Jump to content

The Great Thermal Mass Myth................


Jeremy Harris

Recommended Posts

39 minutes ago, ProDave said:

No living in a static caravan just illustrates how poor a building can be if you only use a VERY small amount of insulation, and that small layer has a fast decrement delay to compound the issue.

 

Also a caravan is very well ventilated to stop condensation so is nothing like airtight (which is a factor in well built houses)

 

note, I lived in a caravan for most of two years during my build.

 

My new build is a “heavy” build (a term I have used since reading about therxxx mxxx! ) mainly because we wanted a brick finish and solid walls and we are very pleased with the constant temps in both cold and warm weather.

Link to comment
Share on other sites

Unlike mass which has a very precise definition (and nothing to do with heat, BTW) , if you stick an adjective thermal in front of it which means "relating to heat" have "thermal mass" a bit of an oxymoron which is a concept and without any accepted definition and can be overloaded pretty much with whatever message you want.  Take the PDF that Ian originally referenced: it was authored by the Concrete Centre and has one dominant message running through it: use lots of concrete in your construction.

 

I've modelled the thermal performance of our house in some detail.   A large part of our slab is the load bearing ring and cross beams and actually plays little active roll in stabilising the house temperature; the floor slab itself does.  However, the entire plasterboard throughout the fabric of the interior has just as large an effect and in terms of the overall decrement delay performance of the house by far the dominant component is the cellulosic filler used as insulation.  This is because this time constant is directly related to the ratio thermal capacity / thermal conductivity, and unlike most wall fabrics which either have very poor thermal capacity or high thermal conductivity, this filler has pretty much the best combination to give good U-value performance coupled with high decrement delay.  The last factor in the stability of my house is that I have relatively small window areas and no "acres of south facing glass" so don't have to cope with the solar gain tiger.

 

So both my and Jeremy's houses have extremely low heat loss and high thermal stability, but these use a construction technique not even mentioned in the paper, and the only mention of timber frame construction is in a table on P14 where it is listed as having the worst decrement delay.  So the message is: don't use TF construction if you want thermally stable construction

 

This shows the bias of the authors, and in the case of a Larson strut construction with cellulosic infill this is quite simple a load of bollocks.

Edited by TerryE
  • Like 2
Link to comment
Share on other sites

5 hours ago, Ed Davies said:

13 hours is pretty short. Are you sure that's right? That would mean that if it was left unheated overnight it'd get down to about half the temperature it had relative to outdoors by morning.

 

 

I learned something there is that the formal definition of decrement delay? I thought @ProDave was saying he can barely detect any decline in temperature overnight.

 

Using the 50% fall definition the decrement delay of my static caravan is less than 3 hours.

Link to comment
Share on other sites

Decrement delay (in hours) is the time between the peak temperature on one surface of a wall or roof and the peak temperature on the other surface.  In general terms, it's the time taken for heat to travel through that part of the structure.

 

It's dependent on both the thermal conductivity of the structure, the thickness of the structure and the heat capacity of it.  A long decrement delay will be given by a thick wall (or roof) with a high thermal resistance (low U value) made from material with a high heat capacity. 

 

 

Link to comment
Share on other sites

16 minutes ago, epsilonGreedy said:

 

I learned something there is that the formal definition of decrement delay? I thought @ProDave was saying he can barely detect any decline in temperature overnight.

 

Using the 50% fall definition the decrement delay of my static caravan is less than 3 hours.

The thermal analysis of my wall build up suggests when the outside temperature peaks and starts to fall, the inside temperature of my house will START to fall 13 hours later.  It would be a VERY long time before it fell to 50%

 

And yes you barely notice the overnight drop in temperature when the heating is off.  

 

In the static 'van there was no concept of storing heat. you want to be warm, you need heating on.  A wood burning stove actually filled that role well

Link to comment
Share on other sites

32 minutes ago, TerryE said:

I've modelled the thermal performance of our house in some detail.   A large part of our slab is the load bearing ring and cross beams and actually plays little active roll in stabilising the house temperature; the floor slab itself does.  However, the entire plasterboard throughout the fabric of the interior has just as large an effect and in terms of the overall decrement delay performance...

 

 

Wow that is quite a revelation, I had idly assumed the floor slab was the dominant heat capacity component of timber frame.

 

Coupling your observation with the @JSHarriscalculation of 1.6kw per degree for the plasterboard and assuming the same for the floor, then a mid sized house should not leach more than 1kw of energy per degree delta overnight to prevent chilly mornings. 

Link to comment
Share on other sites

13 minutes ago, JSHarris said:

Decrement delay (in hours) is the time between the peak temperature on one surface of a wall or roof and the peak temperature on the other surface.  In general terms, it's the time taken for heat to travel through that part of the structure.

 

 

Ok. So when people here quote a whole-house decrement delay are they referring to average air temperature in a central port of the house?

Link to comment
Share on other sites

15 minutes ago, ProDave said:

The thermal analysis of my wall build up suggests when the outside temperature peaks and starts to fall, the inside temperature of my house will START to fall 13 hours later.  It would be a VERY long time before it fell to 50%

 

 

Ok that sounds like a modeled decrement delay. Now I am trying to picture how that is applied when the house is actively heated and the outside wall temp peaks at say 10 and the inside is maintained at 20 until the heating goes off at night.

Link to comment
Share on other sites

3 minutes ago, epsilonGreedy said:

 

Ok that sounds like a modeled decrement delay. Now I am trying to picture how that is applied when the house is actively heated and the outside wall temp peaks at say 10 and the inside is maintained at 20 until the heating goes off at night.

 

In my case the delay is 13 hours.  So if the coldest outside temperature was at say 4AM in the morning, then without any heating, the maximum heat loss from the house would be at 5PM.  Of course the heating would be on then so you would not notice the drop in temperature.

 

But you can begin to see why the long delay means you just don't notice an overnight drop in temperature when the heating is off.

Link to comment
Share on other sites

1 hour ago, ProDave said:

The thermal analysis of my wall build up suggests when the outside temperature peaks and starts to fall, the inside temperature of my house will START to fall 13 hours later.

 

This is a completely different thing from the thermal time constant of the house. See my explanation further up this thread. You could, in theory, have a house with a very long thermal time constant because it has a lot of heat capacity inside but very little decrement delay because the walls have very little heat capacity. Imagine blockwork internal walls with aerogel external walls.

 

(To be honest, I'm not at all convinced that decrement delay matters much in the UK climate most of the year for a well insulated house as the diurnal temperature variations just aren't that large. The only exception, really, is direct surface heating on a few exceptionally sunny days where the air temperature might not vary that much but the temperature of into-sun dark surfaces might. Maybe slightly more important in New Mexico or wherever.)

Link to comment
Share on other sites

7 hours ago, JSHarris said:

It can be worked out from first principles if the conditions are known, but one problem is that often the conditions aren't known accurately enough to make modelling that useful.  One way to try and measure it is to look at thermal admittance, which is a measure of the ability of a material to absorb and release heat into the interior of the house.  However, accurately assessing this is compounded by the way that we tend to have most of the internal structure in a house (at least the part that can act as a thermal buffer) made up of layers that have widely varying thermal conductivity and specific heat/heat capacity.

Yes, that's what I was thinking in respect of how much solar gain could be stored in the floor slab. I was thinking it would be very difficult to work out unless it is a more or less a homogeneous material. I have a concrete slab with a layer of tile adhesive then porcelain tiles. Presumably how reflective the surface of tiles is would also affect the result. Some of the flexible tile adhesives have air bubbles as well, so would act as an insulator. It would be interesting to know how much is being stored in the floor when the sun shines on it but I guess not really possible to work out, although maybe possible to measure.

Link to comment
Share on other sites

39 minutes ago, Ed Davies said:

(To be honest, I'm not at all convinced that decrement delay matters much in the UK climate most of the year for a well insulated house as the diurnal temperature variations just aren't that large. The only exception, really, is direct surface heating on a few exceptionally sunny days where the air temperature might not vary that much but the temperature of into-sun dark surfaces might. Maybe slightly more important in New Mexico or wherever.)

 

I was intrigued enough by this to pop out and measure the temperature of the outside wall of our house (just outside the front door, so facing SSE).  It's currently (at 17:40) sitting at 36.8°C, so about 15° warmer than the inside of the house.

 

 

Link to comment
Share on other sites

With a U-value of 0.1 W/m²·K but no decrement delay that 15 °C will give just under 15 W for a few hours a day to a 2.4 m-high room with a 4 m long external wall. That's not very significant compared with the hundreds of watts coming through any windows.

Edited by Ed Davies
Link to comment
Share on other sites

1 hour ago, JSHarris said:

I was intrigued enough by this to pop out and measure the temperature of the outside wall of our house (just outside the front door, so facing SSE).  It's currently (at 17:40) sitting at 36.8°C, so about 15° warmer than the inside of the house.

If you had a vented cavity behind your outer skin how would that affect the temperature of the main wall construction and hence decrement delay.

  • Like 1
Link to comment
Share on other sites

2 minutes ago, PeterStarck said:

If you had a vented cavity behind your outer skin how would that affect the temperature of the main wall construction and hence decrement delay.

 

We do have a vented cavity behind the skin, so that will reduce the heat transmission a fair bit, but I've no idea how to try and model it.  I could try and poke a temperature sensor into the cavity and see how warm it is in there, as that's probably closer to the temperature of the outside skin of the insulated wall itself.  I might have a go at doing this tomorrow, just to see how effective the ventilated space is at reducing inward heat transmission.

  • Like 1
Link to comment
Share on other sites

22 hours ago, epsilonGreedy said:

Wow that is quite a revelation, I had idly assumed the floor slab was the dominant heat capacity component of timber frame.

 

It depends on the room sizes etc.  We have a 3 floor house with fairly narrow (~3.3m) rooms and use 15mm plasterboard , so the volume of plasterboard is pretty close to that of the 100mm slab.  The specific heat capacities of plaster and concrete are pretty much identical.  

 

The decrement delay of a single-wall TF with PUR insulation is quite low, but that of a twin-wall TF with cellulosic filler is very high.  In our case we also have an external stone skin which also extends the overall DDF.   Ours is over a day, and what that means in practice is that we can ignore diurnal temperature variations is our heating calcs; daily average temperatures are easily good enough.

 

Our SE facing principle elevation does act as good external heat store  and as Jeremy says, it does heat up and hold this heat especially on these clear sunny mornings that we've been having.

Link to comment
Share on other sites

  • 3 months later...

Well, Architects are quite happy working in unitless and qualitative measures; see also visual weight of a design, degree of overbearing or overlooking on neighbours, harmony, speaking the local vernacular, feelings of space and light, etc.

To be honest the vast majority of the population are also quite tolerant unitless and ill defined measures. It's just in the more obscure corners of the internet you find the folks (some would even say, pedants) that get het up about it :)

 

  • Like 1
Link to comment
Share on other sites

34 minutes ago, scottishjohn said:

 

 

With the exception that what they refer to as "thermal mass" is really just another (spurious) name for heat capacity, and has no units of measurement that include a term for mass.

 

The easy way to distinguish whether a property is real or imaginary is just to ask what the units of measurement are.  For example, we can define power in watts, horsepower or probably a few other less commonly used units.  Likewise we can define energy in joules, kilowatt hours etc.

 

The only answer I've heard from any supporter of thermal mass has given the unit of measurement that supposedly applies to it that which has already been allocated for an entirely different property, heat capacity.  Heat capacity on its own isn't very useful though, as you also need to know how well heat can flow into, and out of, any element for to work out how much effect it may have on maintaining temperature stability. 

 

For example, the heat capacity of the phase change material inside my Sunamp is pretty high, yet because the case is highly insulated that stored heat contributes next to nothing to the temperature stability of the house.

 

In contrast, the gypsum plasterboard and plaster skim on all the internal walls has a moderate heat capacity (but less than that of the PCM in the Sunamp), but gypsum has a moderate thermal conductivity, plus there is a large area in contact with the air inside the house, so it has a massively greater impact on maintaining the thermal stability of the house.

 

With specific relation to that article, it refers to concrete for heat storage.  The heat capacity of concrete isn't that high in relation to its mass.  For example, 1 kg of concrete can store about 880 J/K, whereas 1 kg of water can store about 4220 J/K.  Even 1kg of plaster on the walls can store about 1090 J/k, so significantly more than concrete.

 

If someone wanted to improve the heat capacity of their house, the very simplest and cheapest way to do it would be to just double up on the thickness of plasterboard on the walls.  The combination of the reasonable heat capacity of gypsum, its moderate thermal conductivity and the relatively large area in contact with the air inside the house, means that this would be very much more effective at improving thermal stability than adding a bit more concrete to the floor.  It would also reduce sound transmission from room to room, too.

 

 

 

Link to comment
Share on other sites

stashing bottles full of water everywhere would be cheaper 

 

I like stable temperatures and heavyweight construction with the insulation on the outside.

 

I agree with Jeremy on the misuse of the term TM and wish to point out that water is almost unique having an anomalously high capacity to store heat. 

Link to comment
Share on other sites

What's interesting is that our house pretty much proves that mass doesn't determine the thermal time constant, as it's timber framed and timber clad externally, yet has a thermal time constant that is a fair bit longer than a diurnal cycle.  There's not much concrete in the house at all, just a 100mm thick concrete slab, laid on top of 300mm of EPS, and surrounded by 200mm of EPS around the periphery.  The pumped cellulose in the walls and roof helps a bit (300mm thick in the walls, 400mm thick in the roof), but not by that much, as there's a 50mm service void between the VCL and the plasterboard.  Pretty much all of the heat storage that helps to maintain a pretty constant temperature comes from the concrete slab and the plasterboard, in practice.  I guess a little bit also comes from the water in the UFH pipes...

 

 

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...