Fitting wet UFH into a very old cottage

[blankton]

Hi. I have a very old cottage which I am retrofitting with wet UFH. It is in a village with no gas, so heating options were limited to oil, LPG or electric. 

 

The cottage had a damp bitchumen floor, which needed dealing with, so I have laid a new "breathable" limecrete slab, complete with UFH pipes embedded. I plan to heat this from an ASHP. My Plumber (my brother, so I am not getting a new one 😊) doesn't do a lot of UFH, so this will be somewhat of a learning curve for us both. We have got some questions:

 

I was going to do a zone for each room. However, I have been doing some reading and I believe this is now a bad idea, as small zones will make the pump cycle more. I have a 5 zone manifold. If I stick with the 5 circuits (2 downstairs and 3 upstairs) could I wire the actuators for the two circuits downstairs in parallel, so that thermally they were just one "zone", but using two circuits on the blending manifold. Could I then balance the temp in the different circuits by adjusting the flow through each one, as you would with a radiator? likewise could I then link the actuator wiring for the 3 circuits for the upstairs, so that these all open/close together? 

 

The upstairs wont be in a slab. The floor in two of the three rooms has been replaced with new joists and chipboard. I have the ceiling down. I was going to fit UFH pipe from below using the "fit from below" spreader plates. The third room upstairs still has the original oak joists and wide floorboards. The bottom of these are visible from the living room, so I can't use the "fit from below" type system. For this room I was going to use an overfit type board. Is it ok to mix and match both fitting types in the same installation? The floor is already not level and the overfit board will actually work in my favor in terms of levelling up the different floor heights. 

 

For the bathroom, originally I was going to put wet UFH under it, but I am now concerned that there might be times that you want a bit of warmth on the tiled floor in spring/autumn, when the rest of the system might not be on. Do I just fit electric UFH in this small room. On the face of it this will be more expensive to run, but if I end up having the ASHP cycling just to run this small room, then maybe it would be for the best. I am sure this question must have been asked before, what is the usual solution?

 

Manifold placement. I have fitted UFH in my downstairs extension. Manifold placement was simple, as it is only in one room. In the cottage install I have ran the pipes that are already in my slab to the cupboard under the stairs. Now that I've done this I have some reservations! If I place the manifold on the ground floor, I will end up with a messy spaghetti, with the circuits going upstairs looping back on themselves. I wonder if I should have placed the manifold on the first floor, so that all pipes are exiting downwards to the circuit. Again, I am sure this must be a common question.   

 

Thanks for your time. Here are a few photos that might be of interest:

 

Foamed glass insulation under the slab:

 

 

 

 

Original part of the house with oak joists. The overfit system will be above this:

 

 

 

2ft thick walls with a rubble/earth infill:

 

 

 

First floor brickwork in handmade bricks, this was behind a cement render:

 

 

 

Edited February 23, 2023 by jack
Images converted from links

[Conor]

What are your overall U values and heatlosses for each room? No point thinking or planning any further until you've that work done first. As this sounds like a full on renovation and extension I'm sure you're at least at the required building control levels of insualtion? Have you upgraded the glaslzing etc? What's the proportion of old uninsulated walls to new, insualted walls? Is there a loft you can insulate? 

 

Before you plan about how you're going to put heat in to a building, you have to plan on stopping heat getting out.

 

Plenty of topics on here about heating old buildings, have a search.

 

You need a manifold on each floor (various reason, mostly to deal with air in pipes), each with pipes coming from the ASHP ) buffer tank.

 

UFH might not be the best option for you, esp upstairs.

[Temp]

6 minutes ago, blankton said:

I was going to do a zone for each room. However, I have been doing some reading and I believe this is now a bad idea, as small zones will make the pump cycle more. I have a 5 zone manifold. If I stick with the 5 circuits (2 downstairs and 3 upstairs) could I wire the actuators for the two circuits downstairs in parallel, so that thermally they were just one "zone", but using two circuits on the blending manifold. Could I then balance the temp in the different circuits by adjusting the flow through each one, as you would with a radiator? likewise could I then link the actuator wiring for the 3 circuits for the upstairs, so that these all open/close together? 

 

Yes that will work fine. Many houses with large rooms or open plan need two loops in one room. These are driven by two actuators linked to one stat to form one zone.

 

9 minutes ago, blankton said:

For the bathroom, originally I was going to put wet UFH under it, but I am now concerned that there might be times that you want a bit of warmth on the tiled floor in spring/autumn, when the rest of the system might not be on. Do I just fit electric UFH in this small room. On the face of it this will be more expensive to run, but if I end up having the ASHP cycling just to run this small room, then maybe it would be for the best. I am sure this question must have been asked before, what is the usual solution?

 

Have you considered installing both? Only use the electric when the rest of the house is off. 

 

Our bathroom has wet UFH but it's oil fired. It's wired to a stat like the rest of the house so it's probably on when the rest of the house needs heating. We don't turn it on out of season.

 

16 minutes ago, blankton said:

Manifold placement. I have fitted UFH in my downstairs extension. Manifold placement was simple, as it is only in one room. In the cottage install I have ran the pipes that are already in my slab to the cupboard under the stairs. Now that I've done this I have some reservations! If I place the manifold on the ground floor, I will end up with a messy spaghetti, with the circuits going upstairs looping back on themselves. I wonder if I should have placed the manifold on the first floor, so that all pipes are exiting downwards to the circuit. Again, I am sure this must be a common question.   

 

Most people have a manifold on both floors. One good reason for having one upstairs it to make air bleeding easy/possible. Eg the upstairs manifold is higher than upstairs loops so you can fit an air valve one it.

[Nickfromwales]

20 minutes ago, blankton said:

I was going to do a zone for each room. However, I have been doing some reading and I believe this is now a bad idea, as small zones will make the pump cycle more. I have a 5 zone manifold. If I stick with the 5 circuits (2 downstairs and 3 upstairs) could I wire the actuators for the two circuits downstairs in parallel, so that thermally they were just one "zone", but using two circuits on the blending manifold. Could I then balance the temp in the different circuits by adjusting the flow through each one, as you would with a radiator? likewise could I then link the actuator wiring for the 3 circuits for the upstairs, so that these all open/close together? 

This will be very crude, and an absolute bitch to balance out IMO.

 

For the 2 different types of emitters, and the fact they are then over 2 different floors, mandates a second manifold upstairs eg so then 2 different flow temps are possible. 

You 5 million % need room stats and actuators and then each space can keep calling for heat as long as it needs to ( maybe after others have reached the set point and have closed off ). 

 

Don't worry about cycling the ASHP as you will also defo be fitting at least a 100L buffer tank, for this type of dwelling and with the type of system you have fitted / intend to fit.

You underestimate, I think  how much hotter the 1st floor system will have to run to do the same job as the ground floor one, so sending one temp via one manifold will be a recipe for major disappointment, sorry.

 

Air rises, so a GF manifold serving FF UFH is an air trap waiting to happen.

 

Unless this house is super insulated and very draft-proof then I fear this system may fail to meet your hopes and expectations. Do some heat loss calculations and revisit before taking one more step is my advice. 

[blankton]

35 minutes ago, Conor said:

What are your overall U values and heatlosses for each room? No point thinking or planning any further until you've that work done first. As this sounds like a full on renovation and extension I'm sure you're at least at the required building control levels of insualtion? Have you upgraded the glaslzing etc? What's the proportion of old uninsulated walls to new, insualted walls? Is there a loft you can insulate? 

 

Before you plan about how you're going to put heat in to a building, you have to plan on stopping heat getting out.

 

Plenty of topics on here about heating old buildings, have a search.

 

You need a manifold on each floor (various reason, mostly to deal with air in pipes), each with pipes coming from the ASHP ) buffer tank.

 

UFH might not be the best option for you, esp upstairs.

 

 

There is no extension, so all the walls are old. I have done some preliminary calculations on the heat loss. The house has 300mm of insulation in the roof. It is double glazed. The downstairs is 2-3ft thick stone walls with a loose rubble infill. The basic U value tables for these walls suggest they are very bad, but I have found a masters paper where they have measured actual values for these type of walls and they have found them to be significantly better than expected - I'd guess because there is all sorts of voids,  sticks, brussel sprout stalks, straw etc jammed in the infil. The upstairs is a solid 9" brick, but with an air gap and insulated plasterboard. I may overboard this further with some more insulation.  As old houses go I believe it to be well insulated, but there is obviously a huge margin of error in calculating the heat loss on a building with unusual construction such as this. 

 

I has assumed that with the low temperature flow from the ASHP that UFH would be best. I gather that if I go for radiators they need to be massive and potentially have the flow temp of the ASHP turned up higher?

[blankton]

29 minutes ago, Nickfromwales said:

 

You underestimate, I think  how much hotter the 1st floor system will have to run to do the same job as the ground floor one, so sending one temp via one manifold will be a recipe for major disappointment, sorry.

 

I realise that the ground floor slab will take a long time to get warm and a long time to cool. The first floor will have less thermal mass and will heat and cool faster. Are you saying that the UFH flow temp would also have to be higher on the first floor? How high can you go? I thought that you were limited to a floor temp of about 27C as it can damage flooring if it goes higher?

[Nickfromwales]

44 minutes ago, blankton said:

I thought that you were limited to a floor temp of about 27C as it can damage flooring if it goes higher?

Yup, but the amount of heat lost in the transfer from the pipe to the deck and then to the floor will be significant.

I've not seen the job, and I've not done the calcs, but I am speaking from experience of 30 years of heating and hot water, and I say you're off to a wonky start here.

The ASHP is going to be running at almost full capacity, and that will likely also attract frosting when you need it the least.

48 minutes ago, blankton said:

The first floor will have less thermal mass and will heat and cool faster.

The FF UFH will have wooden floorboards over it, and then (?) more finished flooring. Thermal mass here is moot, as you need to worry about getting the energy in the water inside the pipe conveyed to the room interior. That's quite a journey. The spreader plates will all need stapling up from underneath so you have as close to 100% surface contact between the plates and the deck ( floorboards ) as is humanly possible. Neglect this and you'll be retrofitting rads afterwards.

 

1 hour ago, blankton said:

The cottage had a damp bitchumen floor, which needed dealing with, so I have laid a new "breathable" limecrete slab, complete with UFH pipes embedded.

Can you tell us what is under all of this? Insulation thickness ( if any ? ). 

[blankton]

6 hours ago, Nickfromwales said:

Yup, but the amount of heat lost in the transfer from the pipe to the deck and then to the floor will be significant.

I've not seen the job, and I've not done the calcs, but I am speaking from experience of 30 years of heating and hot water, and I say you're off to a wonky start here.

The ASHP is going to be running at almost full capacity, and that will likely also attract frosting when you need it the least.

The FF UFH will have wooden floorboards over it, and then (?) more finished flooring. Thermal mass here is moot, as you need to worry about getting the energy in the water inside the pipe conveyed to the room interior. That's quite a journey. The spreader plates will all need stapling up from underneath so you have as close to 100% surface contact between the plates and the deck ( floorboards ) as is humanly possible. Neglect this and you'll be retrofitting rads afterwards.

 

Can you tell us what is under all of this? Insulation thickness ( if any ? ). 

 

Its a shame the forum doesn't seem to allow photos, as you'd have seen the photos of the slab.  There is about 400mm of a product called "foamed Glass" insulation under the limecrete slab. If you google "limecrete floor" you will get all the information on this. I believe that this will be suitable. 

 

We had already come to the conclusion that the spreader plates needed to be nailed up very tight. We were wondering about also putting some PU glue on them, to make sure they never try and come loose, but obviously this will be another mini layer of insulation. Any thoughts on this? It almost sounds as if you  are saying that the spreader plate solution would never work very well, its always going to have to conduct heat through the floorboards, as they cant be nailed to thin air.  

 

I realised the upstairs UFH wont transfer the heat as well as the downstairs pipes encapsulated in the slab, but I figured that upstairs would also be getting some of the "heat rises" effect from the downstairs (which I expect to be a very stable warm temperature), with the bedrooms likely just needing a boost morning and evening.  

 

I'm finding this really interesting. I've heard lots of stories of people that have replaced a gas boiler with an ASHP through their existing rads, and it hasn't worked due to the rads being too small. I thought I was being smart in going all UFH to have the lowest flow temps, but you seem to be telling me I actually need rads?! I feel I should also reiterate, I have no mains gas (otherwise there would already be a nice efficient combi), so I am effectively competing with storage heaters. I only have a very small yard, so an oil tank or LPG would be quite difficult as well.   

[jack]

The forum welcomes photos but not via external URL. You need to upload them when composing the post.

 

I'll edit your first post to include them when I'm at my desk later this morning.

[jack]

3 hours ago, blankton said:

Its a shame the forum doesn't seem to allow photos, as you'd have seen the photos of the slab. 

Photos added to the first post.

 

For future posts, you can add photos using the tools underneath the text window:

 

[Nickfromwales]

3 hours ago, blankton said:

We were wondering about also putting some PU glue on them, to make sure they never try and come loose, but obviously this will be another mini layer of insulation. Any thoughts on this?

Definitely do NOT put any kind of glue under the spreader plates, particularly . Just staple the living daylights out of them and that will take you to the max potential that you will ever see. Any kind of separation will have a massive impact on surface to surface contact, and the cured glue won't transfer heat. PU is an insulator!

 

 

4 hours ago, blankton said:

I'm finding this really interesting. I've heard lots of stories of people that have replaced a gas boiler with an ASHP through their existing rads, and it hasn't worked due to the rads being too small. I thought I was being smart in going all UFH to have the lowest flow temps, but you seem to be telling me I actually need rads?! I feel I should also reiterate, I have no mains gas (otherwise there would already be a nice efficient combi), so I am effectively competing with storage heaters. I only have a very small yard, so an oil tank or LPG would be quite difficult as well.   

What we try to do here, is tell you the pro's / con's / caveats, and then you can make your own, informed, decisions. You're a bit past there as you've already jumped in and done the GF UFH.

What you can do now is possibly look at the most efficient ASHP for your situation, and then also size it correctly ( so it doesn't get installed, and then simply scream until it dies early ). FYI, most ASHP's will have a stated service life of 7 or ( a few ) more years. After the stated warranty / service period, you're then on borrowed time and the design of the system will dictate how hard a life / how long it is then going to last.

You may benefit from a high-temp split ASHP vs a mono-block for example, but if your brother only knows how to plumb vs how to effectively design a suitable installation and specify the equipment to compliment that design, then you may need to step on the brakes and re-think this before putting good money after bad / not getting the results you think you will.

Not trying to piss on your chips, just better to stop and think for 5 than steam on setting fire to time and money  

[blankton]

On 23/02/2023 at 10:44, Nickfromwales said:

 

What we try to do here, is tell you the pro's / con's / caveats, and then you can make your own, informed, decisions. You're a bit past there as you've already jumped in and done the GF UFH.

 

 

Whatever I was doing, I'd have laid the ground floor underfloor heating, as I needed a new floor, so its literally a couple of hundred quid extra to put some pipe in the floor. I have GF UFH in my solid walled Victorian kitchen and its great, so there is no way I wouldn't have incorporated this on the ground floor, regardless of the boiler type. 

 

However, I am kind of wed to the ASHP solution for other reasons - I've been given a Grant 6Kw invertor heat pump at cost price. We made a bit of a fag packet assumption that  it would be a good size. Since my last post I have firmed up the calculations and I think our back of the fag packet calculation is vindicated. I have made a calculation for each room. Working out the thermal resistance / conductivity of the walls was a bit of guesswork, but I found a few academic studies where they measured this directly on loose rubble walls, so I have used this data. Most walls are insulated and there is a lot of insulation in the loft and beneath the slab.

 

This means I have a loss of 3-4Kw depending on which U values you believe. I have likely been a bit optimistic, as there will be cold bridges and I have a chimney where the construction is a bit of an unknown. So I reckon I could add another 20 or 25% "contingency" on top of the calculated value. Although there is a lot of uncertainty, I think that the pump I've been given is an appropriate size. Or to put it another way; I don't really see how I can be any more accurate in my forecast, when I have such uncertainty on the construction methods!

 

Where this has become a bit of an eye opener is the output needed per M2. In the upstairs rooms I am looking at 30-35 ish, which from what I can tell will require a flow temp of about 45C. This is higher than I thought I would need, and I assume will reduce the efficiency of my heat pump. I have tried to work out radiator sizes, but if assuming a flow temp of 45C, the delta T is only 15C, so there is a 0.7ish reduction factor, meaning that I would need big radiators.

 

So whichever way I look at it (rads or UFH pipe), I will need a flow temp a bit higher than I initially envisaged. 

 

Sorry for the waffle. Does this sound reasonable? Will I be highly inefficient running at this flow temp?   

 

Forgot to mention, there will also be a 5kw log burner. My heat loss was based on a -3 outside and 20C inside. However, I think in all likelihood if it got that cold the log burner would be on. 

 

 

 

 

 

Just realised the above table says kw, but its just watts. 

Edited March 12, 2023 by blankton

[Roger440]

I have nothing much useful to say, other than i did limecrete over foamed  class downstairs in the old part of my house. Its very nice.

 

Rest of house is on conventional rads, and runs off an oil boiler. 

 

Input temp to floor is 40c if that helps.

 

Quite easily holds the room at 21c regardless of outside temp. 9 inch walls, no insulation. Though they are a very lightweight brick which, now they are dry, by any measure of logic makes it twice as good as a conventional, old, cavity walled house. 

[lookseehear]

On 22/02/2023 at 21:37, blankton said:

Hi. I have a very old cottage which I am retrofitting with wet UFH. It is in a village with no gas, so heating options were limited to oil, LPG or electric. 

 

The cottage had a damp bitchumen floor, which needed dealing with, so I have laid a new "breathable" limecrete slab, complete with UFH pipes embedded. I plan to heat this from an ASHP. My Plumber (my brother, so I am not getting a new one 😊) doesn't do a lot of UFH, so this will be somewhat of a learning curve for us both. We have got some questions:

 

I was going to do a zone for each room. However, I have been doing some reading and I believe this is now a bad idea, as small zones will make the pump cycle more. I have a 5 zone manifold. If I stick with the 5 circuits (2 downstairs and 3 upstairs) could I wire the actuators for the two circuits downstairs in parallel, so that thermally they were just one "zone", but using two circuits on the blending manifold. Could I then balance the temp in the different circuits by adjusting the flow through each one, as you would with a radiator? likewise could I then link the actuator wiring for the 3 circuits for the upstairs, so that these all open/close together? 

 

The upstairs wont be in a slab. The floor in two of the three rooms has been replaced with new joists and chipboard. I have the ceiling down. I was going to fit UFH pipe from below using the "fit from below" spreader plates. The third room upstairs still has the original oak joists and wide floorboards. The bottom of these are visible from the living room, so I can't use the "fit from below" type system. For this room I was going to use an overfit type board. Is it ok to mix and match both fitting types in the same installation? The floor is already not level and the overfit board will actually work in my favor in terms of levelling up the different floor heights. 

 

For the bathroom, originally I was going to put wet UFH under it, but I am now concerned that there might be times that you want a bit of warmth on the tiled floor in spring/autumn, when the rest of the system might not be on. Do I just fit electric UFH in this small room. On the face of it this will be more expensive to run, but if I end up having the ASHP cycling just to run this small room, then maybe it would be for the best. I am sure this question must have been asked before, what is the usual solution?

 

Manifold placement. I have fitted UFH in my downstairs extension. Manifold placement was simple, as it is only in one room. In the cottage install I have ran the pipes that are already in my slab to the cupboard under the stairs. Now that I've done this I have some reservations! If I place the manifold on the ground floor, I will end up with a messy spaghetti, with the circuits going upstairs looping back on themselves. I wonder if I should have placed the manifold on the first floor, so that all pipes are exiting downwards to the circuit. Again, I am sure this must be a common question.   

 

Thanks for your time. Here are a few photos that might be of interest:

 

Foamed glass insulation under the slab:

 

 

 

 

Original part of the house with oak joists. The overfit system will be above this:

 

 

 

2ft thick walls with a rubble/earth infill:

 

 

 

First floor brickwork in handmade bricks, this was behind a cement render:

 

 

 

Did you complete this project? We're looking to do something very very similar and would be keen to hear your experience. Would also love to hear how much it cost if you're willing to share, but I understand if not.

[blankton]

On 23/12/2023 at 15:55, lookseehear said:

Did you complete this project? We're looking to do something very very similar and would be keen to hear your experience. Would also love to hear how much it cost if you're willing to share, but I understand if not.

 

Hi Sorry,

I have only just seen this. 

 

Still not finished. I am just getting to the stage of piping up my ASHP. Even without the heat on the foamed glass / limecrete does seem to have completely dried out the ground floor. 

[Dave Jones]

100mm pipe centres and no zones!