Heat loss calculations

[Jeremy Harris]

I'm sticking this here as I've been asked the question via PM, and rather than just give an answer to one member, I thought it might be more useful to stick the answer somewhere were others can also read it.

 

Back when I was first looking at doing some rough "what if" type comparisons, between different build systems, windows, insulation and airtightness levels etc, I wanted a fairly quick way to be able to change one element, say the wall U value, or the efficiency of the MVHR system, and see what impact it had on the overall heat loss of the house.  This model was never intended as a substitute for something like PHPP, which is very comprehensive, it was just intended to give a rough idea so that I could see the scale of some of the changes, and work out where best to spend our limited budget.

 

Having written the spreadsheet for our build, others expressed interest in using it, so I tidied it up and let others have a copy.  Because lots of people seemed to want to use it, and also because it generally seemed to give results that were within 10% or so of more complex models, like PHPP, I put a copy of the spreadsheet up on our website, as a free download: http://www.mayfly.eu/wp-content/uploads/2017/01/Fabric-and-ventilation-heat-loss-calculator-Master.xls

 

This post is a set of very brief instructions on using this spreadsheet.  First some health warnings.  It was never intended to give an absolutely accurate prediction of heat loss, and as such it takes no account of solar gain, wind or incidental heat gain from occupants and appliances.  As a consequence it is generally a bit pessimistic, in that it will usually tend to slightly overestimate the heating requirement.  This is not necessarily a bad thing, as it can be useful to have a bit of heating capacity in reserve for exceptionally cold weather.

 

To use the spreadsheet, you first need to gather all the data needed to complete the white cells.  Most of this should be self-explanatory from the notes in each section.  The U values, for example, should be the true U value of the component, including any additional thermal paths, so the window U value needs to be the Uw value, not the Ug value, and the floor U value needs to be adjusted for any thermal bridging around the periphery.  All the areas are the internal wall, floor and ceiling/roof areas, not the external ones.  The model does not account for geometric thermal bridging at corners, but in a well-insulated house this effect should be very small, anyway.

 

Some of the most difficult to obtain data can be the mean daily air temperature and the mean minimum daily temperature, for each month.  This data is available for your location on the Met Office website, but posting a link seems a bit fraught, as the Met Office keep changing their website and this makes any link out of date fairly quickly.  All I can suggest is that you work your way through the historic data on the Met Office website and find that closest to where you live.

 

Once all the data is filled into the white cells on the spreadsheet, you should get some numerical data in the green cells, plus two graphs will appear.  The graphical data is often the most useful.

 

First, there is a basic heat loss versus outside air temperature plot (the Heat Loss Vs Delta T plot).  You can use this to determine how much heat the house will need to maintain the room temperature that you put into the spreadsheet (it defaults to 20 deg C, but you can change this to whatever you feel comfortable with).  The red line is the total heat loss, the other lines are there so you can see which elements are contributing the most to the total.  If you want to know how much heat the house will need in order to maintain a temperature difference between inside and outside of 20 deg C (say a 20 deg C room temperature when it's zero deg C outside), then just go up vertically from the 20 deg C point on the horizontal axis until it meets the red line, then go across horizontally from this point to the vertical axis and read off the heating needed in watts.

 

The other plot shows the heat loss per month, and this one can be a bit confusing, because, like the other plot, it takes no account of incidental heat gain, from solar heating, appliances, occupants etc.  The best way to use this is to print it off and pencil a horizontal line across where you think you wouldn't have heating on.  For example, If you turn your heating off in April/May and on again in September/October, then draw lines across at about the point where these dates cross the other lines and call that your "no heating" point.  The mean heating needed for each month will then be the difference between those lines and the values on the plots.  You can quickly work this out by just noting the amount of incidental heat gain, indicated by the pencilled horizontal lines, and then subtracting those values from the monthly values.  Be aware that this is really a very rough estimating tool, as there will be big peaks and troughs in daily temperatures within those months that will effect the heating required.  In most respects, the heat loss vs delta T plot is more useful for sizing a heating system.

 

Hopefully the above should make some sense to anyone trying to use this tool.

[Guest Alphonsox]

An independent assesment ?

I used this tool to model the heat loss performance of our planned build prior to construction - I didn't believe the results and as a result I have an over specified heating system. It turns out to be well within 10% for our particular box like passive spec build when 

- unoccupied 

- no appliances running

- zeroish solar gain (Northern Ireland in spring!)

 

The SAP calculation I ran at the same time gave very similar results when occupancy was accounted for.

 

I'm not sure what other information I would have gained using PHPP or similar.

[AliG]

I used this for my house. I have also used this to roughly calculate the expected cost to heat my swimming pool, so it will be interesting to see how close I am when it is finished.

 

I would note if anyone asks a heating company to specify a system, it will almost certainly be enormous relative to the figures the spreadsheet produces as self built houses tend to be way better insulated than they are used to.

 

 

[jack]

I've said it before, but our plumber was aghast at the idea of a 5kW ASHP for a large house - and that was just for heating! He simply wouldn't believe there was any chance it could handle both DHW and heating.

 

He gave an example of the DHW being on for a couple of hours very early in the morning, and wanted to know how we'd feel about the temperature in the house falling significantly during that period. I pointed out that it would take 24 hours without heat for there to be much difference to the temperature, so a couple of hours was nothing.

 

He still refused to believe me and suggested a 9kW ASHP (he wanted 14kW) as a compromise.

 

We went with the 5kW and it works just fine.

[ProDave]

I have input all the data for my house into Jeremy's spreadsheet, and it tells me when it's -10 outside and 20 degrees inside the total heat demand will be 2554W so I am confident my 5Kw ASHP will do nicely.

 

The house will get tested this winter. We will be partly occupying it but at least to start with the ASHP and UFH won't be installed so just an electric convector or 2. That should enable me to reasonably accurately measure the real heat input needed.

 

I am hoping this winter will prove there is no need for a WBS (we will be having one for the static caravan though)

[TerryE]

It depends on your house's construction technology, but in our case with 300mm cellulosic filler and an external stone skin the decrement delay factor is so long (multiple days) that you can ignore diurnal variations in any calculations.  IMO, Jeremy's level of detail is fine, though I use my own variant in my spreadsheet.

 

You also need to understand that there are two separate measures that you need to consider: (i) a calculation based on your local area seasonal temperature averages to give you your expected heating costs, and (ii) the worst case heating demand if you are in the middle of a stuck jetstream cold-snap which lasts for days.  IMO, this second case is so rare that I am quite happy to use a small electrical heater for a couple of days, once every few years if needed to keep the temperature up.

 

IMO, the biggest risk is not in the heating but in some construction flaw which leads to thermal bridging that you have missed in your calculations, and again in my view the answer this is down to you.  The devil is in the detail and you can't just assume that any construction crew will always get this right: trust but verify.

[Stones]

I built upon Jeremy's spreadsheet to include solar gain (both average and peak daily), incidental & household gains, and the impact of wind, as our location is very exposed as wind driven heat loss is probably our biggest variable.  Accurately modelling solar gain was very helpful in being able to determine whether we would suffer from an overheating and how this could be managed.  Well worth spending the time doing this.