General: Decrement delay

61 to 90 of 104

- CommentAuthorfostertom
- CommentTimeFeb 3rd 2009
Posted By: djh Posted By: James NortonThe Viking site has a formula but doesn't seem to tell you what all the variables are

Lambda is k :) - thermal conductivity

rho is density

c is specific heat capacity
What's a? (Just above, Temperature guide number = a2/m)
- CommentAuthorfostertom
- CommentTimeFeb 3rd 2009 edited
Posted By: marktimeDecrement delay seems to be an artifact of mass and I have seen another a reference to a mass equivalence effect where the the internal temperature reflects a higher external temperature than is actually the case. I have one reference which I am unable to follow up, a formula given by an Italian author, Givoni where:

delta T ext. = 0.01 . M eqiv. (unit mass in Kg/m2)
Is this him? http://books.google.co.uk/books?id=rJsVoRw1geoC

Can you re-explain the above, and what its significance is? (sounds interesting)
- CommentAuthorfostertom
- CommentTimeFeb 3rd 2009
Posted By: Paul TThe sense of touch - hot an cold - is not as dependent on temperature as you might think. What we are actualy sensing is a rate of temperature change in our nerve cells. So two materials witht he same temperature can fee very different. So it is quite possible for a 'warm' wall (say 20) to feel cold compared to a cold piece of insulation. This is one of the reasons why the actual surface material is the most important layer of all.

- Wood cladding is possibly the best surface material you can chose:
It has a low conductivity (warm sup quickly)
It has a lower emissivity than most surfaces (0.75), reflecting more heat back.

If it's the 'actual surface material' that's important, how about wallpaper - just thin wood! Actually it has to be a bit more substantial than that, to work as a warm-feeling surface - how thick?
Say we had the ideal medium density wall material - say Ytong aerated conc block, as claimed http://www.viking-house.ie/downloads/Ytong%20Dec%20Delay.pdf (presumably Thermalite etc too), themn plasdtered it with something cold-feeling like gypsum, much comfort would be lost? What warm-feeling alternative? clay plaster? lime plaster? or just wall paper?
- CommentAuthorralphd
- CommentTimeFeb 4th 2009
Blown fiberglass is worse than cellulose mainly because of convection cells, not because of the lower thermal mass.
http://www.ornl.gov/info/ornlreview/rev26-2/text/usemain.html

Here's a site in Alaska that clearly shows fiberglass insulating less than cellulose
http://www.gwscientific.com/cchrc/rtf_data/roof/roof.html
(the top of the fiberglass is much warmer than the cellulose)
- CommentAuthordjh
- CommentTimeFeb 4th 2009
Posted By: Viking HouseHi djh

I actually believe that house B will use less fuel to heat the house than house A ... If they are building a house for a client the heating cost for the house is reduced dramatically when they choose to Externally Insulate with Dense Paroc Rockwool Insulation over Polysterene. ... If it is really Cold outside you will choose your heaviest coat which would have the same U-value (thickness) as your lighter coat if tested.

When they started studying heat loss with 3D Thermal Imaging cameras in Sweden they discovered that heavy insulation worked better and that heat moving in cylindrical patterns through dense insulation cannot be detected by a hotbox test.

Hi VH,

Please don't take this the wrong way. I've already said that I believe there is a place for heavy insulation, thermal mass & decrement delay. But the subject can be somewhat complicated and confusing so I believe it is important to understand exactly what works in what circumstances and why. So I'd like to add my voice to the requests for evidence - ideally independent peer-reviewed papers - for any of the effects you've mentioned above.

As regards my clothes, I'm afraid you're exactly wrong. I do have a wardrobe with many woollen jerseys including heavy fishermens but I no longer wear them. I have a lovely thick and heavy wool coat but I only wear that when I go to important meetings. In real life I wear layers. When it's cold that's polypropylene next to the skin, one or two fleeces and a lightweight windproof outer. Fleece-lined nylon trousers. All as light as I can get them. As much insulation as possible (i.e. low U-value) and an airtight barrier!

I do have a pair of trousers that claim to buffer heat using phase-change material - probably the closest thing to the form of construction we're discussing. They're pretty useless in the winter and too hot in the summer but they're good in the swing temperatures of spring and autumn.

Basically, the underlying physics of clothes and buildings are the same but the requirements and engineering are completely different.
- CommentAuthorfostertom
- CommentTimeFeb 4th 2009 edited
VH, I also really like your approach and contribution here, and I'm rootin' for yer, to see whether we can get good explanation for, and understanding of, your experience of all-season benefit of dense insulations e.g.

'The findings of the TNO indicate clearly that simulation calculation and temperature behaviour of the roof are in practice comparable, and confirm the advantages of dense insulation for summer and winter thermal protection' (did you say you were going to seek translation from Dutch?)

The Scandinavians I have spoken to speak of building houses with Cold Materials and Warm Materials and the fact that the house feels Colder by a few degrees when you build with Cold Materials.

I actually believe that house B will use less fuel to heat the house than house A which was one of the reasons for putting up the webpage. Many of my Polish builders believe this instinctivly and have seen it in practice. If they are building a house for a client the heating cost for the house is reduced dramatically when they choose to Externally Insulate with Dense Paroc Rockwool Insulation over Polysterene. The U-value of the wall doesn't change just the density of the insulation.

I feel the story's not over yet - there's something there!
- CommentAuthorfostertom
- CommentTimeFeb 4th 2009 edited
Posted By: Viking HouseExternally Insulate with Dense Paroc Rockwool Insulation over Polysterene
This sounds in principle like
Posted By: tomsuswebour walls from outside in are - Timber cladding, air gap, 60mm softboard (tongue and grooved to help with air tightness), 150mm stud frame with cellulose insulation, 11mm OSB (racking and vapour control layer), 40mm softboard (internal humidity buffer, extra insulation and improved acoustics), plasterboard
Posted By: tomsusweb
Posted By: PaulTHowever Cellulose materials are not High thermal mass - they are medium thermal mass. THis means that you have a medium amount of thermal mass available with a medium amount of thermal resistance. This means that the wall can asborb a relatively significant amount of heat at a slow rate. The home will quite repsonsive to heating and have slow release quantity of thermal mass.
That's how I see it too. And I'd say that's what makes it work well for us in the Irish climate - we need a reasonably responsive house to heating because the weather changes so often but also need a bit of slow release to even out the temperature swings
What about this idea - lightweight insulation core (low k, so much U-value in small thickness) faced one or both sides with dense insulation?
- CommentAuthorfostertom
- CommentTimeFeb 4th 2009
As well as decrement, other concepts in this area are admittance, diffusivity and effusivity. We need to understand these too.
- CommentAuthorViking House
- CommentTimeFeb 5th 2009 edited
My friend stays in his log cabin in Sweden 500 miles north of Stockholm for a few weeks every Winter. He often spoke to me about how the cabin works to Cold weather! He heats the house with a log stove which tends to go out about 2 O'Clock at night. Often it gets as Cold as minus 20-25 degrees there at night but the Log Cabin loses heat relatively slowly and is typically 10-12 degrees in the morning, 15-20 degrees warmer than outside. The first thing he does in the morning by the way is light the stove again. While he was telling me this story I was thinking of the winter nights I spent in a caravan in Ireland with a gas heater. The caravan would reach the external temperature of maybe 8-10 degrees in under an hour after the stove was put out. The caravan has, when you calculate it, a better U-value than the Log cabin so why is there such a big difference in heat loss?
I know again that this is again not Scientific but I want to put these examples out there to get people thinking.

This same friend studied Architecture in Sweden and one of his teachers was Bo Adamson who started the Passive House movement, so he was learning about energy when I was only a chap. My friend has asked some of the Energy Professors in Stockholm University about this whole Decrement Delay thing and they agree that it is a factor in heatloss. The problem is that all the information on this is in either German or in a Scandinavian language.

They use a layer of Softboard on 65% of the new roofs in Switzerland, why is this if its only for U-value? there are much cheaper ways to get a good U-value!

I will email Prof Michael Ham from TNO University tomorrow to root out the Paper on this, I had no time to do it today!
- CommentAuthordjh
- CommentTimeFeb 8th 2009
Posted By: fostertomWhat's a? (Just above, Temperature guide number = a2/m)

Good question, Tom! And nobody's answered it. In the formula with lambda, c and rho it's "thermal diffusivity" (see e.g. http://en.wikipedia.org/wiki/Thermal_diffusivity ). As "Temperature guide number" in the table of "weight and performance of various construction materials", I have no idea! The one thing I am sure of is that "a" there doesn't mean the same as "a" below and that neither are defined. And google's never heard of temperature guide number. So that's another one for VikingHouse to answer, I think ...

There's also a similar concept called thermal effusivity - http://en.wikipedia.org/wiki/Thermal_effusivity - which looks similar but is what determines whether something feels hot or cold, I think.

This page - http://www.evitherm.org/default.asp?ID=277 - might help resolve confusion.

Peter, that link on "Thermal-Mass-and-R-Value" is excellent. You can find the original article at http://www.buildinggreen.com/auth/article.cfm/1998/4/1/Thermal-Mass-and-R-value-Making-Sense-of-a-Confusing-Issu and as you say, it points out some limitations of this concept.

Since we haven't had any of the references we were asking for, I went googling. I found one interesting link that says the whole thing is hokum - http://www.bing.org/DOWNLOAD/DOCS/BING_FACTSHEET10_Efficient%20insulation%20-%20the%20key%20to%20effective%20thermal%20protection%20in%20summer.pdf - oddly and I'm sure coincidently it's connected to PUR insulation manufacturers.

I also found something that's probably more independent - http://www.ornl.gov/sci/roofs+walls/research/detailed_papers/effects/influence.htm - but I haven't studied it in detail yet. It's basically comparing the effectiveness of using various layers of thermal mass and insulation in a wall. As I understand it, the point is that the heaviness of the insulation is not paramount. It's the structure of the whole wall that matters.

Cheers, Dave
- CommentAuthorBrianR
- CommentTimeFeb 8th 2009
I looked at the explanation of decrement delay on the viking web site.

Decrement delay appears to be the delay from the peak of outside temperature to the peak of inside temperature assuming fluctuations in the 24hr temp a bit like a sine wave.

The decrement delay then comes from the phase delay caused by the integration effect of heating a material with high thermal mass. So a heavy church wall retains its heat from the day into the evening. By having a series of insulation and then thermal mass elements you could get any thermal delay you wanted 6 hours 12 hours or even 18 hours.

Not sure how useful such an effect is as I would expect the ideal to be very good insulation on the outside of a church wall causing the wall to take weeks to heat up.
- CommentAuthormarktime
- CommentTimeFeb 8th 2009
Thanks to fostertom for the heads up on Baruch Givoni. Mass equivalence seems to be the subject of Peter's link, i.e. useful when exterior temps fluctuate around internal set point.
- CommentAuthorali.gill
- CommentTimeFeb 27th 2009
http://www.gutex.de/en/index_prod.html

Click on the yellow sun icon next to 'Dealers' for a brief overview of decrement delay and some useful, though not referenced graphs, showing lag and also comparing different materials.

Interestingly the diffusion page (blue icon) states "Necessary: The water vapour diffusion resistance must be lower towards the interior and higher towards the exterior"
- I may be wrong but i thought another thread touched on a rule of 5x more resistance from interior to exterior_ i.e the opposite of what it says above.
- CommentAuthormike7
- CommentTimeFeb 27th 2009 edited
Posted By: djh
Posted By: fostertomWhat's a? (Just above, Temperature guide number = a2/m)

Good question, Tom! And nobody's answered it. In the formula with lambda, c and rho it's "thermal diffusivity" (see e.g. http://en.wikipedia.org/wiki/Thermal_diffusivity ). As "Temperature guide number" in the table of "weight and performance of various construction materials", I have no idea! The one thing I am sure of is that "a" there doesn't mean the same as "a" below and that neither are defined. And google's never heard of temperature guide number. So that's another one for VikingHouse to answer, I think ...

I'm not a viking but I'm sure 'a' is the diffusivity as defined by the formula a = lamda over ( c times rho) as was given, and to be consistent with the other data would be expressed in units metres squared per second. In the table of material properties, 'a' is still diffusivity, but is expressed in units of cm squared per hour and has magically been renamed 'temperature guide number' . The 'a*2/m' heading is thus a misprint and should read :- a [cm*2/hr]

It all checks out across the table with the values given, except 'a' given for Paroc and Hemp is wrong. Does this make me geek of the week?
- CommentAuthorGotanewlife
- CommentTimeFeb 27th 2009
Mike7 - I love/hate people like you!
- CommentAuthorPeter Clark
- CommentTimeMar 12th 2009
Posted By: Peter ClarkHow likely is it that an existing roof or dormer structure, with no insulation, or rockwool, will be able to bear the extra weight of pavatherm or something similar?

Pavatherm has a density of 160Kg/m3

so 8 Kg for a m2 at a thickness of 50?

Can anyone enlighten me?

I recently talked to a structural engineer about this, informally. He suggested that most roofs, including mine with concrete tiles, are carrying a load about 10 times the 8Kg/m2 for Pavatherm. he suggested that there was nothing to worry about putting that load on the existing dwarf attic walls and dormer cheeks, didn't mention the dormer roof explicitly, but he seemed to imply "go for it".

Does anyone have any advice?

Peter
- CommentAuthorsinnerboy
- CommentTimeDec 14th 2009 edited
I stumbled accross this http://ecotect.com/node/480 - it contains 2 methods to calculate decrement delay

But i'm stuck - I have entered the formulas on the attched excell sheet to try to compare 300mm of Fibreglass vs Cellulose . For the purpose of the exercise - same thickness , same lambda , different heat capacity .

But the 2 formulas give differrent results . would not be the first time a mis typed of formula in excell - so

1. help guys
2. think this is useful ?
- Dec delay.zip
- CommentAuthorjon
- CommentTimeDec 14th 2009 edited
Are we talking about the effect of the thermal mass of the insulation itself?
(as in thermal mass, whatever it is, slows down the fluctuation rate of an insulated body)
If so, why call it decrement delay: Doesn't this just confuse everyone?

I don't really see the point: If you need thermal mass then it's better to go for the best insulation you can and add thermal mass (or even 'heavy' insulation on the inner surface): That way, the energy storage has double the life of whatever storage you would get from the same 'mass' of insulation.

Unless I've missed something; frequently happens.
- CommentAuthordavidfreeborough
- CommentTimeDec 14th 2009
It's often cited, though I don't have a reference, that only the 100mm of concrete closest to a habited space can usefully contribute to thermal mass & thereby decrement delay. Presumeably, this is because the thermal conductivity of the concrete (k=1.4) is sufficiently low to decouple the more distant concrete from the air in the room.

For insulation with a 35 times lower thermal conductivity (k=0.04) the layer which is effective in thermal mass terms must be proportionately thinner? 3mm? Surely it would be more effective to insulate to meet your U value targets & line with something thermally massive?

David
- CommentAuthorsinnerboy
- CommentTimeDec 14th 2009 edited
I was hoping someone would open the excell file and see if there are any errors in the formula cells . I have tried and fear "not seeing wood for trees" has set in.

The formula IS narrow - intended to assign a thermal lag to the insulation layer only - academic I know . The problem I have now is that 2 different formulas are indicating different time lags so I am doubting the entries I made in the excell file .
- CommentAuthorfostertom
- CommentTimeDec 14th 2009
In latest Winter 09 Green Building mag, Gavin Harper's begun a series of much-needed articles about building physics, beginning with Moisture and Humidity - based around the Psychrometric chart.

Wd be fantastic if the series rolled on to eventually cover Decrement and all that.

Tho it takes me back to those 1967 classes, the Architecture students jointly with the Building Technology students (the likes of John Willoughby), at Bristol Coll of Advanced Technology (became Bath Univ), I regret I can't remember it all and have sadly lost my notes and text books. What a treasure those would be now! The knowledge and understanding was there, but the building industry was completely ignorant of Decrement etc, sunk without trace for 35yrs until now we're scrabbling to understand it, which is the key to new modes of thermal performance that go far beyond the simplistic limitations of U-value alone.
- CommentAuthordjh
- CommentTimeDec 15th 2009 edited
Posted By: jonAre we talking about the effect of the thermal mass of the insulation itself?
(as in thermal mass, whatever it is, slows down the fluctuation rate of an insulated body)
If so, why call it decrement delay: Doesn't this just confuse everyone?

It's another name for thermal diffusivity http://en.wikipedia.org/wiki/Thermal_diffusivity

It's typically used when talking about the structure of a wall made up of layers of insulation (light or heavy), structural materials, facing materials etc and is usually expressed as the number of hours that heat is delayed travelling through the wall.

I don't really see the point: If you need thermal mass then it's better to go for the best insulation you can and add thermal mass (or even 'heavy' insulation on the inner surface): That way, the energy storage has double the life of whatever storage you would get from the same 'mass' of insulation.

Doesn't always work out that way, apparently. There's an interesting web page by one of the US government labs that explores various different layerings of insulation and mass - http://www.ornl.gov/sci/roofs+walls/research/detailed_papers/effects/influence.htm
It also strongly depends on what climate you're in.

Posted By: davidfreeboroughIt's often cited, though I don't have a reference, that only the 100mm of concrete closest to a habited space can usefully contribute to thermal mass & thereby decrement delay. Presumeably, this is because the thermal conductivity of the concrete (k=1.4) is sufficiently low to decouple the more distant concrete from the air in the room.

Indeed, the 100 mm only for heat that is absorbed on the surface of the concrete on the inside of the building and only when considering diurnal changes. If you're using masses of concrete and the time constant is longer (e.g. Hockerton) then the heat has time to travel deeper.

When talking about decrement delay, it's pretty much always about how heat absorbed on the outside of the building is transmitted to the inside of the building, so the whole depth of the fabric is always involved.

For insulation with a 35 times lower thermal conductivity (k=0.04) the layer which is effective in thermal mass terms must be proportionately thinner? 3mm? Surely it would be more effective to insulate to meet your U value targets & line with something thermally massive?

It all depends what you're trying to achieve!

sinnerboy - I'll take a look at your spreadsheet but it'll take longer.
- CommentAuthorsinnerboy
- CommentTimeDec 15th 2009 edited
Agreed Tom - the lab environment employed to arrive at a lambda value fails to take account of real life environments where a materials air permeance , density and specific heat capacity all impact on the performance of the building fabric . The formulas I posted seem to give a method to calculate a value taking into account density and specific heat capacity ... together with lambda so still an imperfect calculation .

But as a starting point I would love to get the 2 formulas to agree .They formulas are to be found here http://ecotect.com/node/480 and for clarity I cut and paste them here now

1st one
Thermal lag = 1.382*thickness * sqr(density*specific heat / 3.6*thermal conductivity)

2nd one
tl = 1.38 * L * SQRT(1/a)

where:
tl = time lag (in hours)
L = Thickness of component (unit?)
a = thermal diffusivity, = thermal conductivity / (density * specific heat)

Both formulas begin by muliplying the thickness by 1.38 .

The first formula then seeks the square root of - density x specific heat divided by 3.6 x conductivity

The second formula seeks the square root of of the inverse of conductivity divided by density x specific heat

Any mathematically capable posters care to shed light ?

Edit - thanks djh - we were typing at the same time it seems so i had not seen your post when I was writing this
- CommentAuthorsinnerboy
- CommentTimeDec 15th 2009
I found this link on the subject

http://www.selfbuild-central.co.uk/green-design-overview/saving-energy/insulation-properties/decrement-factor/

( Note it links to the VH site - so full circle - well done S . )
- CommentAuthorjon
- CommentTimeDec 15th 2009
Thanks djh

But the references seem to just say that we are talking about the effect of the thermal mass of the insulation itself?

So I can't see how it would be possible (in a physics sense) for thermal density within a wall to beat black body internal equivalents except where the black body surface is significantly smaller than the insulation surface (ie less than 50%); even then it's about short term effects: Over longer term cycles (say a week or a season) the internal thermal mass must, by definition, always win.

Have I missed something?
- CommentAuthordavidfreeborough
- CommentTimeDec 15th 2009 edited
Jon

Yes, it's about short term cycles & minimising the day/night temperature variation. If each infinitely thin piece of insulation has both thermal resistance and thermal capacity then I guess we need to employ a bit of calculus to work out the overall time constant.

I've not studied thermo-dynamics, so for me it's easier to think of this in terms of electronics with resistances charging capacitors, but I'm not sure that would help everyone! So, in other words, the thermal capacity of each layer is "charged" by the thermal resistance of the layers to it's outside. But the "charging" temperatue seen by a layer is "slugged" by the thermal capacity of the layers to it's outside. So I assume you need to integrate the effect of each layer to work out the overall time constant.

David
- CommentAuthorjon
- CommentTimeDec 15th 2009
A good argument for the idea of putting tiles between floor joists?

Or for using concrete instead of timber for frames
- CommentAuthordjh
- CommentTimeDec 16th 2009
sinnerboy, sorry I haven't got around to doing much with your formulae yet. But as you show above, they are basically the same formula except one divides the result by sqrt(3.6) ~ 1.9

That suggests to me that one has different units to the other. They look like roughly what I'd expect but I'm not familiar enough to instantly tell what units differ. I suggest trying to find out what units were quoted in your sources.

There's probably an answer in chapter 5 of this online book - http://web.mit.edu/lienhard/www/ahtt.html - but I haven't got time to look right now.

Cheers, Dave

PS David, yes I use the electrical analogy to think about it too. It's an RC ladder filter, or in the continuous case, a waveguide, I think.
- CommentAuthorfostertom
- CommentTimeDec 16th 2009
Great book - wish I could remember my Alevel maths. On p19,

thermal diffusivity (aka decrement?) is proportional to conductivity divided by volumetric heat capacity.

A classic high-decrement material, with high conductivity and high volumetric heat capacity, like concrete, may have similar decrement to a material with medium conductivity and medium volumetric heat capacity, like hemcrete, or aerated concrete block.
That's the special genius of hemcrete or aerated concrete block - high decrement coupled with quite decent insulation.
If we take conductivity further down the scale, to the level of a conventional insulation material, such materials' volumetric heat capacity doesn't fall proportionately, so their decrement is low.
- CommentAuthorjules
- CommentTimeDec 17th 2009
Sorry to divert this discussion with a very dumb question, but does all this mean - assuming perfect installation with no gaps etc - that 100mm solid insulation with a U value of say 0.2 is better than 200mm mineral wool with a U-value of 0.2?

It so happens that in my loft the bottom 75mm between the joists is solid insulation, with wool on top and then across the joists (cos that's what i happened to have to hand).

61 to 90 of 104



	Site Map \| Home \| View Cart \| Pressroom \| Business \| Links	Contact Us
Logout
© Green Building Press

Green Building Forum

Categories

Feeds

General: Decrement delay