29 Jun 2010

Into the Metrics

One of the things which PassivHaus has brought us is a new way of measuring energy efficiency. It's being called a performance metric and it's expressed in kilowatt-hours per square metre per annum. Or kWh/m2/a. It's a much clearer way of presenting the issues at hand but it's not without its problems - of which more in a minute.

The subject is dealt with at some length in the document produced by the Zero Carbon Hub late last year. It looked at using another metric, kg/co2/m2/a, but rejected it because of the confusion relating to the different carbon factors in various fuels - i.e. if you selected to heat with biomass, you could virtually ignore energy efficiency. It also looked at using the so-called Heat Loss Parameter which is a sort of amalgamation of all the U values surrounding the building, but felt it was too little understood and no one outside the UK has ever heard of it. And it alighted on kWh/m2/a because it was the one pushed by the PassivHaus standard and it has some sort of meaning for ordinary mortals. If you can just figure out how many bloody square metres there are that you are trying to heat, then it gives a good indication of what the fuel usage will be.

Now the Zero Carbon Hub has come in for a fair bit of stick from darker greens than me for setting a target which doesn't look so very hard to beat, compared with the Passivhaus standard. These comparable metrics look like this:

• PassivHaus standard: no more than 15 kWh/m2/a
• Zero Carbon Hub suggestion: no more than 39 kWh/m2/a for flats and mid terraces, no more than 46 kWh/m2/a for detached houses, semis and end terraces.

By way of comparison, the last time I carried out this exercise was in producing the 2008 edition of the Housebuilders Bible, when I put my benchmark house through its paces. I configured the house in different times and the performance metrics looked like this:

• 1975 (pre-insulation) standard: 270 kWh/m2/a
• 2006 (as built) standard: 60 kWh/m2/a

Interesting that the 2006 standard really isn't that far short of what the Zero Carbon Hub are suggesting for 2016. Which, I guess, is why they have come in for so much flack. But, to their credit, they have chosen a metric which makes it much easier to compare one standard with another. Would that Part L would do this as well, instead of creating maximum confusion with its over-complex DER and TER calculations.

And what of the problems? Well none of these metrics is perfect and a metric based on kWh/m2/a can be abused. You might think that a kWh is a kWh, but are we talking about heating demand or consumption? The definition clearly states that it is demand that is being measured, so the efficiency of the delivery method shouldn't be taken into account. Nevertheless, expect one of two heat pump suppliers to misuse the metric to suggest that they can heat draughty old piles for much less than the Zero Carbon standard.

Another old chestnut is playing around with the footprint that you are measuring. We still don't have a standardised way of measuring floor areas, so there are advantages in exaggerating the floor area. This factor was graphically illustrated on the PassivHaus study tour to Hanover that I went on with the BRE in 2007, where Carsten Grobe, the architect/selfbuilder we visited, admitted that he had chosen to include his basement within the heated envelope because it made the floor area grow by 30%, and therefore made the task of meeting the 15kWh/m2/a metric that much easier.

There will be confusion, also, about what's included in the kWh measurements. Does it include provision for hot water? No. It's too dependent on other factors and has nothing to do with the built fabric. Does it include anything for cooling? Sort of. As far as I can see, it only takes the cooling effects of thermal mass and natural ventilation into account, and assumes that no active cooling gets used. Lighting? No. Ventilation? No. That's because they are not insisting on mechanical ventilation for the 2016 standard. Really, it's just space heating.

Finally, there may be a danger that consumers will take the metric as a guarantee, and we may find that if they use more energy than this, they start complaining. Heaven help us! The idea of the metric is that this amount of energy will be capable of keeping the building warm and comfortable throughout the year, but the actual performance is still very much dependent on how the house is lived in, and experience of Passive Houses to date reveals that the best performing homes use about a quarter of the energy of the worst, and that the metric score lies somewhere in the middle. It can in reality be very hard to distinguish space heating demand from hot water, and so it's far from straightforward to tell just how successful a scheme has been just by analysing fuel bills.

5 comments:

  1. The other issue with kWh/m2/yr is that it completely ignores the size of the house and number of occupants. I would imagine from the case studies I've seen that the typical passive house is much larger than a typical developer-built house. This means the emissions of the passive house could well be higher than a house built to meet the Zero Carbon Hub standard.

    Also, as most large houses have lower occupancy per m2 they could easily have higher emissions per person. In my view some sort of measure in terms of kWh/occupant/annum with an assumed level of occupancy would be a truer measure of the efficiency of a house. There is an argument to be had over the way of estimating the number of occupants. At the moment there are at least two methods I know of and probably more that I don't. There's the complicated method in SAP (to calculate N) or the simpler number of bedrooms plus 1 used in the Code for Sustainable Homes rainwater harvesting calculations.

    There is of course the danger of the perfect becoming the enemy of the good but on this issue it seems worth getting it right.

    ReplyDelete
  2. Entirely endorse the suggestion that occupancy be taken into account - (I note that this is at least a field in the new AECB/TSB low carbon buildings database, if not part of the calculations).

    I want to see what standard is realistic in modest homes, then I think the rate should be ramped up as the home gets bigger for the same number of people - and/or a factor included for the embodied energy, which might offer a similar incentive towards modesty. I realise this would be more complicated, but it is absolute energy we have in our sights, in the end, isn't it?

    Don't really see why detached houses should be let off the hook, either. Its an inefficient form and customers should know it!

    Yours in the unfashionably red corner!

    Kate

    ReplyDelete
  3. I though rooms are cubes and therefore volume is the measurement to use not area of floor space. Or are they just assuming a standard height? In which case what does double height hallways do to the calculations?

    ReplyDelete
  4. You make another good point. There have always been grey areas involved in measuring floor areas, and the double height gallery effect is one. The modelling for these metrics is all carried out on social housing or developer house types, and often anything a little unusual doesn't fit at all well.

    A better metric would be kWh/m3/a, wouldn't it? But PassivHaus as an idea now has such traction that it's imperfect metric is being adopted because it's easily recognised and understood.

    ReplyDelete
  5. Mark BennettJuly 01, 2010

    In PassivHaus (if I recall correctly) a double height gallery would only have the floor area counted once. This has the effect of increasing the wall/window areas and therefore heat losses but without being able to claim any floor area for it.

    Consequently, the losses need to be compensated elsewhere, so it does "punish" this kind of feature.

    In fact, PassivHaus is actually quite strict about counting Treated Floor Area to make sure only usable space is taken into account. Of course, there are still ways to manipulate this - heated basement storage etc. but no system is going to be perfect.

    I don't know how this is handled in SAP or any other modeling tool.

    ReplyDelete