of their quality process. The review must include
errors as well as successes. Lessons learnt from these need to
be effectively disseminated throughout the organisation and
the information incorporated into design guidance
information and databases.
Whatever your
building services
requirement contact
BSRIA Limited:
T: +44 (0)1344 465600
F: +44 (0)1344 465626
E: bsria@bsria.co.uk
W: www.bsria.co.uk
Old Bracknell Lane West,
Bracknell, Berkshire,
RG12 7AH, UK
Offices in Bracknell, Cadiz and Toulouse

BG1 2006

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

BG1 2006

Uploaded by

Copyright:

Available Formats

Applying HVAC building

BSRIA Guide BG 1/2006

See Guide to HVAC Building Services Calculations, BG 30/2003

ISBN 0 86022 661 1

Printed by Multiplex Medway

MODEL DEMONSTRATION PROJECT

See Guide to HVAC Building Services Calculations, BG 30/2003

1 Introduction to the building process

2 The demonstration project

3 Outline and detailed proposals stage

4 Final proposals/production information stage

B Outline design heat gain calculations

C Outline design heat loss calculations

E Final proposals/production information

G Executive summary from engineering design

MODEL DEMONSTRATION PROJECT

See Guide to HVAC Building Services Calculations, BG 30/2003

1 INTRODUCTION TO THE BUILDING PROCESS

After practical completion

Mechanical and electrical designers may be appointed, either to

Cost estimates are calculated from a properly defined

ACE Conditions (2002)

Project management techniques particular to building services

Table 1: RIBA Plan of Work stages.

There are three critical success factors for projects to remain

The completed building provides the functional requirements

What the completed project is for

MODEL DEMONSTRATION PROJECT

See Guide to HVAC Building Services Calculations, BG 30/2003

1 INTRODUCTION TO THE BUILDING PROCESS

The process by which a client informs others of needs,

Briefing is an essential and important part of the project process.

MODEL DEMONSTRATION PROJECT

In situations like this, the buildings success will be determined

The clients business function

The statement of needs is entirely in the hands of the client and

See Guide to HVAC Building Services Calculations, BG 30/2003

1 INTRODUCTION TO THE BUILDING PROCESS

Details of the project team

The consolidated brief develops alongside the proposals from

Appointment and duties of the building services

Fee tender and qualifications, whereby a designer is selected

Precise services routes

Design ideas competition, whereby a designer is chosen

MODEL DEMONSTRATION PROJECT

See Guide to HVAC Building Services Calculations, BG 30/2003

1 INTRODUCTION TO THE BUILDING PROCESS

Obtain information regarding utility services to the site

Prepare detailed design drawings

Assemble documents for tender

Activities not defined by ACE, but important to performance

Obtain and inform the clients brief

ACE Agreement B2 Detailed design service*

RIBA work stage

* Summarised by Fulcrum Consulting

The new BSRIA guide BG 6/2006: A Design Framework for

MODEL DEMONSTRATION PROJECT

Conflicts can arise when building services designs are based on

See Guide to HVAC Building Services Calculations, BG 30/2003

1 INTRODUCTION TO THE BUILDING PROCESS

In turn, this may lead to a series of subsequent revisions. Such

There are many instances where the expertise of the building