I was hoping I could get some feedback on my first attempt at using OpenBEM, which is apparently based on SAP 2012.

I have written this up in full on my blog: http://www.elstensoftware.com/blog/2015/03/28/first-openbem-assessment/ (sorry this post is HTML so no linky linky)

Summary:

• SAP Rating: 68.5


• Annual Energy cost: £1589


• Floor 85.3 W/K


• Roof 27.9 W/K


• Walls 99.8 W/K


• Windows 168.5 W/K


• Ventilation 178.5 W/K

A few thoughts:

The ventilation figure is assumed given the data I entered. Until I get an air tightness test it can't be accurate, but I would not be surprised if it *is* the largest factor.

It surprising how much the windows appear to be losing. Conventional wisdom amongst many appears to be replacing windows does not derive large savings. I suppose it depends on a case by case basis.

Thermal bridges are ignored in SAP, right? So things I have such as the following will not be considered:

• Some (not many, thankfully) plasterboard walls (I felt a draught through a plug socket yesterday!)


• Unfilled external sides of cavity walls being brought into the thermal envelope following extension, the whole cavity then acts as a 'chimney'' into the loft (I've done a thermal imaging camera survey that showed this. Another blog post, when I get around to it)


• External walls bridging the thermal envelope into ground floor extensions

Do I need to worry about these types of things, considering I am not aiming at Passivhaus levels (we have neither the time nor the money to achieve this)? Let's say I have my eye on AECB Silver which would mean cutting our heat demand by about 60% per m². The trouble is that because they aren't considered in the model, I can't do a "what if" scenario to see if they make much difference.

What does SAP occupancy mean? The number of 3.1 seems low for a house of this size if it means the average number of people living in the house.

Is a fan-assisted flue (as we have on our oil boiler) the same as an open flue?

I have attached all the input data (it's JSON encoded but I had to name it txt to get it attached), all opinions appreciated.

Thanks Tony.

Posted By: tonyHall u value is definitely wrong I only dream of 0.05

Agree, thought that was wrong, but that's what the formula gives for zero external walls.

Posted By: tonyAir infiltration is different from ventilation, I may start a thread on this.

Are you referring to the ventilation losses?

Posted By: tonyGenerally the U values in the existing house seem high.

Do you mean I have entered higher than you'd expect or they are higher than you'd want?

Posted By: tonySeems overly complicated doing a room at a time, could try house plus extension? Should be able to model what ifs by running same thing with the relavent changes added.

There are actually five extensions!

I did it took by room because that's how I happened to have the measurements already.

Thanks Tony.

Posted By: tony1) even with a solid ground bearing floor it would be higher, if suspended floor then a great deal higher.

It's solid uninsulated. Clearly I trusted the formula too much. Maybe I'll edge that up a bit higher... suggestions? No external walls remember.

Posted By: tony2) no ventilation losses are necessary though the heat can be recovered, air infiltration losses are voluntary, uncontrollable and the heat cannot be recovered.

I see... well I guess this might be an area where the model might not factor in the difference. Do you mean "voluntary" as in it's my choice to leave the gaps, or did you mean "involuntary" in the sense that when the losses happen it's not because of a choice at the time of loss?

I still think working on these types of losses is a good idea. So many things still to do... loft hatch, downlighters, cracks between frames and walls, plasterboard tops and bottoms, joist ends if I ever get brave enough...

Posted By: tony
3) higher than I would like them to be, with so many extensions at least parts of some of the old external walls must be covered partially.

Agree but the U=0.5 walls are the best you will get them without I/EWI (filled 1950s cavity) and the U=2 wall was dealt with here: http://www.greenbuildingforum.co.uk/newforum/comments.php?DiscussionID=12874

Even the assumed better U values on the more recent extension is probably partly negated by the plasterboard used to line it. Thankfully almost all of the rest of the house is wet plastered.

Window U values were assumed to be U=3, 6mm air fill old fashioned 2G. Uninsulated frames obvs. Black mould forming on them this winter.

Comment left on your blog re graphics there not working.

Re the insulation: saying it has zero perimeter doesn't make sense unless your house is infinitely large. How about taking the whole area (all the rooms) with this floor and treating them as a single unit to get an average U-value across the lot then applying that to all of them?

But the rooms in your non-infinite house do have an external perimeter - it's just that they're less directly attached to it than that simple formula assumes. The “right” way to do it would be to have a 3D heat flow model of what's going on in the ground which would be a right pain to set up (and require some doubtful assumptions) so the question is how to get a better approximation with the formula.

Just setting the perimeter to zero is obviously wrong. If you just had a one-room house you could separate it into the outer 10 mm round the wall with a lot of perimeter and therefore a high U-value and the rest with “no” perimeter and therefore a low U-value but that'd be silly.

If you've got an area of poor insulation like an old inner room surrounded by extensions with somewhat better floors then maybe a better approximation would be to assume that the inner bit is sufficiently protected by the extensions as to have the same U-value as they do.

gravelld,

SAP does NOT ignore Thermal Bridges/Junctions. Far from it, they can play a large part in your SAP rating calcs.
You have choices, in SAP: y=0.15 (No calcs for thermal bridge/junction details), adopt ACDs, or calculate the specific Htb from known Psi-values.

Good luck

Ah! I found out where I came to those floor U values... http://www.samsung-heatpumps.co.uk/documents/U-Values_from_Domestic_Heating_Design_Guide_v1.0.pdf (5.8).

Note "for irregular shaped areas". I think I extrapolated that, probably incorrectly, to think this was a general formula because when I ran through the values in the tables later on in the doc they seemed to obey the formula as well. However, looking again that isn't the case.

I have altered the U for all the rooms according to the tables in the above document. Some have gone up, some down. The overall effect is to actually reduce the heat loss through the floors.

I have also arranged to visit the local council to get the microfiche out to find out the dates of construction for each extension, to get more accurate U values.

So I am minded in my strategy to, short term, continue fixing air leakage.

Long term, replace the windows (our render is looking a bit shabby so it would be fantastic to do this at the same time as EWI but... money).

If we did that, plus got air permeability down to 3 m³/h/m² that would put us about 55kWh/m², so getting close to AECB Silver for the space heating.

What I don't quite understand is that the ventilation losses appear to be very difficult to get under 100 W/K... even specifying very air tight with an air infiltration test. Must be something I'm missing here.