Heating design, calcs and process - ASHP/UFH

[SuperJohnG]

I've read various threads and views and back and forth and done some calculations based on information I have found here and of course from the very helpful spreadsheets provided by our friend @Jeremy Harris.  I had pretty much decided ages ago I would get an 8.5kW Ecodan, which might go up to an 11.2kW ecodan unit.  However I have not really advanced it beyond basic assumptions, what others done and some quick back of the fag packet calcs. I'm about to start foundations shortly hence, need to firm up on the UFH design for the in slab pipework and would appreciate some sense checking/ a wee nod that it seems ok. 

 

Build is SIPS with Kore insulated foundation, aiming for high levels of airtightness but not near PH (basically as best as I can get it without major hassle), Vent Axia Sentinel MVHR.  ASHP which does UFH and DHW (UVC 3-400L tank). Secondary heat source of Log burner (I'm aware it's overkill likely - but I have 2 acres of woods, vaulted ceiling and need a backup in the event power ever went out).

 

I would like to map out the steps in layman's terms to ensure they are right and I have followed it correctly from what I have read, then hopefully the thread can then be followed by others. 

 

Step 1 - Heat loss calculation: 

 

So I completed the heat loss calculator - see attached. I have a few checks to do on areas and might make minor adjustments but near as dammit it's close enough. I have added the OAT values for my local area (West Scotland, south of Glasgow) from the MET office.  Now do I basically just use the 'Total daily heat loss power for average OAT (W)' value to find the maximum based on historical data and that defines the minimum amount of heating system required such below: which seems 3.6kW heat loss for January.

 

The simply - I select a heating system which can supply this amount of heat as a minimum (ignore losses for now). So just simply selecting an 8.5kW unit would do the job easy? 

 

Step 2 - UFH design

 

So Having found out the maximum heat loss above (3602W) then we can just plug this into the Heat loss and UFH calculator, which I have also attached. And provides the following output.

 

 

So essentially, going on bare minimum I would need an UFH system which provides 27.7W/m^2 , based on 130m^2 of floor - so essentially if I don't use some slab area then I need to compensate and adjust the calc. 

 

Step 3 - UFH loop spacing

 

So I know the total heat loss, I know how much heat input I have to supply per m^2 if I cover the whole area of 130m^2. So I get loopcad, then draw the circuits. Aiming for counter flow circuits as they provide the best option for even heat distribution, keeping circuits less than 100m in length. Now this is where I get a little lost, I have a figure of  27.7W/m^2 for the whole area to get my required heat input, however how do I correlate that with loopcad and also deciding what my spacing should be?! I have seen @PeterW mention quite a few times about the the spacing had to be adjusted but I am struggling to find the route here to confirm easily what it should be.

 

 

 

My loopcad drawing (I've attached my current draft) showing for example 105W/m^2 in my living room, but I'm not sure how this should read in relation to my calculated figure. I'm pretty sure I'll need 200mm centres, but I'd like to ensure that was right. I've some adjustments to make to my circuits, but it's not far away. I was finding that the auto generate function for pipework isn't ideal but it works ok. The garage can be ignored -  I ma installing pipework in there pre-empting possible future conversion of that area and hence pipe is cheap, so why not add it now. 

 

The manifold will go below the stairs - seems to make sense to me, I have a plant room directly adjacent and was going to put it in there but seemed better to get it out of there as I will have lots of ducts and this would clutter that up. Is it ok to run underneath the stairs? I assume I just have to tell the Joiners not to fit the stairs to the floor with big screws?! 

 

 

Step 4 - Zones

 

I am only having downstairs heating, so it is a single manifold. 

 

I would prefer as few zones as possible, but a single zone wont work. I was thinking I can have two /possibly three. 

 

Zone 1 - Main family room area

Zone 2 - Everywhere else excluding shower room and garage

Zone 3 - Shower room

Zone 4 - Garage (permanently isolated at the moment. 

 

I assume you can have multiple zones on a manifold, and can just split them up by using salus actuators?. So a single thermostat for family room can control three loops (each with an acutator) in that room? and so forth?.

 

 

It's a long post , and actually I wasn't going to talk about zones but thought it was worth adding in. 

 

Thanks. 

Heat loss calculator - SuperJohnG.xls.xlsx Floor heat loss and UFH calculator - SuperJg.xls.xlsx Looopcad - SuperJohnG.PDF

Edited November 9, 2020 by SuperJohnG
forgot tags

[SuperJohnG]

@LA3222 this should make more sense now I've added the rest. 

 

??

 

@LA3222 said

Quote

'For what it is worth, I suspect we are thinking along similar lines. I used the spreadsheet you referred to and I too have settled for an 8.5kW Ecodan. My house is approx 280m2 and the calcs suggest I could prob plump for 5kW but as many others on here mention, I'd rather upsize slightly and run it easier than have a smaller one struggle. I'm planning on using the O/P from that to heat an UVC and also a buffer tank for the UFH. I've started puzzling through this but am not at the finish line yet. Broad brush, it all makes sense to me, I'd like to grasp the finer details though in terms of components used and how they all go together.'

 

Edited November 9, 2020 by SuperJohnG

[SteamyTea]

7 hours ago, SuperJohnG said:

Now do I basically just use the 'Total daily heat loss power for average OAT (W)' value to find the maximum based on historical data and that defines the minimum amount of heating system required such below: which seems 3.6kW heat loss for January.

No, you need to use the minimum OAT that the Met Office supply, and preferably the daily data not the monthly data.

https://catalogue.ceda.ac.uk/uuid/b37382e8c1e74b849831a5fa13afdcae

I think soil temperatures are buried in there as well.

 

 

To give you an example, my Feb (coldest month) temperatures have a mean of 4.3°C, but a minimum of -0.8°C (the whole month, in 1986)

7 hours ago, SuperJohnG said:

So I know the total heat loss, I know how much heat input I have to supply per m^2 if I cover the whole area of 130m^2

Is this just the footprint of your house, or the total floor area, or even the total liveable area?

Just thinking that @joe90's place is about 200m² and his HP is 4 kW, and that does his DHW as well (I know that we are in the very mild SW, but you are not so different being on the west side of Scotland (Isle of Sky climate is very similar to Cornwall, within a degree).

Edited November 10, 2020 by SteamyTea

[LA3222]

Have you considered giving the UFH design over to WundaTrade (as an example)? They will then draw it all up for you and give you a schedule of parts etc. 

 

On another note, (caveat, I am not an expert) I think with a well insulated house and the UFH pipes in the slab means that you don't need to get hung up on w/m2 O/P. I had a quick google last night and saw that max O/P in a screed is 100w/m2 which is well OTT. Having had a quick look at my UFH pipe layout Wunda designed it for 200mm spacings and a 70w/m2 O/P. Again this is well OTT and using the slab as a massive heat store means that the UFH will be run at a much lower temperature than the industry norm.

 

The O/P of the UFH doesn't matter as such because that doesn't heat the house directly, it is the O/P of the slab which matters as that provides a continuous source of heat, the job of the UFH is to dump sufficient heat into the slab.

[dpmiller]

IIRC when I was doing the calcs for ours, LoopCAD always came out very pessimistic compared to jeremy's calculator.

[SuperJohnG]

5 hours ago, SteamyTea said:

No, you need to use the minimum OAT that the Met Office supply, and preferably the daily data not the monthly data.

https://catalogue.ceda.ac.uk/uuid/b37382e8c1e74b849831a5fa13afdcae

I think soil temperatures are buried in there as well.

I maybe haven't worded that correctly - I have used the minimum OAT and added this to the spreadsheet. I am using the line for total daily heat loss to get the output. I tried that link, but seems extremely difficult to find the data. But essentially we are saying the same thing I believe. 

 

5 hours ago, SteamyTea said:

Is this just the footprint of your house, or the total floor area, or even the total liveable area?

Just thinking that @joe90's place is about 200m² and his HP is 4 kW, and that does his DHW as well (I know that we are in the very mild SW, but you are not so different being on the west side of Scotland (Isle of Sky climate is very similar to Cornwall, within a degree).

Its the footprint of the thermal envelope, where the UFH will be (I.e. excludes garage) as it's about heat loss through the floor.  I suppose doing it accurately I would include the area in the rooms above garage as albeit they are internal floors - they will loose heat like the ground floor as the garage is not heated and vented. I did a quick check and doubled the size and made a tiny difference increasing the power required by 150W. 

[SuperJohnG]

3 hours ago, LA3222 said:

Have you considered giving the UFH design over to WundaTrade (as an example)? They will then draw it all up for you and give you a schedule of parts etc. 

 

On another note, (caveat, I am not an expert) I think with a well insulated house and the UFH pipes in the slab means that you don't need to get hung up on w/m2 O/P. I had a quick google last night and saw that max O/P in a screed is 100w/m2 which is well OTT. Having had a quick look at my UFH pipe layout Wunda designed it for 200mm spacings and a 70w/m2 O/P. Again this is well OTT and using the slab as a massive heat store means that the UFH will be run at a much lower temperature than the industry norm.

 

The O/P of the UFH doesn't matter as such because that doesn't heat the house directly, it is the O/P of the slab which matters as that provides a continuous source of heat, the job of the UFH is to dump sufficient heat into the slab.

I did and may well - but I wanted to understand for myself. I was planning on using wunda stuff as it gets a good rep (probably with an Ivar manifold as per @Nickfromwales recommendation - but that's a whole other thread ?, which I'll link with yours)

 

So did you give wunda a required w/m^2 value (for exmaple simialr to my 27W/m^2 - then tell them to work it out? seems as through 200mm spacings will eat the requirement no problem. 

 

1 hour ago, dpmiller said:

IIRC when I was doing the calcs for ours, LoopCAD always came out very pessimistic compared to jeremy's calculator.

That's the bit I was trying to understand - are those w/m^" values  what loopcad calculates as the required heat input.  

[LA3222]

I didn't give them a w/m2 value, just gave them the plans and left them to it. They came back with 150mm spacings at first iirc but I told them to drop it down to 200mm spacings as 150mm is OTT. Their plans say the 200mm spacings gives 70w/m2 O/P but they must be assuming a specific flow temperature for this value which they don't state.

 

Just checked the docs they sent me and I see two temperatures mentioned, 35 degrees and 45 degrees. So 200mm centres will be more than sufficient with a flow temperature in the low 20s.

 

As per most others on here I anticipate the flow temperature being down in the low 20s and I dont anticipate there being any issues at 200mm spacings. 

Edited November 10, 2020 by LA3222

[SteamyTea]

Using min OAT and adding a bit to allow the temperature to change is probably good enough.

I was, just because I was up early and bored, knocked up a very basic calculator. The thermal losses via ventilation dominated.

 

[LA3222]

1 hour ago, SteamyTea said:

The thermal losses via ventilation dominated.

 

It is surprising just how significant this can be. I played around with the spreadsheet a lot for my own house and additional insulation to the structure made little difference. Hence the argument can be made that the ROI for additional insulation is so long that it may not be worth it. Airtightness however does make a significant and tangible difference to heating loads and as such is well worth pursuing if one is that way inclined.

[SteamyTea]

3 minutes ago, LA3222 said:

It is surprising just how significant this can be. I played around with the spreadsheet a lot for my own house and additional insulation to the structure made little difference. Hence the argument can be made that the ROI for additional insulation is so long that it may not be worth it. Airtightness however does make a significant and tangible difference to heating loads and as such is well worth pursuing if one is that way inclined.

I think the trick is to get both right at the start of the design stage.  Then it is relatively easy to do the rest.

The one thing that is hard to model is the effect of windwashing.  A stiff SW wind is generally warm, but even a relatively slow NE is anything but.

 

I may try and incorporate some solar gain and wind effects later.  The trouble with even a basic model is verifying it against real data.  I can make just about any model fit existing data, not so easy the other way around.

[Adam2]

3 hours ago, SteamyTea said:

The thermal losses via ventilation dominated.

 

I'm intrigued by your ventilation losses, the only variables I think impacting this being (with my figures):

 

Volume (780m3), air-changes/hr (0.8), mean min daily temp delta (21.5), MVHR efficiency (85%), 

 

Appreciate my Jan mean min temp delta may be lower as I'm in the south and of course you may have a larger volume but I end up with ventilation losses of 645W (min in Jan) vs your 1045.

 

Would you mind sharing a bit more info, if only so that I can double check my workings/understanding?

 

2 hours ago, SteamyTea said:

I may try and incorporate some solar gain

I found the PVGIS website really handy for modelling this 

[A_L]

26 minutes ago, Adam2 said:

Would you mind sharing a bit more info, if only so that I can double check my workings/understanding?

 

I pretty much agree with your estimate (645W)

 

The formula I use is Volume(m³) x a.c.h. x  dT x 0.33 x MHVR inefficiency 0.15 which gives 780 x 0.8 x 21.5 x 0.33 x 0.15 = 664W

[SteamyTea]

I did not include MVHR in my calcs. Just trying a basic model out and I tend to see how close it gets to my actual data collection.

 

[SuperJohnG]

so does anyone have any thoughts - is it as simple as taking that maximum heat loss (3.6kW in my case) allowing some losses etc and then specifying an ASHP from there? I was thinking I would have to possibly go to a bigger unit than the 8.5kW but it doesn't seem like it now? 

[LA3222]

That's how I am working, feels a bit pie in the sky, however in the absence of an answer to the contrary from someone more experienced/qualified it is how i an rolling. 

 

That spreadsheet, coupled with various conversations littered throughout the forum lead me to believe that an 8.5kW ASHP should be more than sufficient for my needs, which in terms of heating are very similar to your own.

[SuperJohnG]

31 minutes ago, LA3222 said:

That's how I am working, feels a bit pie in the sky, however in the absence of an answer to the contrary from someone more experienced/qualified it is how i an rolling. 

 

That spreadsheet, coupled with various conversations littered throughout the forum lead me to believe that an 8.5kW ASHP should be more than sufficient for my needs, which in terms of heating are very similar to your own.

Defo does. But I'd agree generally where there is no feedback seems like as good as a yes, looks fine. 

 

8.5kW and 200mm spacing it is, seems fine to me. Let me know if we have any issues @LA3222 I'm definitely glad you're a year ahead and pretty much same build. Very helpful. 300 or 400l UVC for you? 

[Adam2]

@SuperJohnG what will you do for hot water? Also - I made a note above but didn't tag you relating to ventilation losses as yours seem very high, may be worth checking

[SuperJohnG]

1 minute ago, Adam2 said:

@SuperJohnG what will you do for hot water? Also - I made a note above but didn't tag you relating to ventilation losses as yours seem very high, may be worth checking

Apologies- I thought that was for @SteamyTea. 

 

3 or 400L UVC heated by the ASHP, with secondary immersion. 

 

I had put in 2 ACH as I thought this might be conservative. But maybe its overly conservative? It should be fairly tight out the box as it is a SIPS build. Is that what you were referring to? 

[SuperJohnG]

@Adam2 quick check and if I reduce my ACH to 0.8, the ventilation heat loss drops from 1045 to 418W 

[Adam2]

Yes that's probably more realistic for air tightness, maybe check with supplier re what is achievable. 

 

You'll need to maybe think of heating all water on a cold day and base size on that for worst case. I'm no expert, just what I've done for mine. 

[dnb]

@SuperJohnG I must confess to not reading all the maths in great detail, but it looks to me like the heat loss numbers aren't too far out. The house ground floor area isn't too different to mine and the construction is comparible, and so is the heat loss. I too agree with @SteamyTea that the infitration losses dominate at least in the model. This is why the inside of my house is beginning to look like a very untidy satellite factory.

 

The model is of heat loss. If you aim to just service this loss, then there's no scope in the system to change the temperature of the house at anything other than a snails pace. Imagine the house at say 18 degrees, and you want it at 20 when it is at your worst case design temperature, then the heat input will mostly be offsetting the losses and not acting to change the temperature. My rule of thumb was to double the energy required to hold a steady temperature in the worst expected case, so in your case something of the order of 8kW.

 

I should point out that my "as designed" SAP calculation says will boil everyone in the house unless it's like Siberia outside. I remain unconvinced of this.

 

[LA3222]

@dnb unless I am missing a trick John has not opted to 'just service' his calculated heat losses.  His peak loss is estimated at 3.6kW for approximate dT of 20 degrees and 2ACH and opted to go for 8.5kW ASHP.

 

I would say the estimation of heat loss is sound and the ACH which has a significant impact is, if anything, a conservative estimate. So there is scope for the losses to come down. An 8.5kW ASHP will be more than sufficient for the task at hand.

 

The mean minimum temperature is 1.4 degrees for January, so you could argue a case for times when it goes sub zero. The model shows that even at -5degrees the loss is 5kW, so well within the capacity of an 8.5kW ASHP. Also it is unlikely to remain at -5degress for a 24hr period.

 

Snails pace - my understanding of using the floor slab as a 'radiator' is that it does act at a snails pace. It will not react rapid the same way that pipes in a screed will do. It takes a long time to dump sufficient heat into it to reach your desired house temperature. Equally it will take a long time to lose that heat so it serves to iron out the peaks and troughs and maintain the house temp at constant levels, requiring periodic 'top ups' to maintain temp.

 

One final comment ref the accuracy of the excel model used. Anecdotal evidence from members whom have completed their homes and modelled their energy usage suggest that the model is a very good starter for ten and is accurate enough to base your projected energy demand upon whilst designing your home.

Edited November 11, 2020 by LA3222

[SteamyTea]

34 minutes ago, LA3222 said:

The mean minimum temperature is 1.4 degrees for January, so you could argue a case for times when it goes sub zero. The model shows that even at -5degrees the loss is 5kW, so well within the capacity of an 8.5kW ASHP. Also it is unlikely to remain at -5degress for a 24hr period.

Using that mean temperature, and a standard deviation for the whole month of 4°C, -5°C will happen for 3 percent of the time.  22 Hours.

 

A 1m by 1m by 0.1m concrete block, heated at 25W/m² will take 2.39 hours to raise up 1°C.

If it looses 5W/m² to the ground, then 3 hours.

If another 12W/m² goes to heating the air (which is what you are after) then 7.5 hours.

 

(only had 1 mug of tea so far so may be wrong)

 

Edited November 11, 2020 by SteamyTea

[LA3222]

56 minutes ago, SteamyTea said:

Using that mean temperature, and a standard deviation for the whole month of 4°C, -5°C will happen for 3 percent of the time.  22 Hours.

 

A 1m by 1m by 0.1m concrete block, heated at 25W/m² will take 2.39 hours to raise up 1°C.

If it looses 5W/m² to the ground, then 3 hours.

If another 12W/m² goes to heating the air (which is what you are after) then 7.5 hours.

 

(only had 1 mug of tea so far so may be wrong)

 

Interesting numbers, confirms my understanding  in so far as using the slab for heating is the long game. Zero point in trying to use it like a standard radiator. I expect that once my house is occupied it will take a period of adaption to get used to how to live with it as we tend to be used to instant heat when a dial is turned up.