Margins for error

Many engineers believe that designing adaptable or flexible building services and oversizing plant go hand in hand – we must design-in large safety margins to create space plant capacity for future requirements. But how many of these anticipated building changes actually happen? Even if they don't, is oversizing plant still a way of achieving good adaptability?

As well as increased capital costs, designing spare services capacity can detrimentally affect system performance and increase energy costs. It is therefore extremely difficult to know how safe to play it.

The CIBSE recently surveyed building owners and designers on the requirements and consequences of designing for adaptable building services. Building owners said unequivocally that adaptability was more important when their organisation has developed new premises or purchased an existing building (53% said of major importance, 23% said it was vital).

What clients want are adaptable buildings that can respond to changes in use, in a way that is cost and time-effective. What they don't want are buildings that restrict change – such as the low floor to ceiling heights that restricted IT expansion during the 1980s. Nor do they want poorly designed central building intelligence systems that have to be reconfigured – at great expense – to cater for simple changes.

In the short-term, office buildings need to be flexible enough to cater for physical changes in the working environment like repositioning people or furniture – the 'churn factor'.

Managing this change is the specialist discipline of facilities managers. Of those surveyed, 75% reported that their average churn rate (the percentage of people relocated in one year) has been more than 10% over the last three years. A third of respondents reported churn rates of over 50%. A high proportion of this movement meant desks had to be relocated and, to a smaller extent, a switch from cellular to open-plan and rearrangement of partitions. Many of these companies incurred significant costs in modifying their data, communications, electrical, lighting and hvac services.

Clients and designers agreed that the service sector had the highest churn rates. The main reasons given were reorganisation of team/department structure (96%), space use intensification (63%), expansion (48%) or downsizing (34%).

Plant that's simply too big

To cater for these changes, the CIBSE survey suggests that designers oversize to 15-25% on mechanical services, and to around 25% on electrical services. Such levels of oversizing are confirmed by the other studies^2,3.

However, despite the good intentions of using design safety margins, preliminary results from the latest BSRIA research in the UK suggest that up to 70% of existing services plant may be significantly more oversized than designers are admitting. These results are based on a current BSRIA project that is monitoring the oversizing of plant on a sample of 50 UK sites.

It seems that much of this excessive oversizing arises in the design office. Other causes of the problem have also been attributed to reduced loads that have come from changes in building use and improved energy efficiency. Examples of better energy efficiency that affect oversizing include:

• increased building insulation;
• improved local environmental control;
• improved lighting provision, eg by using high frequency ballasts and controls.

The BSRIA has found that oversizing is most prolific in heating and cooling plant. At the extreme, plant has been identified to be nearly five times too big for current requirements. In such cases, oversizing can significantly increase energy costs and create problems of poor control, occupant discomfort and shortened plant life.

Typically the BSRIA estimates that plant oversizing in the UK accounts for 10-15% of hvac-related energy consumption⁴. For a large office building with an annual energy cost of £250 000 for hvac plant, this translates into a cost for the occupier of up to £37 500 per year.

Looking ahead

The CIBSE survey suggests that clients have difficulty in thinking beyond their present needs for new building developments. 70% of designers said that their clients aren't able to express their requirements for adaptability clearly.

More than four-fifths of them said they have to prompt clients to think about it, ideally early on in a project – during feasibility or outline design. Designers suggested taking a whole life approach to project economics, or considering future scenarios that the design might have to cater for – the more radical the better.

Scenario 1: status quo

Consider an office that will be used continuously with little or no change over time. This is what engineers mostly design for now – specialised buildings are difficult to adapt to changes in use⁵. The office may be flexible enough to cater for small changes in work practice – thanks to the tradition of plant oversizing – but this designed-in flexibility brings penalties in terms of operating problems and greater energy use.

Of course, this scenario is extremely unlikely to occur – even though it is the one designers concentrate upon.

Scenario 2: all change

Consider the same office but, more realistically, one that is subject to constant change with high churn rates. Staff may become annoyed with a lack of privacy, prompting a redesign of the original open-plan layout in favour of small offices and small, open-plan team areas.

A few years later, home-working may be in vogue, so that less office space is required – the redundant space being converted into flats. To allow this kind of change, some designers said they advocate a loose-fit standardised approach, with oversized services located away from the structure.

Others recommend a shell and core approach – stating that it is easier to adapt individual areas in a building. To help evaluate the requirement for adaptability, the designers polled said they would discuss different things. Three-quarters said they would talk about future occupancy patterns. Two-thirds said they'd talk about possible changing work practices, with another two thirds citing future load changes. Slightly fewer (63%) said they'd discuss space layout.

Strategies designers might use included:

• designing on a grid basis with distribution routes that are inexpensive to change;
• using busbars within risers;
• plug-in or local services, like point of use hot water systems;
• structured cabling (and probably wireless technologies in the future) for IT and communications.

A rule of thumb says that a raised floor becomes economical for churn rates above 30%, and typically pays for itself in two complete churns⁶.

Scenario 3: turning up the heat

What would happen if local climates suddenly became much hotter or cooler? Or if the heat gains from office equipment changed sharply? A radical change in load like this will clearly affect the plant's capacity requirements. Sizing with loose-fit safety margins will cope where loads have increased but if loads fall, plant will become even more oversized. In most cases with falling loads will reduce plant efficiency.

A good distribution of services infrastructure, providing some overcapacity where it is expensive to add later, seems to be a popular strategy. But for central and terminal unit plant, the general rule for energy efficient adaptable design seems to be "if in doubt leave it out", providing contingencies to allow future occupiers to adapt the building and add services when necessary.

Alternatively, the design team can use a diverse supply strategy. By, for example, using lake cooling and/or ice storage, or providing for additional terminal units by offering spare distribution connections.

Scenario 4: high maintenance

Higher fossil fuel costs are likely to be a fact of life. But how expensive will they become? If energy costs increased tenfold in real terms, then the cost of plant oversizing would rise proportionately. Higher energy costs would also increase manufacturing costs for plant with a high embodied energy content. So replacing such plant may become more expensive – meaning that oversized plant may have to last more years than expected.

Good operation, control and maintenance will help to minimise energy costs by, for example, ensuring that plant is off when it is not required. Well-engineered modular control of plant, and capacity control of oversized central plant, are popular ways to meet changing loads efficiently. Some designers suggest using plant such as variable speed drives on pumps, inverter-driven fans, modular boilers and even modular plantrooms.

While certain designers make use of fan-coil, displacement ventilation or variable air volume systems for adaptability, others realise the benefits of using natural ventilation. They choose local air conditioning for hot spots and an upgrade path to full or partial air conditioning if it is required in the future.

Finding the more sustainable solution

Most building clients said that building adaptability will continue to be very important (51% said it was of major importance, 35% said it was vital). However, the future requirements of buildings are impossible to control. Clients are realising only too well that their plans do not necessarily happen.

Considering more than one possible scenario early on in the design process may help to accommodate future changes in a cost and time-effective manner. It may also help to avoid providing buildings that lock future occupiers into costly, inflexible situations.

This doesn't mean providing extremely loose-fit buildings that are inappropriate for their intended functions. The key is to balance good adaptability and the cost of plant oversizing with the business objectives of the building owner and occupiers.

Whole life approaches to design can help to compare different design solutions for a particular brief. The British Standard HB 10141⁷, the draft ISO standard on service life planning provides one such method.

The adaptable building services design project will produce a CIBSE-endorsed design guide next year. This is intended to help designers find a more sustainable solution – optimising cost and time and making savings for the owners and occupiers over the life of the building.

The project will also produce a complementary document for clients. This will include contributions to the British Council for Offices Specification of statements on satisfying clients needs in this area.

References and further reading

¹Kincaid D, 'Adaptive reuse of buildings', Facilities Management World, Issue 5, BIFM Aug/Sept 1997.

²Parsloe C, 'Over-engineering in Building services', Technical report TR21/95, BSRIA

³Lawrence-Race G & Parand F, 'Engineering design calculations and the use of margins', Research Report 4, CIBSE, 1998.

⁴Brittain J, 'The cost of oversized plant', Building Services Journal, 4/97.

⁵Leaman A, Bordass B & Cassels S, 'Flexibility and adaptability in buildings: the killer variables', Usable Buildings, London, 1998.

⁶Brand S, How buildings learn, Phoenix, London 1997.

⁷BSI, Buildings, Service life planning: General principles, BS HB 10 141, 1997.