[Petrochemicals (OP)]

We are always told that air is a thermal insulator in such things as building insulation. In a cavity wall in a house though we place insulation of one sort of another that is a material aerated. Be it foamed polystyrene, blown glass fibre, wool, polyethylene, they all have one thing in common, material that is not particularly low in thermal conductance created into non continuous pieces interspersed with air. Yet air is the majority of the volume still.

Why not just stick with an empty cavity still?

[Bored chemist]

Convection.
(also advection but it's not usually the big effect)

[evan_au]

By having lots of small air cavities, there is not much temperature difference between one side of the cavity and another, so there is not much convection.
Many small cavities means that there are many of these low-convection obstacles to overcome before heat can enter (or leave) a home.

[alancalverd]

There is a critical channel width (about 1.5 cm) below which air convection is not sustained at normal ambient temperatures - this is the basis for double glazing. The usual brick cavity wall has a 10 cm cavity (the width of a brick) which pretty well prevents conduction and increases rigidity, but doesn't prevent convection.

[Petrochemicals (OP)]

Sorry i cannot believe convection, if air can convect and conduct in a gap, smaller gaps mean the same thing, especially if what breaks up the gaps is a solid with high density and a higher thermal capacity.

 It is just air with bits in the middle, air still has to draw the energy in the same way as covected air has to, it still has to deposit the energy in the same way a covection flow does.

I do understand if for example the cavity in a brick wall is uncapped, you have a chimney and a single skin of bricks, the air would be drawn in and expelled at the top, perhaps the insulation slows this effect. Many buildings with cavities where infact built with uncapped empty cavities for many decades.

[alancalverd]

If the gap is too narrow, you can't get laminar flow down one side and up the other - turbulent mixing interrupts the convection pattern. That's how gas-filled double glazing works, regardless of the belief system of the observer.

Not sure how air is "drawn in" to a cavity wall acting as a chimney. Mine are supposed to be wind- and water-proof all the way to the ground. 

As for the density and conductivity of the filler, polystyrene or polyurethane foam is about 96% air by volume - you can pretty much ignore the plastic content. 

[Petrochemicals (OP)]

If the gap is too narrow, you can't get laminar flow down one side and up the other - turbulent mixing interrupts the convection pattern. That's how gas-filled double glazing works, regardless of the belief system of the observer.

But what difference does it make if the air undergoes heating, convects and then cools to air that under goes heating and cooling between pockets in a linear fashion? I cannot see the difference

Not sure how air is "drawn in" to a cavity wall acting as a chimney. Mine are supposed to be wind- and water-proof all the way to the ground. 

Building standards are still not what they should be, 20 years ago you could have a cavity open to the loft space, as well as the plasterboard dot and dab being open. Walls today have exterior ventilation placements in the brickwork as well as an airgap infront of the insulation to guard from condensation. Years ago air would go into the open hollow cavity at the eves and would eventually exit at the ridge, not warm but no condensation.

As for the density and conductivity of the filler, polystyrene or polyurethane foam is about 96% air by volume - you can pretty much ignore the plastic content

I wondered whether the change of density, (gas to solid) would limit transmission, it takes much more air to raise a solid by the same temperature. Hence slowed transmission, or insulation?

I simply cannot see much difference between air in non convecting horizontal containers or a big cavity? Even though one is moving, they are both at some point conducting to and from other materials?

[evan_au]

I simply cannot see much difference between air in non convecting horizontal containers or a big cavity?

Size matters!

There is a concept called the Reynolds number which indicates the transition from laminar flow to turbulent flow.
- And turbulence occurs more in small spaces, while laminar flow is possible in large spaces
- A model aeroplane in a wind tunnel is not the same as a full-sized plane, because the Reynolds number is different for the smaller model

I vaguely recall a comparison which went something like this: A bee flying in air is like a human swimming in honey.
- A bee flying has a Reynolds number around 1,000
- Unfortunately, the following list doesn't include a human swimming... https://physics.info/turbulence/
- Although I saw estimates of around 50,000 (for a spherical human in water...)

https://en.wikipedia.org/wiki/Reynolds_number

[Petrochemicals (OP)]

And turbulence occurs more in small spaces, while laminar flow is possible in large spaces

But what is the difference between air convecting and air transmitting, it all has to "pick up" heat one side and deposit it the other, whether it is replaced in a flow, or is static, all of the air has to perform this task.

Would a cavity wall with 10cm high partitions in a 10cm cavity or even 10cm cubes really perform better than a 10cm open cavity?

[alancalverd]

But what difference does it make if the air undergoes heating, convects and then cools to air that under goes heating and cooling between pockets in a linear fashion? I cannot see the difference

It wouldn't make a difference but it can't happen. Convection depends on establishing contrary laminar flows, like vehicles on a dual carriageway. If you narrow the road to a single track, you can't maintain a continuous high speed in both directions.

 

it all has to "pick up" heat one side and deposit it the other,

Now replace the cars with a single rogue dustcart.  Convection means driving down one side of the street to collect stuff from n houses, then driving back up the other side, depositing rubbish as you go, so you shift all the stuff in a distance 2l where l is the  length of the road.
For conduction, you shuttle between the sides  of the road, taking one bin at a time so you have to drive 2nw to shift the same amount of stuff across the road, where w is the width of the road. Replace the dustcart with a molecule and n becomes very large. 

So conduction is less efficient than convection unless w is very small indeed. 

So in the case of a window we make w just small enough, and in a cavity wall we break l into small bits, to prevent convection.

[Petrochemicals (OP)]

Convection means driving down one side of the street to collect stuff from n houses, then driving back up the other side, depositing rubbish as you go, so you shift all the stuff in a distance 2l where l is the  length of the road.
For conduction, you shuttle between the sides  of the road, taking one bin at a time so you have to drive 2nw to shift the same amount of stuff across the road, where w is the width of the road. Replace the dustcart with a molecule and n becomes very large. 

Yes but in the case of conduction you have more dust carts working all at the same time doing one bin, where as with convection they are all queued up doing an entire side of the street one after another. I would expect the working all at the same time to win to be honest.

It all relies on each air "cart" conducting heat at one side and depositing it the other.

[Eternal Student]

Hi.

For conduction, you shuttle between the sides  of the road, taking one bin at a time so you have to drive 2nw to shift the same amount of stuff across the road, where w is the width of the road.

  (i) Those formulae looks suspicious.
   The cart can't just move back and forth between the same houses that are on opposite sides of the street.   After doing a shuttle run it must move down to the next pair of houses straddled across the street.   That's a movement of  δL = L / (n/2)    =  total street length / # pairs of houses.   It does this n/2 times to reach each pair of houses, so it moves a total  distance (n/2) . L / (n/2) =   L  down the street.     So the formula you want   is     2nw + L or something like that.   Maybe doubled if it's allowed to turn round at the end of street and start the shutlling of energy again the other way up the street.

   (ii)   It's a poor analogy anyway, I would have thought.   Conduction is not the movement of moelcules each carrying some energy payload.   That's more like convection.

- - - - - - -
   This is the sort of analogy that woud have been used in my school....

Conduction is:   The transfer of heat from one molecule to the molecule next to it.   In a solid, it's reasonable to assume the molecules can't make huge movements from one place to another, they are tied down to more or less one location.   However they can be agitated and jostle about that position.    When they are given some quantity of heat, Q, they become more agitated and really dance around that position.   This starts to jostle the molecule next to it until that molecule is also very agitated and now has some extra heat energy.    It's possible to imagine that some quantity of heat, Q, that you put in at the left is effectively passed along a line of molecules in this way until it reaches the right hand side.
    Now air is a gas at standard temperature and pressure,  so its molecules are much more free to move around than a solid.   However, that movement is quite random, the molecules go off in all sorts of directions.   So you've now got to imagine a group of ladies (or gentlemen) dancing around their hand bag but now the hand bag is often kicked by someone else accidentally and the ladies move to stay with their hand bag.   They definitely have some large scale movement but it's random, they go in all sorts of directions.    The main transfer of heat is still based on agitating the molecule next to themselves.    In air, this conduction process is a slow one.

    Convection is a bit different.   Here there is a very organised flow of molecules.   When some quantity of heat, Q, is passed to a molecule on the left, it doesn't need to try and pass it along to the next molecule by the usual agitation process.   There is an air current that will fairly quickly move the molecule and it's not random, it's reliable.   The molecules with the extra heat are swept in the current to the place where it was cold  (the right hand side), where they can now deposit their extra energy on the right hand side.    In air, convection can be a much quicker process.    As opposed to passing a bucket of stuff (heat) along a line of molecules (which is conduction),  the molecules take the bucket and just run with it along an organised route to where it must be deposited.

      Where convection is possible, the rate of transfer of heat from the left side to the right side can be very high.   However, if you can prevent convection then the air can only use conduction to transfer heat from left to right which is slower.   As the earlier posts have suggested,  in small spaces, Convection is prevented.  Any attempy to establish an organised flow of air is quickly disrupted and just becomes a chaotic flow of molecules in random directions. 

Best Wishes.

[Bored chemist]

This sort of thing

Sorry i cannot believe convection

Is the reason why I posted stuff that led to this
"For reasons of repetitive antagonism, this user is currently not responding to messages from;
BoredChemist"

It's not just that he can't science, it's that he refuses to science.

[Bored chemist]

And turbulence occurs more in small spaces, while laminar flow is possible in large spaces

But what is the difference between air convecting and air transmitting, it all has to "pick up" heat one side and deposit it the other, whether it is replaced in a flow, or is static, all of the air has to perform this task.

Would a cavity wall with 10cm high partitions in a 10cm cavity or even 10cm cubes really perform better than a 10cm open cavity?

Viscosity.

[Petrochemicals (OP)]

      Where convection is possible, the rate of transfer of heat from the left side to the right side can be very high.   However, if you can prevent convection then the air can only use conduction to transfer heat from left to right which is slower.   As the earlier posts have suggested,  in small spaces, Convection is prevented.  Any attempy to establish an organised flow of air is quickly disrupted and just becomes a chaotic flow of molecules in random directions. 

Best Wishes.

So a line of molecules 1 thick in a circulation that is based on conduction from on one side and conduction to the other seems scientifically more transmissive than a column containing many particles each conducting heat from one side to the other? It is all based on conduction. The temperature difference in the air is one factor, if you heat for longer it will achieve a faster cooling, but the heating likewise will take longer, ea molecule with ten times the temperature will take >10 times to heat. The heat conduction of air remains the same and central whether transmitting in a circulatory manner or a conduction manner.

And no one seems to want to answer this. It's no good just saying "science"

Would a cavity wall with 10cm high partitions in a 10cm cavity or even 10cm cubes really perform better than a 10cm open cavity?

[Eternal Student]

Hi.

So a line of molecules 1 thick in a circulation that is based on conduction from on one side and conduction to the other seems scientifically more transmissive than a column containing many particles each conducting heat from one side to the other? It is all based on conduction.

     Some of that was right.   One thing is always due to conduction.   The air molecules get their heat from the left side by conduction from the wall molecules to the air molecue.   The air molecules also depoist their heat on the right side by conduction (from the air molecule to the wall molecules).     However, those are the only bits of the process that have to be done by conduction.     With convection the air molecules are quickly moved from one wall to the other wall.    If you prevent convection than the only way heat can get from the left wall to the right wall is by multiple instances of conduction from air molecule to the next air molecule in the space.

Would a cavity wall with 10cm high partitions in a 10cm cavity or even 10cm cubes really perform better than a 10cm open cavity?

     Yes, you would have though so.
     However, how thick are the bricks that form these partitions?  If they are 10 cm wide then you replacing a lot of what could have been air (a poor conductor) with something that is a much better conductor.    So you'd get better results with some thin material that is still fairly resistant to air passing through it.
     Also, heat insulation is only one thing you consider when building a house.   You do actually want some air circulation to prevent damp problems and other stuff.

Best Wishes.

[Petrochemicals (OP)]

.     With convection the air molecules are quickly moved from one wall to the other wall. 

At 1 molecule thick it would require a flow rate of considerable velocity, something we know is not present. The mixing of particles in a stream would mean the turbulence as mentioned by @evan_au, and ruled as destructive by @alancalverd, if however these stream are undergoing conduction it would rule the advantage nil and void.

However, how thick are the bricks that form these partitions? 

I was thinking more along the lines of cubes of air in thin plastic containers, such as you find in giant bubble wrap packaging just placed within the cavity. It would be many many times cheaper than filling the wall with expensive glass fibre.

[alancalverd]

Bubble wrap works pretty  well (ask any rough sleeper) but tends to collapse under its own weight and is a fire or toxic smoke hazard, so it isn't used in buildings.

[vhfpmr]

Anyone who thinks thermal insulation doesn't work could try taking all their clothes off and see if they notice any difference. This time of year is an ideal opportunity for an experiment.

[Petrochemicals (OP)]

Bubble wrap works pretty  well (ask any rough sleeper) but tends to collapse under its own weight and is a fire or toxic smoke hazard, so it isn't used in buildings.

Very helpful, but unfortunatley in this senario the cubes are made from fire resistant non smoke highly durable super resiliant uber duber kevlar stuff that poses no threat to the building, thermal losses or the cost.