I recently had an airtightness test carried out on my house and I think the results might be interesting for others. My reasoning for commissioning the test is that I want to install an MVHR but I also believe my 9kW Daikin is significantly oversized - no need to go into detail on that here because it is well covered in other threads. I had a local company in Kent do it for the very reasonable price of £265. This website lists registered providers: https://www.bcta.group/…/air-tightness…/united-kingdom/
The result was 4.11 ACH50. I understand that ACH50 is not the ACH used for heatloss surveys. Heatloss surveys, I believe, use ACHnat. ACH50 has a pressure set by the door blower fan of 50 Pascals whereas ACHnat, I believe is at approximately 4 Pascals. ChatGPT tells me that 4.11 ACH50 is 0.27 ACHnat.
Post 2006 new builds need to have an ACH50 up to 8 Pa max. The tester told me that when they are involved in the design from the start they aim for 5 Pa max. Out of interest a passive house needs an ACH50 of 1 Pa or less.
On the face of it it seems surprising that a 1988 126.92m2 4 bed detached house would be more airtight than a new build but there are a number of factors that explain why. The only chimney was removed down to the first floor ceiling and sealed. It has cavity wall insulation from new, 300mm insulation in loft, no extractor fans, no trickle vents bar one in a sliding door. Other than adding draft excluders to a couple of doors and replacing window seals the only other upgrades are a sealed loft hatch and a composite rear door. I have attempted to locate and seal up any gaps where services enter, although I missed two. I removed redundant overflow pipes and sealed up the holes.
My heatloss survey for the heat pump came in at 7.59kW. Using ‘default’ MCS air exchange rates for different periods of property on heatpunk I get.
Pre 2000 - 8,452W
Post 2000 - 7,069W
Post 2006 - 5,750W (My ACH is even lower than this!)
I have an ESPAltherma feeding emoncms and that suggests my actual heatloss is about 5.5kW (taking the top line not the middle). This is also consistent with my prior gas usage.
This suggests that the lack of air tightness testing for heatpump heatloss surveys is quite a blind spot.
I’ve just put the ACHnat estimate of 0.27 into heatpunk and it is saying the heat loss is 4767 W. Almost bang on the HeatpumpMonitor.org estimation of 4620.
and have been trying to raise awareness of this issue. There is an interesting 2nd component to this and that is how much margin is wise to add on top of an accurate heat loss for heat pump sizing see:
Thank you for posting those specific links. I have read quite a bit on here but I am not sure I have read all of those pages before - I’ve skimmed through but I’ll have to study them carefully. The link to a cheaper download of EN 12831-1:2017 is very helpful. I gave up looking for a free or cheap download so that is great! I’ve been working on the assumption that the ‘Integrated’ values on the Daikin Spec for the 8kW unit take into consideration DHW and defrost cycles, do you have any insight on that?
That would suggest a reliable average output of ~7.3kW at -3.2
I’ve also been wondering about wind. My current thought is that the main regression line on the heat demand tool is for average wind speed and if you fit the the upper and lower lines to the outliers then the top line, where it intersects 23.2 degrees, would give a good indication of a heatloss on a very windy day at design temperature (of course there are other factors for day to day variance but I think it might give an indication). Or does the heatpunk calculation factor that in by scaling up the provided ACHnat?
I think in my situation the 6kW Daikin is probably big enough, but given the 8kW is the same hardware it would be a safer choice and would more than cover a sensible oversizing factor.
I had the same air tightness test and as a result Octopus Energy agreed the proposed 8kW Daikin was oversized and installed a 6kW. Data proves the 6kW is a good fit.
That is encouraging that an Air Tightness test persuaded them to downsize the unit - albeit at design stage. I have asked a number of people a number of times if they can take into consideration air tightness but they say their model does not allow it to be changed.
@matban Matt this post was really helpful for me. I’ve been thoroughly puzzled by my seemingly excessive ACH8.9. I hadn’t given sufficient notice to the fact that it was at 50pa. So you have helped enormously. I have been having ongoing chats with chat gpt too, so I was able to go back and update the input and solve the mystery that had appeared to be a meaningless number.
Not taking measured ACH results into account is just plain poor practice. We had a measured result and an MVHR system and no installer doing a survey used this information on first pass but 4/5 were bullied into using it ‘after getting help’ from the boss/ software supplier etc. on how to do it.
