Somehow I lost this post for 6 months, could it be PASH ?(passive annular simple head). what great links and info folks.
I'm about to start some bore holes this winter and find out what the subsoil depth is etc. we are on boulder clay with limestone base, I put some deep drains, 100m away and slightly (3meters) down hill from the proposed heat-store site, several years ago. we got 5 meters down and were still in clay, that was the limit of the digger and deeper than we needed for that job, while they were their it was worth the extra time as I was hoping to find the limestone !.
So I'm proposing drilling holes 100mm dia and lining them with stainless steel pipes, I was aiming for a depth of 6 meters (6 months @ 1meter/month) and then having a coil in the bottom meter of the liner circulating an antifreeze solution from solar panels during warm weather. perhaps a solar powered pump would be sufficient in a closed circuit. The question is how many bore holes and at what spacing ?. the building is 10m x 5m approx with open land east,south and west.
Tom

mike7, we didn't complete our previous discussion about this. You showed me that the temp gradient gets steeper (square-law curve) as you move in from infinity to the core of the heat store. But to me that doesn't prove your point about losses only becoming acceptable the bigger the store volume is; i.e. that single-house schemes are doomed. Given sufficient time (years) the temp gradient will get steadily flatter and flatter, so loss steadily diminishes. That's true even if that very flat curve is steeper near the core - despite that fact, it can still get very flat even close-in.

Tom - sorry but the answer really does lie in the maths.... eg. if the store was one-dimensional, say a long perfectly insulated rod, then it would work as you suggest. In two dimensions apparently it would too, although I find this harder to visualise. Also, it would approach the steady no-leakage state much more slowly, if I'm getting it right.

The third dimension of a real store gives a solution which has a continuous leakage at final steady state. Hence your problem.

Posted By: Jeff Norton (NZ)I have been waiting for someone to simulate thermal storage with dynamic software?

Love to! and I have done some preliminary stuff. Unfortunately require funding to take it further. To my mind there is several YEARS of research [and software development] before we are at a stage where we can confidently predict results with DSM software.

I too started some thermal simulation software with a view designing a PAHS type house then dropped the matter a bit over a year ago when I had an attack of common sense and decided that the whole thing was too risky. In particular, I was worried about having to dig the whole thing up again if it was found that water was flowing through the bit that was supposed to be storing heat.

mike7, I will pass your comments on to Wates and Arups, I am sure they will be delighted that you are calling them academics. You might have missed the point about SURegen, it is funded by the government (£2.5M) and by industry £500K and is designed to benefit industry. It is a project that has been created in partnership with leading practioners over the past few years.

Posted By: shelterreOne of the best performing buildings in the UK which uses PHS is the Mile End Ecology Park Building, London

Gorgeous buildings/bridge, but things have moved on quite a bit since then, haven't they? What's the state of the art now?

How many years before I (as oppossed to some expert - acedemic or not) can get an indication of how much money I might save by making my 2 top stories (SE and SW walls) Trombe walls (about 100m2 worth not including glass)? The aim of this stuff being, I hope, to make a difference to the people a fit further down the food chain - and I don't mean at the bottom just not people doing energy calcs as a reguler part of ones life.

I don't want to hijack this thread but this topic has the information:

http://www.greenbuildingforum.co.uk/newforum/comments.php?DiscussionID=2849&page=1#Item_11

Which is making progress but its painful.

Jeff, please do get in touch: jlittlewood@uwic.ac.uk.

Try it longhand:-
The store is a complete sphere radius r1 immersed in an infinite field of uniform properties:
Conductivity K = 1.0 W/mC. Vol. Spec. heat V = 2.0 Ws/m3.C 10*6

The temperature distribution in the region beyond the influence of the annual fluctuation is in a steady state, i.e. that which the system will tend to eventually.

Far ground temp T0 = 10C, Store minimum T1= 30C, Max T3= 70C, store average T2= 50C.

Store capacity Qs = 4/3pi.r1^3 V (T3 - T1) Ws which is 93.r1^3 kWh

Heat loss rate = 4pi.K.r1(T2 -T0) = 503.r1 W
Therefore annual loss QL = 365.24/1000 (503.r1) = 4400.r1 kWh

r1 can be eliminated using the above to give QL in terms of Qs:-

QL = 970.(Qs)^1/3 kWh (that's the cube root of Qs there)

With this I can make a table of the total heat input required annually for various sizes of heat store:-

Qs 10,000 kWh......QL 21,000........Qtotal 31,000 ie multiplier of 3.1
....20,000 ...............26,400.................46,400....................2.3
....50,000................35,800.................85,800....................1.7
...100,000...............45,200................145,200....................1.45
...500,000...............77,200................577,200....................1.15

Note that a complete sphere will have lower losses than a real store even if insulated.

If the outer limit were say 10 times the r1 radius instead of infinity, the losses would only be about 10% greater - it just makes the maths easier.

Soil properties chosen are on the favourable side, but not the best. This is the formula if you want to try different values:

QL = 1224K(Qs/V)^1/3 kWh

Using a different value of T3 doesn't affect the results much. 30C seemed the lowest realistic value for T1 unless a heat pump is used.

I've been assuming that all the central store capacity is available to meet the heating load, and none is required to buffer the heat loss, other than that provided by the zone between the core and the steady state region (marked A and B on a diagram I have yet to post). My assumption may be wrong.

Hope this helps!

Note please that my various caveats mean that you will need not less than 25000 kWh - how much more you'd actually need I don't know. I wouldn't fall off my chair if it turned out to be twice as much. Nor do I know how to calc the borehole design, but a bunch of shorter holes would be better than one long, perhaps in a wigwam-style pattern to reduce losses near the surface.

How far down before you hit water where you are?
I salute your indefatigability!

The size of the usable core of the store depends on the details of the heat delivery/retrieval system.
I'm assuming an array of pipes (not a single insertion point) within the core so that ideally all parts of the core will achieve the max temp when full, and the min temp when empty, during a typical yearly cycle. Surrounding this core is the A and B labelled zone or shell which will respond to the annual fluctuation of core temp, but to a lesser extent the further out one goes.(Using the 6 metre per year rule of thumb, one could suppose this shell might be around 3 metres thick). Outside this AB shell, the annual fluctuation would not be felt, as indicated in my graph of temp versus radius.

The curve is perhaps more neatly described as a plain old reciprocal curve ... T = Constant/R. The heat required to build up this 'bund' of temperature has of course been built up over several - or many - annual cycles until the steady state is approached. At the steady state, the gradient the reciprocal curve MUST have to satisfy the maths of a spherical geometry is the gradient that gives rise to the continuous heat loss, oozing away in three dimensions rather like a puddle of spilt oil would in two (but the curve would be different!). The bund thus needs fresh dollops of heat annually to maintain its 'height'.

I hope this clarifies it. Pip pip

Thanks fostertom!

A single point source would have too high a temperature: T = Constant/R, so small R means high T.

Doesn't have to be all at the max/min temp, but the sums are easier. And the store is more compact, the losses lower.

Further out? Too thin? Maybe, maybe not. Show me your sums!

"Is it fair to say...." Sorry, you've lost me there. But one thing a canny store designer/user might do would be to charge the centre of the core first, and draw on the outer regions of the store first. That might help a bit. We need it.

My conclusion so far.
Instead of bore holes/ducts (fairly expensive), if I dug a trench say 4m deep by 1m wide 4m long (house foot print app 10m x 7m) down the center of the foundations and installed a water storage tank a 1m tall by 800mm wide 3.8m long,

then insulated the sides and top with 100 PUR and back filled.

I could then charge it during summer with an indirect coil to allow stratification, eventually the whole tank would heat and thereby transfer some heat downwards into subsoil. from that point the 6m rule will/should kick in.. etc,etc.

Having a liquid store makes good sense for DHW with a smaller tank/coil upstairs, maybe even UFH.

Your excellent comments please... I'm trying to keep up with them, but need to read them several times before they sink in

Tom