Sound guidance

The Institution knows these issues well, having ridden the rocky rapids over revisions to Guide A. Now the important sections of Guide B are up for a make-over, and the CIBSE is keen to ensure that the changes are an appropriate and justifiable improvement over existing requirements.

Three sections are being revised: Guide B2, B3 and B12. B2 covers the requirements for ventilation and air conditioning; B3 covers ventilation and air conditioning systems; and B12 deals with noise and vibration control.

The BRE and Oscar Faber are taking responsibility for sections B2 and B3, while revision of B12 is being managed by acoustics consultant Dr Geoff Leventhall.

The Institution has already carried out broad-based consultation on the changes, and created a task-group comprising manufacturers, practitioners and researchers to provide advice. Given that it's been some time since Guide B was updated, the revision authors not only have to consider content, but also how the new edition will link to other publications, particularly those produced by ASHRAE.

The target audience may also be wider, such as architects, and reach out beyond traditional building services into other sectors. For instance, the current Guide B is quite comprehensive on animal farms, but less so on theatres.

Given that the last few years have seen a plethora of technical memorandums and guidance documents on ventilation, the next Guide B will need to be positioned to avoid overlap and repetition. It will also have to be consistent across all sectors and technologies. Indeed, on that basis there is a fundamental question whether Guide B should remain a textbook or become a practical manual. The BRE and Oscar Faber would like readers' opinions on that.

Guide B2/B3 – the suggested changes

Among its many recommendations, the Guide B task-group suggested the following revisions to sections B2 and B3:

• include ventilation system design and control for transient spaces (airport lounges, shops and fitness rooms, etc) and non-sedentary activities;

• place less emphasis on allowing for needs of smokers;

• include ventilation of plantrooms;

• include more on natural ventilation system design and night cooling;

“The existing Guide shows a straight-line relationship between boiler output and ventilation requirements, so beyond 1 MW the ventilation rates become ridiculous” – Andrew Pascal

• include more on mechanical ventilation and mixed mode design;

• increase emphasis on car parks, computer rooms, dwellings and sports centres;

• include museums, art galleries, libraries, theatres and conference rooms.

The task-group also asked for Guide B to cover control strategies, not only for systems but for occupant controls. Additionally, the Guide should recognise the increased use of thermal mass in controlling internal temperature, and take note of emerging energy sources like photovoltaics.

The big question which the Guide's authors will need to address is how much information it will provide to designers. Will it just contain basic facts, leaving it to the raft of technical memorandums produced by BRE, BSRIA and the like to supply the details, or should it give a level of information that enables a designer to narrow down the options? In discussion, the CIBSE workshop delegates voiced opinions on structuring the advice to suit an architectural audience as well as the traditional services engineering sector, and requests for the Guide to carry advice on specific fan power and airtightness.

"We have an opportunity here to define specific fan power to show how designers can get to 1 W/litre/s," said Derrick Braham. "Air leakage also needs to be included, because if it isn't then room ventilation will not work as designed, especially the displacement types.

"We also need to define whether the ventilation system is balanced and at what point perimeter heating becomes necessary." Casimir Iwaszkiewicz of Construction Resources suggested that the Guide would benefit from a graphical representation of when and where systems "fall over".

"A coloured chart showing the boundaries of system performance through high infiltration or low insulation for example, would show when systems become unusable," he said.

FBE's Paul Evans agreed. "The Guide needs to be a more visual document than before," he said, "to help people like architects during the selection process. The current Guide is very wordy [sic] when it should be more strategic." Delegates told the authors that the ventilation guidance in B2 will also need to be in agreement with that published elsewhere in the Guide and in CIBSE AM10.

WSP's Andrew Pascal said that much of the combustion ventilation guidance in Guide B "is way out of date" and based on old heat gain data from boiler manufacturers. "Boiler technology has moved on, and the insulation is much better," said Pascal. "But the existing Guide shows a straight-line relationship between boiler output and ventilation requirement, so beyond 1 MW the ventilation rates become ridiculous.

"We need a non-linear graph in section B13, and advice that is consistent with other parts of the Guide," he concluded.

There was considerable debate on the kinds of buildings Guide B should cover. Call centres for example, are a growing building type, and many thought the Guide should give particular guidance on deep plan, densely loaded, trading floors.

“...no matter how venerable the NR system, most people think in dBA” – Richard Galbraith

Section B12: Sound control

Dr Geoff Leventhall is heading up the revision of the acoustics section of the guide, and this too provoked considerable debate, particularly as the ASHRAE has made significant changes to its sound guidance.

"There have been strong suggestions that Guide B12 should be ditched and we should use ASHRAE data," revealed Leventhall. "but then again, ASHRAE is not happy with its data either, and is continually updating it".

Leventhall concluded that the main requirements of a revamped section B12 of the Guide are a modern noise prediction method, data to support that method, and criteria to assess both the results of predictions and the results of measurements.

"We have to consider what frequency range is needed," said Leventhall. "A lot of CIBSE data starts at 125 hz and goes up, but ac equipment produces substantial noise at 63 hz and lower, so perhaps we need to extend the frequency range downwards.

"We also need to know the source levels (particularly from noisy sources like fans), the numerical values at each of the frequencies we are going to use for either the attenuation of the duct and duct fittings, or the regeneration by duct components," added Leventhall. "We also need to know what happens with the sound when it gets out from the terminal unit into the room, and account for room effects." A debate followed on the merits and demerits of NR and NC curves. Leventhall revealed that the Noise Criterion method used by the CIBSE was never intended for the use to which it was subsequently put.

"The data on which it is based covers very low frequency noise, and the spectra drop off quickly below 125 hz and sometimes 250 hz, so the data is not useful for air conditioning," he explained.

Last year the ASHRAE adopted the Room Criterion method along with a quality assurance index which enables designs to assess whether the noise is rumble, roar or hiss. By judging the quality of the noise you can get a feel whether there will be a problem.

The big question is whether those developments are appropriate for the UK. As Leventhall said, US air conditioning systems tend to produce more low frequency noise, but if UK installations don't, then what is the need for a method to test for it? Acoustics consultant Richard Galbraith argued cogently for replacing the NC and NR curves with something that clients and lay designers can easily understand. The obvious sound assessment tool, he said, is dBA .

"Everybody knows what it means. The reality is that no matter how venerable the NR system, most people think in dBA. So it may be better to go with dbA and develop a 'fudge factor' that attaches rumble or screech to it.

"That must be progress, because in the end it's about good communication between clients and those in the know." Geoff Leventhall wondered whether dBA was too coarse a measure than having NR with third octave bands, but Galbraith took the view that tagging the noise with a correction factor, like rumble, would get round that shortcoming. "Surely it is not beyond our wit to devise a method which can deal with that," he said.

The conclusion reached by the B12 working party was to devise two sets of noise assessment criteria: one set to design to (which might involve frequency conversion using NR or another method) and one for commissioning, which needs to be a simple as possible for clients and property developers to understand.

The Guide B workshop is but one stepping stone on the path to completing the revisions to Guide B. Readers wanting to get more details on the proposed structure of the Guide, or who wish contribute to the revisions, should contact Hywell Davies on 01234 351269 or Denice Jaunzens at the BRE on 01923 664522.