What can go wrong with MVHR

I live in a Victorian detached house which has been extensively renovated throughout; since most of the rooms are now fairly well airtight we chose to have mechanical ventilation and heat recovery installed (MVHR).  It has proved itself well. It is almost silent and it provides the ventilation we need without generating draughts. Air is extracted from wet rooms and used to heat up fresh air pumped into the other rooms, so the fresh air isn't so cold. Also the fresh air comes in at ceiling height rather than foot or neck height and mixes in with the rest of the room so we don't notice it. That is how it is supposed to work, and it does for us, but I recently read an assessment on the functioning of MVHR in 10 new homes near Slough where it didn't work well at all [1]. As building regulations for new houses tighten up it is very likely many more homes are going to be built with MVHR. It would be a great shame if they didn't work properly.

The 10 homes in the assessment were completed in 2010 and monitored during 2010-2012. The performance proved so poor that the MVHR systems were recommissioned and some components replaced. Here are some of the problems revealed:

• The same size unit was installed in all 10 houses, from 1 bedroom flats to 3 bed semi-detached. The unit installed was half the size of the one originally specified and it was under-powered for the larger homes.
• MVHR units can be noisy because of vibration from the fan. They should have rubber feet and the duct connections should have a short run of flexible tubing to prevent the noise being transmitted. In this installation they had the rubber feet and in some cases they had flexible tubes - but quite often not.
• MVHR should have different types of valve for extract and supply, because the supply valves are designed to push the air out sideways rather than straight down, to aid mixing. The extract ones just suck in and, in reverse, will jet straight out - when the vents are in the ceiling they will blast straight down. Unfortunately these homes had extract valves fitted everywhere even in the supply pipes. To reduce the resulting draughts the valves were closed off to let just a trickle through. This meant the fans had to work extra hard and monitoring found that in several of the houses the system was running on boost pretty much all the time - adding to the noise and also to the energy consumption.
• Why were the draughts from the supply vents so cold? Isn't that the point of the heat exchangers, to warm the air up? Of course the system can never be perfect but for these houses it was not nearly as good as it should have been. The MVHR units were fitted into the cold loft space and there were long lengths of ducting also in that space. Even though most of the ducts had some insulation (the flexible sections mostly did not) they were still leaking considerable heat into the loft. So the air extracted from the wet rooms was cooling down quite a bit before it even got to the heat exchanger and the supply air was cooling down further before it got into the rooms.
• Also there should have been condensation traps in the ducting to outside, to prevent moisture running back into the MVHR unit - there were not.
• While the ducting was being installed there was a lot of fitting out work going on generating dust which found its way into the pipework. When the fans were switched on this dust was swept into the filters clogging them. The filters should have been changed or at least cleaned before people moved in but they were not. Also some of them were installed the wrong way round.
• Also these units had a fly screen to stop crud coming into the system from outside. Over time they block up and need cleaning - but they were in a place that was difficult to reach.
• Finally, the original specification was for CO2 and humidity sensors to be connected up to control the MVHR unit so it could run gently most of the time and boost when necessary. These were supplied but not connected up.

Although the occupants did not notice a problem with air quality in their homes they complained that the system was noisy. Also the units were drawing at least 30% more power than expected, in some cases nearly twice as much power.

The residents in this case were lucky. The study found so many problems that it was decided to recommission their installations and correct some of the mistakes. They got the right sort of valves fitted and adjusted so that the flow was less restricted. Also the filters were repositioned to make them easier to change (previously this required removing 10 screws). After this the systems were much less noisy and used less power.

How did this appalling installation happen? Clearly there were failures at a number of stages.

• Firstly, the initial specification should have selected appropriate units for the different size houses. Also, given that the MVHR had to go in the loft it would have been better to run it as a warm loft with insulation in the rafters (as we do).
• MVHR needs regulate basic maintenance like clearing filters. The designers should have ensured easy access. 
• There was never a detailed plan of wiring or ducting produced. The equipment was ordered on a supply-only basis, not taking advantage of the suppliers design service. The plumbers and electricians installing the system did not have any specific training. How should they know that you need different sorts of valves for extract and supply?
• Since no-one designed wiring for the CO2 and humidity sensors the electricians did not install them.
• And so on.

Fundamentally, the problem was that there was no-one on site with knowledge of this type of system who could have guided the fitters and flagged up mistakes.

The worst thing about this story, though, is that if the researchers had not happened to look at these homes the mistakes would never have been rectified. The occupants would tell their friends that MVHR was horribly noisy and, if they bothered to check their bills, might have decided it was not nearly so efficient as it is supposed to be. At some point the MVHR units would fail because of condensation in the system and then they would say MVHR was unreliable as well. Also, the builders would go on installing other MVHR systems in the same way and never learn anything was wrong. In fact they might have done several previously we don't know about.

MVHR isn't new technology but it is fairly new in this country. Anyone intending to install it needs to make sure that the system is designed by an expert and the installers are properly supervised, even though this may cost a little extra. The same is true for other new technologies such as heat pumps.

If this story is representative, it is no wonder that new technologies tend to struggle with a poor reputation in the early years. Of course this doesn't mean that all new technology works perfectly from day one. For example early condensing boilers were renowned for icing up in winter. Solutions have now been found such as siphon traps which stop the condensate dribbling. Condensing boilers are widespread now mainly because government regulations forbid less efficient systems being installed, so customers don't have any choice. MVHR isn't mandatory, and probably never will be but in my experience it is a perfectly satisfactory option, when configured and installed sensibly.

[1] Assessment of MVHR systems and air quality in zero carbon homes NHBC Foundation (August 2013)