A prototype fan coil unit has achieved such amazing energy savings that its developers fear for their credibility

We’re worried engineers won’t believe our data,” frets Terry Farthing, technical consultant at Trox Design Bureau. He has a point. Tests of a prototype variable air volume (VAV) fan coil unit in Trox’s Thetford laboratory showed the unit has the potential to cut a fan coil installation’s fan energy consumption by an astonishing 90%.

But the 90% figure is only part of Farthing’s credibility concerns. Once the laboratory results were plugged into a computer simulation programme and translated into an energy performance certificate rating, the modelling revealed energy savings so significant they have the potential to push up a building’s EPC rating. What’s more, computer modelling showed that if a cautious designer oversized an installation’s VAV fan coil units, the savings would be even greater.

The energy savings were far larger than Trox initially expected when it started development work on a VAV fan coil unit. “It was the magnitude of the savings in relation to the total building energy that surprised us,” says Farthing. To realise why the savings are so significant it is necessary to understand the difference in operation between a VAV unit and a traditional constant air volume fan coil unit.

Fan coil units are essentially mini air-handling units containing a filter, a fan and a heat exchange coil (or pair of coils, one for heating, one for cooling) connected to a building’s heating and chilled water circuits. Each unit serves principally to recirculate the room air; outside air to meet occupants’ needs is provided independently.

On a traditional fan coil unit, with an AC fan motor, the fan can operate in two modes: off (at night or weekends) or at a set speed. With the fan running continuously, the room temperature is maintained by increasing the flow of heated or chilled water through the unit’s coil by opening or closing the control valve.

VAV fan coil units operate differently. Rather than use an AC motor, the fan is driven from an electronically commutated (EC) motor so its speed can be varied simply by adjusting the voltage. This is the key to the energy savings, because the power required by the fan changes as a cube of the flow induced by it. In other words, if the fan speed is reduced to 60%, the power required by the fan is reduced by the same factor cubed, ie 0.6 x 0.6 x 0.6 = 0.2. The exact figure will depend on the type of fan, but at 60% speed a fan will use about 80% less power (see Figure 1). Reducing the speed will also reduce the unit’s fan noise.

Trox knew the theory, but before it launched into manufacture of the VAV fan coil units it needed proof of how they would work in practice and likely energy savings. In July 2008 a team of experts was assembled to enable a valid comparison to be drawn up. Trox pulled in Stuart East, MD of consultants John Noad (Building Environment) and, for a developer’s input, Neil Pennell, head of sustainability and engineering at Land Securities. It also brought in Alan Jones, MD of building simulation specialist Engineering Design Solutions Ltd (EDSL), supplier of the TAS Building Designer simulation software.

Using TAS, Jones built a virtual 30m x 30m, open-plan, air-conditioned office to work out the impact on the total building services carbon emissions. The virtual building was designed to comply with the Building Regulations Part L2 and its perimeter and core areas were divided into zones according to the National Calculation Method (NCM). The model used typical figures for internal heat gains and weather data based on a typical representational year (TRY) for London weather data. The fan coils were sized to provide 10 air changes per hour and their operation was based on 7am to 7pm occupancy.

Trox then tested the performance of the prototype units at its Thetford factory. Because the TAS simulation package is based on a component by component hourly plant performance simulation, Jones was able to input the data from the lab tests directly into the modelling package. “We used the actual data to create a performance algorithm for VAV fan coils,” he explains. He admits these algorithms are at an early stage of development, but claims they are of sufficient detail to allow the potential benefits of VAV fan coils to be assessed.

For the VAV units Jones used a specific fan power (SPF) figure of 0.25, which was based on the Trox test data, and a maximum fan “turndown” to 60%. For the comparison Jones used an SPF of 0.8W/l/s, taken from the NCM, for constant air volume fan coils.

Computer modelling showed that if a cautious designer oversized an installation’s VAV fan coil units, the savings would be greater than 90%

Thermostatic controls on the heating and cooling system ensured the heating was designed to operate between 18C and 20C and the cooling between 24C and 22C with a 2C dead band between.

The simulation provided hourly profiles of the heating and cooling. It showed that because of internal heat gains, cooling was needed throughout the year. A small amount of heat was used primarily to preheat the offices in winter before occupancy. To achieve maximum energy savings the controls on a VAV fan coil are set up so the fan runs on minimum speed for as long as possible, with variations in the heating or cooling load accommodated by adjusting the hot/chilled water flow rate through the heat exchange coil (Figure 2).

Farthing says it is normal to vary the fan speed between 100% and 60%, to maintain air circulation. “Below 60%, there is the possibility of cold air being dumped on occupants unless an anti-dumping device is fitted to the diffuser,” he says.

The simulation showed that for about two-thirds of the year, the VAV fans would run at 60% speed. Buildings often contain scores of fan coils, so even a modest energy saving will be replicated many times.

The modelling showed that for this particular installation a VAV fan coil unit produced considerable energy savings: the fan on the constant air volume fan coil with its AC motor consumed 13800kWh, significantly more than the 1405kWh annual fan energy use for the VAV fan coil unit with its EC motor. In other words the model showed that by simply varying the fan speed it was possible to save 90% of the fan energy in a fan coil unit. Figure 3 shows how the two installations compare.

The advantage of having the VAV’s performance data in the simulation programme meant Jones could run EPC software. The reduction in fan energy for the VAV fan coils pushed up the building’s EPC rating from a C to a B.

The savings assumed the FCUs had been sized correctly. Jones then re-ran the calculations on the assumption that the engineers had been cautious in their choice of units. The results were even more impressive. The simulation showed that if the VAV fan coils were oversized by 40%, the unit’s fan would run at the minimum speed of 60% all year, giving an energy consumption of 1330kWh, which is slightly less energy than the correctly sized units. The fact that oversized FCUs do not consume any more energy than correctly sized ones “should offer a huge amount of comfort to the engineer,” says Jones.

Land Securities’ Neil Pennell sounds a note of caution. He says on many of Land Sec’s developments, the fan coil heating circuit is designed to run at a lower temperature than the TAS model, so the VAV’s fan speed may have to be increased in winter to meet the room’s heat demand.

Jones decided to check the VAV’s performance on a real scheme. EDSL had recently modelled a new office building in Leeds which used fixed speed fan coil units. He ran the model again but this time using VAV units in place of the constant volume fan coil units. “The proportion of energy saved was almost identical to that for the notional building we’d modelled previously,” he says.

The EC motors cost more than the equivalent fixed speed model, but Farthing expects payback to be less than three years. However, cost is not the real challenge. At the moment the Building Regulations Simplified Building Energy Model fails to take account of potential energy savings from varying a unit’s fan speed. So the greatest difficulty for specifiers who do not run TAS software, for now at least, will be quantifying those savings.