UFH pipes in the walls, not floor? Silly idea?

[Dreadnaught]

UFH pipes go under the floor but, when linked by a ASHP,  they can be used for cooling as well as heating, @JSHarris-style.

But… air convection works against the cooling effect at the floor. So wouldn't it make sense to put the UFH cooling pipes in the ceiling not under the floor?

 

Or… is the best compromise for both cooling and heating is not put them in the floor or ceiling but in the walls instead? Is this a silly idea?

 

(I thought of this because, in my build, I may not have a concrete slab and have an insulated timber-frame floor instead. It strikes me that it could be just as easy to put the pipes in some of the internal walls as it is it put them on the floor.)

 

 

 

 

 

[Jeremy Harris]

The problem with putting pipes in the walls is people like to bang nails and screws into walls to hang and fix things.

 

In practice, cooling the floor works very well, better than I expected, and I suspect that's because air in an occupied house tends not to stratify very well, and is almost constantly moving around, especially with MVHR.

[Nickfromwales]

1 minute ago, Dreadnaught said:

UFH pipes go under the floor but, when linked by a ASHP,  they can be used for cooling as well as heating, @JSHarris-style.

But… air convection works against the cooling effect at the floor. So wouldn't it make sense to put the UFH cooling pipes in the ceiling not under the floor?

 

Or… is the best compromise for both cooling and heating is not put them in the floor or ceiling but in the walls instead? Is this a silly idea?

 

(I thought of this because, in my build, I may not have a concrete slab and have an insulated timber-frame floor instead. It strikes me that it could be just as easy to put the pipes in some of the internal walls as it is it put them on the floor.)

As cooling will make up a very small percentage of your 'climate control', id seriously rethink not having a slab.

With primarily heating being required, and some sort of emitter necessary to facilitate that effect, look at how your going to deliver heat into the dwelling and how best to produce, store and emit that in line with what your heat source will be. Ive just spec'd a build ( Beatie PH ) with 360^o EPS and no slab or UFH. Getting an ambient via a Genevex HP based MVHR system is being both difficult and unreliable. Luckily for me A.N.Other is doing supply and fit of the MVHR and they have to demonstrate that it will work. PHEW! @PeterStarck has a similar arrangement in a much smaller dwelling and is surviving with that, but admits he keeps his space heating temp quite high ( ~24^oC iirc ) to ensure a constant ambient and again iirc he also fortifies that with auxiliary ( supplemental ) heating.

 

With a slab you have a good medium to store for, eg, heat energy produced via PV > ASHP > SLAB in the daytime and allow that heat to dissipate gently into the house through the late evening / night time, dependant on your slab insulation and heat loss figures of course. With no slab and 'direct emitters' your heating will be nigh on linear to it being required so you may find yourself 180^o out of phase with the 'free' energy you should be consuming and utilising to offset such demand. The B PH I refer to above is basically an EPS cage with very poor decrement delay, and has been a royal PITA to arrive at a space heating solution for. In the end I left the client and the MVHR designer to it, as I just kept on finding flaws in their proposal every time they submitted it for sign-off and ( deposit ) payment. Got on my tits in the end TBH, and now the guy has to fit a huge feature radiator in the open living space and keep it on, plus heated towel rads in all 3 bathrooms which all also have to stay on in order to maintain an ambient of 20,5^oC. To get the bedrooms to 22^oC they have advise fitting auxiliary heating or to light his 3kW WBS !!! P.I.T.A.

The next one I get with the same remit, I shall not be taking it on. Way too problematic. Slab for me every day of the week. So simple, so effective and will cool sufficiently well as documented on his blog, and quoted above, by the beard of wisdom ( in a house that is 'live' and performing admirably ). 

[Dreadnaught]

29 minutes ago, Nickfromwales said:

As cooling will make up a very small percentage of your 'climate control', id seriously rethink not having a slab.

 

 

Thanks @Nickfromwales. Very interesting and good advice. I only wish I could have a slab! The problem is a tricky site with protected tree roots and a risk of clay heave. The solution looks like screw piling with a void beneath the floor. I am still in the design phase and am hoping things might change by talking to the right SE. Listening to you, I am keen that they will.

 

(Oh and I wont have PV either. I have a whopping (if somewhat sick) chestnut tree slap bang due south and inches from the edge of my plot. I am in a conservation zone. Trust me, I shall be having a close look at this tree).

 

29 minutes ago, Nickfromwales said:

The B PH I refer to above is basically an EPS cage with very poor decrement delay

 

Very good point. I raised the question of decrement delay with Beattie recently but did not receive an answer. I even asked if they would substitute blown cellulose insulation for EPS beads but they said they would not. 

Edited August 27, 2018 by Dreadnaught

[Dreadnaught]

@Nickfromwales, thinking some more, and taking note of what you said about emitters and the wish to maintain a constant ambient temperature across the day/night cycle in the winter, I tried to imagine the simplest solution possible.

 

Assuming a passive house, this is what I came up with:

• E7 electricity
• electric mat underfloor heating throughout, which is on at night with E7
• Willis heater and Sunamp* for DHW, again charged at night from E7
• cooling: install a cheap (sub £1k) air conditioner (otherwise known as an air-to-air heat pump), if needed

Strikes me that this hits all the marks. It would not give the SPFH2 gains from a heat pump, nor take advantage of the cheapness of gas. But, my word, it would be relatively cheap to buy and install and low maintenance! Nick, what do you think?

 

* I believe the Willis heater can now be built in to the SunAmp

 

[Construction Channel]

There is also potential for a lot of air locks, especially if you went up and down between the studs. 

[Dreadnaught]

Thanks @JSHarris & @Construction Channel.

 

That's clear then! It goes in the draw labelled Silly Ideas.

[PeterW]

28 minutes ago, Dreadnaught said:

electric mat underfloor heating throughout, which is on at night with E7

 

Issue with this is where it’s installed - which is usually just below the floor finish. What you want is a slab absorbing the heat which lends to a solid floor with UFh pipe in it and a Willis heater on this - this is how @TerryE has done his. 

[Dreadnaught]

1 minute ago, PeterW said:

Issue with this is where it’s installed - which is usually just below the floor finish. What you want is a slab absorbing the heat which lends to a solid floor with UFh pipe in it and a Willis heater on this - this is how @TerryE has done his. 

 

True. Terry uses off-peak electricity to store heat in his slab at night, via wet UFH. That heat is then released into his house gradually over the subsequent hours.

 

But with a passive house, I am not sure the slab heat-buffer is needed. A passive house needs so little heating that even direct heating at night using electric mat UFH of a few kW might be sufficient to maintain a constant temperature throughout the day/night cycle even in the depths of winter.  Of course I would need to do the calculations but this is my hunch.

[Jeremy Harris]

The major issue to address is the thermal time constant of the house.  Insulation is a part of this, but perhaps more important is the combination of the heat capacity of the inside of the house (to a maximum of around 100mm depth from the internal wall/ceiling/floor surface) and the thermal conductivity of that same layer.

 

It's also important to ensure that the insulation doesn't just have the right U value, but that it has a nice long decrement delay.  The idea is to delay heat getting in or out of the house so that the natural diurnal variation in temperature isn't reflected by similar cyclic changes inside the house.

 

In our case, the combination of a fair bit of sensible heat stored in the concrete slab, which also has a pretty good thermal conductivity, plus the use of a relatively thick layer of high'ish decrement delay insulation in the walls and roof, result in very slow changes to the internal house temperature.  This makes it relatively easy to maintain the house at a steady temperature all the time.

 

Most of the heat stored in our house comes from the slab and the plasterboard (plasterboard is better at storing heat than concrete - gypsum has a higher heat capacity than concrete). 

[Dreadnaught]

@JSHarris, yes. Thank you. I have learnt this from you.

 

Decrement delay: I have rejected EPS beads for this reason and am only targeting frame systems that use blown-cellulose insulation.

 

My design has a very poor form factor, 4.9. Its a bungalow with a flat roof. As I am following a fabric-first approach and passive house, this probably means 450 mm of blown cellulose all around, possibly even in the floor. A benefit of this will hopefully be a desirable decrement delay. Is is in this context that electric mat UFH is appealing to me for space heating.

Edited August 27, 2018 by Dreadnaught

[Nickfromwales]

11 hours ago, Dreadnaught said:

(Oh and I wont have PV either. I have a whopping (if somewhat sick) chestnut tree slap bang due south and inches from the edge of my plot. I am in a conservation zone. Trust me, I shall be having a close look at this tree).

Does it have a TPO on it? If not @Construction Channel has a job for his 24"er. Buzz, Buzz, chop, Chop, Ta, Ta.  Plant some new ones where you want them, to alleviate any guilt, or just wait for the guilt to naturally reside........about a day or two max should suffice  

PV is a must IMO, and id never consider a PH without it, even more so if your to go all-electric. 

 

11 hours ago, Dreadnaught said:

Very good point. I raised the question of decrement delay with Beattie recently but did not receive an answer. I even asked if they would substitute blown cellulose insulation for EPS beads but they said they would not. 

Exactly what happened in the case I recounted. A blown eps jacket over the walls and roof, and EPS / PIR afoot makes for near transparency. Switch the heating off and watch the temperature drop. No interest in changing to cellulose / other so a bit one-trick and very niche imo. B PH is for the true eco warrior AFAIC, and doesn't lens itself to sympathetic "gentle" heating like a slab does. 

 

10 hours ago, Dreadnaught said:

Assuming a passive house, this is what I came up with:

• E7 electricity
• electric mat underfloor heating throughout, which is on at night with E7
• Willis heater and Sunamp* for DHW, again charged at night from E7
• cooling: install a cheap (sub £1k) air conditioner (otherwise known as an air-to-air heat pump), if needed

Strikes me that this hits all the marks. It would not give the SPFH2 gains from a heat pump, nor take advantage of the cheapness of gas. But, my word, it would be relatively cheap to buy and install and low maintenance! Nick, what do you think?

What do I think?......LOL.  Let me just fetch my gun  

E7 - yes

Electric - UFH no. And thinking you can only run that at night also a no. With that kind of decrement delay you'll need a constant trickle of heat, day and night, in the shoulder and winter months. 

Willis heater and Sunamp for DHW on E7 - now this is where it gets interesting;

You can, firstly, go for a wet cylinder OR a SA unit, as both will achieve the same in my following proposal. Pros and cons first......

 

Wet cylinders eg buffer and UVC;

PRO : cheaper capital outlay vs SA

PRO : more plumbers will quote for fitting wet cylinders vs the many who go running away when someone mentions you want to fit a SA

CON : physically bigger in comparison by a factor of ~3 per same kWh of heat energy stored

CON : much much higher standing losses, to a point that it may be problematic and require mitigation measures a la @JSHarris issues with his wet TS

CON : certified installation for G3 required and also annual G3 inspection / service for life @ ~ or > £100 P/A

CON : multiple complex pressure reducing valves, pressure relief valves, associated discharge pipework and more

 

Sunamp units;

CON : more expensive to buy than SOME UVCs / buffers etc, but not all. Some HP UVCs are more than an equivalent SA

CON : less plumbers will understand and warm to the idea of quoting for and installing a SA so you will need to persevere with that

PRO : physically smaller as above, and a neater 'modular' box set to fit in under a kitchen / utility worktop ( up to the size 9 unit, the size 12 unit is over 1000mm tall )

PRO : very low standing losses so the heat you store stays in the SA for longer, allowing you to further offset storage and times of consumption eg store heat off PV and then consume at night.

PRO : no compulsory annual inspection but obviously needs routine / annual inspection to check things are operating A1. To assume something is fit n forget would be wrong. The benefit of the SA is that this is far easier / simpler to undertake than the alternatives as there is basically nothing to do. I will double check this tomorrow and update as to whether or not the SA can be DIY 'maintained' as in who must fill out the annual Benchmark service interval record etc. 

PRO : can be connected directly to the mains as the plate heat exchangers are rated at 10bar. That means full flow from the cold mains to DHW vs the multi block ( aka control group ) that is required with an UVC, and something that has been demonstrated here that actually restricts the flow rate somewhat.

 

OK, that due diligence taken care of.

Now for my 2 penneth.

 

I would fit aluminium spreader plates and wet UFH. Its as cheap and quick as electric systems but is far more reliable. Also, electric systems can have far less longevity / be far more prone to being damaged during the installation and that will often cause premature, but not immediate, failure of the delicate heater wire. 

Run that from whichever type of heat source you decide suits you best, wet cylinder or SA, and then you have good cause to get a chunk of PV fitted. Store the excess PV into the chosen storage device and then utilise it outside the hours of PV generation. Fortify with a Willis heater, or even easier via the supplied immersion heaters ( regular immersion in an UVC / buffer / TS, and there is,

10 hours ago, Dreadnaught said:

* I believe the Willis heater can now be built in to the SunAmp

 an OEM 3kW immersion in the 'e' series SA units ).

 

That means there is no specific need for an external willis heater with a wet cylinder / SA based installation. You only need to fit one if you chose no storage device for space heating, a la @TerryE, or as failsafe with an ASHP. FYI an Uniq eDual PCM58 SA unit will drive both your heating and hot water from one unit, as it has 2 hydraulically separate heat exchangers, both in 22mm pipework. Driven from E7 or more suitably E10, IMO, that is the neatest and most efficient way to go, plus no G3. Adding PV will offset the amount of electricity you import. Simples.

 

10 hours ago, Dreadnaught said:

But with a passive house, I am not sure the slab heat-buffer is needed. A passive house needs so little heating that even direct heating at night using electric mat UFH of a few kW might be sufficient to maintain a constant temperature throughout the day/night cycle even in the depths of winter.

What other medium is going to give you storage of heat energy? Certainly not a P5 floor with electric UFH on top. Also with an electric system there is no modulation, so its either on 100% or off 100%, whereas with a slab or even wet UFH in spreader plates you can modulate the flow temp to match precisely, and just offset, the immediate heat losses.  

I think you can get away with out the slab as you have excellent depth of cellulose under the heated floor, but id be wary about over / under shoot with something that you cannot modulate. At best you may be able to get an electric UFH system that gives 50w/m2, but that will roast you if left on too long. Having a very accurate room thermostat would suffice but you'd probably need them local to each zone, and multiples of them ( stats and zones ).    

Edited August 27, 2018 by Nickfromwales

[Dreadnaught]

@Nickfromwales, wow, thank you for your comprehensive reply. That was very helpful indeed. I am learning fast. Quite a lot of that was known to me but not all.

 

42 minutes ago, Nickfromwales said:

Does it have a TPO on it? 

 

As its in a conservation zone, effectively yes (or at least the slap a TPO on the moment you reach for the saw as your required to give advance notice).

 

42 minutes ago, Nickfromwales said:

no

 

Thank you. Two people have now shot down the idea of electric UFH. You plus one on private message too. So I am moving on from that idea.

 

The next idea is to use a MVHR post heater for space heating. And, with a unit like the Zehnder ComfoCool, for cooling too.  

• Airflow rates with MVHR is relatively low so heating (or cooling) capacity is also low
• In mid winter, if more heating is required then the air flow of the MVHR would also need to be increased, with possible problems of noise (whistling) and low humidity (possible to be remedied with a enthalpy heat-exchanger)
• I think that using the ducting for heating and cooling will require for all the supply ducts to be insulated
• Duct heating can also lead to a burnt-dust smell

Any thoughts? Ever seen such a solution used, Nick?

Edited August 27, 2018 by Dreadnaught

[Jeremy Harris]

The air flow rate at normal MVHR background ventilation rates is low, typically no more than one whole house air change every couple of hours.  We have a 1.5 kW duct heater/cooler build in to our Genvex.  Yes it is just about enough to heat the house, and provide a bit of comfort cooling in hot weather, but it's extremely slow to change the house temperature and left to it's own devices the Genvex controller will just put the MVHR into boost mode in order to try and circulate as much heated/cooled air as possible. 

 

It does work, and because the house has a long thermal time constant we could get away with using it as a gentle background heating system, but I find the air a bit dry when it's in heating mode and downstairs there's a very significant benefit from having the UFH heating on.  Somehow, having very slightly warmer feet adds to the perceived comfort level a fair bit.

[pdf27]

Duct heating has the virtues of being cheap and simple (=reliable) compared to just about anything else, but is never going to give quite as good control as something like a wet underfloor system which has lots of heat capacity and will inherently spread the heat around the whole house more evenly than something which can only supply heat over the MVHR ducts. It depends what your priorities are and how accurately you feel the need to control temperatures.

 

One other option to consider would be mini-split air conditioners. They're very popular in the USA for Passive Houses (the US has their own Passive House institute that uses a different spelling), typically with one or two head units. Again, cheap and can provide cooling, and should be pretty reliable but depending on your house design shifting the heat around may be a challenge.

 

Finally, don't assume that because you don't have a slab you can't have underfloor heating in a screed - plenty of systems out there that use a screed on timber floors, and you could even make it fairly thick to give quite a bit of heat capacity if you wanted to. The only real limits are the built up height of the floor (which might be getting quite big with insulation and 100mm of screed!) and the strength of the floor joists.

[Gone West]

6 hours ago, Nickfromwales said:

 @PeterStarck has a similar arrangement in a much smaller dwelling and is surviving with that, but admits he keeps his space heating temp quite high ( ~24^oC iirc ) to ensure a constant ambient and again iirc he also fortifies that with auxiliary ( supplemental ) heating.
 

'Surviving' sounds as though it's a struggle whereas it seems to be a simple and effective solution in my house. The reason we keep the internal temperature high is because that is what Wendy likes. The primary heating is via the electric towel rails in the three bathrooms.  In our view the rooms you want warmest are the bathrooms so it's an effective way of achieving that and then extracting from those rooms via the MVHR. The EASHP in the Genvex Combi will increase the supply heat to the other rooms if required. So primary heating is from the towel rails.

The problem with quoting PH alone is that it is a maximum requirement for space heating of 15kWh/m2/yr which may result in a whole house heating system being needed but when the specific requirement is significantly less that 15kWh/m2/yr there are novel ways of achieving that without wet UFH.

Decrement delay is mentioned a lot but it is not simple to model and how important it is to achieving a comfortable heating/cooling balance is going to be dependent upon the wall and roof construction, shading and weather. What are the effects of ventilated spaced surfaces within the wall construction or having the house shaded by large trees all day or living where it is always windy. Not so simple .

[Nickfromwales]

1 hour ago, pdf27 said:

Finally, don't assume that because you don't have a slab you can't have underfloor heating in a screed - plenty of systems out there that use a screed on timber floors, and you could even make it fairly thick to give quite a bit of heat capacity if you wanted to. The only real limits are the built up height of the floor (which might be getting quite big with insulation and 100mm of screed!) and the strength of the floor joists.

My point above, and with a chunk of decent decrement delay insulation below it you can also likely get away without the screed altogether. The remit is to be able to not exceed the required heat input and get the hysteresis to where there is little or no over / undershoot. This is VERY easy to achieve with a wet blended system utilising pre-blended ( at source ) low grade flow temps at or near to the required loop temps.  

IMO adding a thin screed is more of an annoyance than a benefit, as once you accept that you'll still need a constant low temp flow even with the thin screed, then why would you bother ? More expense beefing up the structure, more labour and materials, and your also then conducting heat downward into the substrate which again I think is pointless. 

IMO its slab or no slab, and if no slab its plates. 

[Onoff]

I could quite happily "survive" in Peter's house!

 

It's the free from draughts feel and overall quietness does it for me.

[Dreadnaught]

@Nickfromwales, would you ever design electric UFH in bathrooms for its higher temperature and responsiveness in comparison to wet UFH?

[Nickfromwales]

26 minutes ago, Dreadnaught said:

@Nickfromwales, would you ever design electric UFH in bathrooms for its higher temperature and responsiveness in comparison to wet UFH?

Every time tbh. If however this is specific to a single story dwelling then the maths all change, and then fitting wet elsewhere and electric in the bathrooms becomes madness.

To get more heat out in a particular space but with a unified flow temp you just put more pipe in per m2, so you can still get 'higher temps' eg a warmer bathroom compared to the living room by engineering it to be so, but also you could open the loop up and feed into a big towel rad from the same loop thus increasing the size of emitter in that one space. 

Re responsiveness, in comparison, if you have wet UFH in plates you'll not really be waiting that much longer TBH. You'll just be heating the P5 up first before the room, plus yopull have a wider choice of floor coverings as electric in bathrooms can only go under certain flooring types. If you want a thin covering you would have to bury the electric heating wire in at least 5mm of SL compound to comply with the manufacturers installation instructions. 

All depends if you plan to store PV derived energy for use at 'night' or whether you intend to use E10. If the latter then you may just want to go for electric for the bathrooms, or go wet everywhere with a Willis heater.

You'll need to store DHW ( as instant electric heaters are dire on a good day ) so best to consider what your going to do for that and then expand from there. The SA for DHW only is a neat solution, so you could take a page out of @TerryE's book and go Willis + SA. Next chunkier option is Willis + UVC which would be slightly cheaper capital cost ( but suffer long term with G3 ).

I guess a few questions need answering.

Is the tree going bye-bye?

Will you then fit PV?

If not, and no, then the last option seems the right one, using whichever DHW solution best suits your remit.  

[Nickfromwales]

1 hour ago, PeterStarck said:

'Surviving' sounds as though it's a struggle whereas it seems to be a simple and effective solution in my house. The reason we keep the internal temperature high is because that is what Wendy likes.

Yes, sorry I could have worded that differently. It was more about my underlying scepticism that the house may not be so comfortable ( perceived ) if the ambient was sat at a more typical 22^oC which would be a full 2^oC cooler. I just wonder what the differential between temp underfoot vs temps at head height when stood up would be like. I fear the difference would be significant as I cannot help repeating to myself that heat rises.

 

With airflow being the key medium for delivering the heat I have reservations about the performance / ambient temps being equal and maintained in rooms which don't have emitters in them, but that is more specific to larger or 2-storey dwellings. I do know that the Genevex does rely on increased airflow rates to achieve their target room temps, how that translates into audible airflow I've no idea. Based on that I've suggested for my aforementioned project that the client instruct the MVHR installers to put low level inlets ( eg 2 per room in the opposite corners to the door, in the few problematic rooms that they say they cannot guarantee ever getting over 18.5^oC, without auxiliary heating, even though other rooms will regularly achieve 20.5^oC with the same shared 'flow' temps. The MVHR supplier suggested at-inlet inline duct heaters, but I think having a warm draught in my bedroom would be less than favourable. Admittedly there are few test cases for this type of installation and I appreciate all the feedback that @PeterStarck has provided. 

He may just get that pint ( of scotch ) after all.  

[Dreadnaught]

Thank you very much @Nickfromwales. That makes sense.

 

23 minutes ago, Nickfromwales said:

I guess a few questions need answering.

Is the tree going bye-bye?

Will you then fit PV?

 

Spot on, Nick. I will be working in the coming months on the fate of that tree, be in no doubt. If the tree can be removed then I shall indeed be embracing PV.

[Gone West]

24 minutes ago, Nickfromwales said:

Yes, sorry I could have worded that differently. It was more about my underlying scepticism that the house may not be so comfortable ( perceived ) if the ambient was sat at a more typical 22^oC which would be a full 2^oC cooler. I just wonder what the differential between temp underfoot vs temps at head height when stood up would be like. I fear the difference would be significant as I cannot help repeating to myself that heat rises.

 

If there is a greater differential between floor and head height as there would be with a 24C temperature then it would be more noticeable. I'm sure you asked me to measure floor and air temperatures and IIRC the difference was very small. As heat rises, with UFH you should have a greater differential, but you won't unless there is high temperature water in the UFH.

[Gone West]

@Nickfromwales

I’ve been thinking about your comments on PH space heating in relation to floor temperatures. A PH isn’t just a conventional house that doesn’t require much heating. In the short time we’ve been living in our new build  I’ve come to realise just how true all aspects of PH are that we had to consider when we designed the house. In our house, walls and floor as well as all the other surfaces, are at much the same temperature as the air. Even the glazing is similar and it is that similar temperature which creates the comfortable feel in the house. If the air is at 20C, 22C or 24C the surfaces including the floor will be at a very similar temperature.

If you create significant convection currents you start to lose the feel of comfort. UFH creates convection currents so in order reduce the negative effects of convection the floor temperature is kept at only slightly above air temperature. When we designed our house we considered UFH but we wanted a ‘dry’ house, a house without water based heating. We found from modelling our house in PHPP that is was feasible to have a warm house without conventional heating. If our house had been larger we could not have used our system. In the house design in which you are currently involved it seems that there isn’t a suitable product which will provide the level of warm air heating required due to the house size.

With the PH concept I like the idea of moving away from traditional wet systems with large tanks, boilers and UFH and with Sunamp, Genvex Combi etc it should be possible to design novel systems.

[Onoff]

Doesn't MVHR itself "create" convection currents?