Latest crackpot scheme (of mine)...

[DamonHD]

I'm still working on it, and there are some pieces I don't know how to finish yet, but it's not looking utterly absurd (by my low standards) yet...

Summary: 60,000l of water in stainless-steel tanks heated by 13m^2 of solar thermal panels may be able to provide DHW and space heating all year round in London instead of mains natural gas.  By comparison, zeroing out our grid electricity use would take maybe twice the roofspace.

http://www.earth.org.uk/milk-tanker-thermal-store.html

Rgds

Damon

[MACM]

Hi Damon.

I heard about a german firm that had parafina in the acc. tanks instead of water. Much higher energy desity and selfconsavation of the tank.

//MACM

[DamonHD]

Hi,

Surely very little that's liquid and non-toxic at 'room' temperature and pressure beats the specific heat capacity of water, and water won't corrode a stainless-steel tank AFAIK.  What am I missing?

Rgds

Damon

[oztules]

Well, there's 50000ltrs of water in 2 plastic water tanks just outside here filled from the roof run-off. It wouldn't be too difficult to insulate them with 300mm of insulation batts wrapped around the outside and then plastic wrap them.

Although with plastic tanks 6-8mm thick, max temp would be 85-90deg C.

Each tank is about $3000 each. so if you don't already have the tanks it may be an expensive operation. But if it doesn't make expectations, at least you will have your own water supply.

Water is cheap (free off the roof) and is very high on the list of liquids with good specific heat figures.

...........oztules

[DamonHD]

One possibility is to simply bury the (top of the) tanks more than about 6m deep (since the temperature gradient moves about 1m/month) and let the surrounding soil add to the thermal mass of the store providing that there is nothing to carry excess heat away such as moving groundwater...

Anyway, yes, sounds like you have your seasonal store right to hand!  You might be more interested in the summer-cooling possibilities than me too.

Rgds

Damon

[feral air]

What would work best (I think!) is to pull cold water from one tank, heat it and move it to the other. That way it can't spread that little bit of heat across 60,000l of cold(er) water....

I'm imagining (/equating it to) trying to boil a big pot of water with a candle and it doesn't work fast enough. But if you boiled a tablespoon at a time and put it in a separate insulated container it would work ok (assuming minimal lid-lifting losses ;-).

I'm not sure how you would have it work when you're actively using the heat though. I mean, if the water drops 2° on its way through the house would you put it back in the hot tank or in the cold one? I'm guessing you'd want to put it back in the hot tank near the bottom (heat rises and all that)...but when would it be best to send it to the cold tank?

Would it be better to stand the tanks up or lay them down? Standing it up would stretch the gradient and separate the hot and cold by a longer distance...no?

I doubt many people that live in town could do this but it seems like it ought to work somehow.

[Ungrounded Lightning Rod]

You aren't running a heat engine so you only care about total stored heat, not what temperature the storage medium was when you added it.

For storing heat pull water from the bottom of the tank, run it through your collectors, and dump it into the top.  For using the heat, pull the water from the top and return it to the bottom.

That will make your water stratify, hot on top, making a single tank act like a hot and a cold tank with the boundary shifting as necessary.  It will also maximize your heat transfer from the collector (due to the cold input water encouraging heat transfer into the water) and do similar good stuff for pulling the heat out into the house.

In both modes it reduces the amount of water pumping you need to do, compared to other connections, to get a given level of performance.

[Chagrin]

Google "slab on grade" for how ~2" thick poly foam can be used to insulate a slab foundation. As absurd as it seems, you could insulate a few feet below the surface to reduce your required digging depth.

At any rate, it certainly is an interesting crackpot scheme.

[DamonHD]

Thanks, am looking now...

Rgds

Damon

[DamonHD]

Hi,

I hadn't thought of stratifying within the tank: an excellent idea.

I had thought of having several separate tanks, both for engineering robustness in case one sprang a leak or whatever, and also so potentially I could keep one for space heating nearer 30C and for DHW nearer 60C to reduce the thermal gradients that the heat-pumps have to work against to improve system efficiency.  That would mean added complexity, and possibly needing two heat pumps, so might not be worthwhile.

Rgds

Damon

[feral air]

"You aren't running a heat engine so you only care about total stored heat, not what temperature the storage medium was when you added it."

Except you want as much hot water as you can make, not 50,000l of luke-warm water that you can't really use. By putting it back into the same tank you're letting the cold mix with the hot to some degree. If your panels can't heat all 60,000l to your target temp over the summer then any mixing is bad for business.

Also worth thinking about is the surface area differences. With one tank the water is always contact with the whole tank, leaking the heat into the ground. With a separate hot tank there would be less water in contact with the tank (until there isn't) to leak the energy.

Another thing that could increase efficiency (and comfort) is to use the cold water to cool the house before sending it to the roof to get heated more. This would be easier to do if you had a separate cold tank since anything coming out would be cool by default. If you had a single tank you would need a bypass in case the water was too warm.

I dunno, I still think 2 tanks would be better. Not orders of magnitude better but somewhat...and it could be more convenient in some ways. take it easy

[DamonHD]

I've updated the page with a little diagram.

The design now incorporates an over-size hot-water tank (ie thermal store) for more than 24 hours' heat/DHW so that the heat-pump can be driven by solar PV to avoid needing grid-power (though the diagram shows the PV grid-tied).  A bigger in-house store allows for more days without decent sunshine without needing grid power.

The link is: http://www.earth.org.uk/milk-tanker-thermal-store.html

Rgds

Damon

[Ungrounded Lightning Rod]

Except you want as much hot water as you can make, not 50,000l of luke-warm water that you can't really use. By putting it back into the same tank you're letting the cold mix with the hot to some degree.

If you put the cold into the bottom and the hot into the top (with the plumbing aimed horizontally to avoid vertical stirring) the degree of mixing is minimal.  Without heat below / cold above to drive convection it takes literally decades for water to mix a vertical inch by diffusion.

Drop into the chemistry building of a major university.  You're likely to find a demo of this:  A vertical glass tube full of water in a closed display case (to keep the temperature changes slow), with a few crystals of potassium iodide at the bottom to color it and a marked scale giving the dates of the level of coloration (or a set of such tubes side-by-side, a new one installed every few years with the initial color level at a given height.)  I know there's one of these in the chemistry building of the University of Michigan, and it's one of those classics that's likely to be replicated in most colleges.  Gives you a real feeling for the glacial speed of diffusion versus other forms of mixing - which is a significant issue in the dynamics of chemical reactions.

Heat travels faster by conduction.  But you're still talking months, not minutes, to make a dent on temperature equalization.  You're interested in hours, and you're just fine at that timescale.

Also worth thinking about is the surface area differences. With one tank the water is always contact with the whole tank, leaking the heat into the ground.

Which is a fantastic insulator as well as a helpful storage medium.  Your main concerns are vertical conduction by the tank wall and heat transfer to/from ground water if you have significant underground flow.  The former is a small but non-trivial issue with a metal tank.  With a fiberglass or poly tank vertical conduction is not a significant issue either.  The latter is a problem regardless of whether you bury one tank or two.

I suspect you'll actually be ahead to use a single tank with vertical stratification due to heat storage in the ground around it tending to help you out rather than fight you.

Also:  A single tank has the advantage that it is always full of water at roughly the same density as the ground around it.  This reinforces it against collapse.

[Ungrounded Lightning Rod]

I'd only go for two tanks if you also want to store cold for cooling.  B-)

I notice your consultant quoted you a vacuum tube system.  That may be overkill unless you want to store some VERY hot water.

There are three main types of (non-concentrating) solar collectors for water heating:

 - Unglazed (black-painted metal sheet with tubes soldered to it, serpintine or coil of black tubing, or outdoor storage tank painted black)

 - Glazed (ditto under glass in insulating box for greenhouse effect and to keep the wind off)

 - Vacuum-insulated.

Unglazed is for collecting heat at roughly the ambient air temperature, so little or no insulation is required.  Typical use is for heating a swimming pool - bringing it up to temperature after a fill and keeping it warm against evaporative cooling.  The "black tank" is also useful for hot showers in the afternoon and is often used at campgrounds (and typically reserved for staff use B-) ).

Glazed is usually adequate for hot water heating and the like.  You get much higher temperature output because you're not losing very much heat to conduction.

Vacuum insulated is where you pay some big bucks to get extremes of temperature.  It might be good for your stoarge system, since it increases the amount of heat you can store in a given amount of water.  Then again it may be overkill.

If you do go for vacuum insulated collectors, I'd consider glazed also, with the two in series:  Glazed first to get the temperature up cheaply, then vacuum to really peak up the temperature just before the water goes into storage.

(Myself I'd go for trough concentrators and antifreeze-solution storage with a heat exchanger before worrying about vacuum tube collectors.  Heat exchangers are easy:  Pipe-in-pipe using standard plumbing-grade copper tube and fittings, wrapped in insulate-your-pipes foam.  Or peel the insulation from your hot water heater temporarily and solder a coil of copper tubing wound around the tank.  Remember to use counter-current for efficient heat transfer.  Pump the heat storage solution in at the top out at the bottom.)

[DamonHD]

Thanks for that.  I simply chose the vacuum tube as the most 'efficient' collection mechanism per unit area as roof space is potentially limiting, and considering other costs, I hadn't looked at the collectors in any more detail.

What you say about glazed and vacuum in series sounds very interesting.  I don't know if you've revisited my page very recently, but I've thought a fair amount about having an in-house thermal store to cover more than a days' energy to ride out cloudy days and to allow the heat-pump to run only when local solar PV is available.  The water in that over-size hot-tank should be kept above 60C to kill Legionella, so higher temperatures from the solar thermal collectors (to minimise need for a the heat-pump when it would be rather inefficient) would be good to help achieve that.

Actually, roofspace would seem under greater attack from the solar PV needed to drive the heat-pump according to my sums!  I'm running out of fingers and toes... B^>

It's now starting to look to me that to be truly carbon-neutral all year round, even each and every mid-winter day, with solar thermal and PV as the RE sources, is looking more like £200k, which just shows what the grids (electricity and gas) and a stable society (to allow them to exist) do for us!

Our current house probably isn't going to be the one to get the treatment, but when we choose our next one, we'll be looking for something that can support some or all of these mechanisms.

Rgds

Damon

[DamonHD]

BTW, I've quoted your idea on my page, thanks again...  B^>

[feral air]

If you put the cold into the bottom and the hot into the top (with the plumbing aimed horizontally to avoid vertical stirring) the degree of mixing is minimal.  Without heat below / cold above to drive convection it takes literally decades for water to mix a vertical inch by diffusion.

Yup. My pond (which I've been mentally comparing this against) holds about 18k gallons by mid-summer and that's pretty obvious if you dip a foot in. The top 6" is nice and warm but below that it's shrink-your-twig'n'berries cold. Even after the kids go swimming(=mixing) there's not a noticeable change in temp...I think I was splitting hairs.

One thing that could be tough is aiming the pipes horizontally. Most of the tankers I've seen have the port in the bottom (the center of a wall if it's vertical). Most normal people wouldn't want to try punching holes in the tanker so they'd have to route the pipes in through the existing port and around the inside of the tank....the hot pipe will be in contact with some cooler water and some convection will occur.

Mixing doesn't matter if you can heat the whole tank though and at this size it may not matter anyway. When you get right down to it, that's a lot of water.

Heat travels faster by conduction.  But you're still talking months, not minutes, to make a dent on temperature equalization.  You're interested in hours, and you're just fine at that timescale.

Why hours? I thought the idea was to build up the (majority of the) heat over the summer for use in the winter, which means you're storing it for months. If that's not the case then why is there so much water in the system? Maybe there's a misunderstanding on my part.

Which is a fantastic insulator as well as a helpful storage medium.  Your main concerns are vertical conduction by the tank wall and heat transfer to/from ground water if you have significant underground flow.  The former is a small but non-trivial issue with a metal tank. With a fiberglass or poly tank vertical conduction is not a significant issue either.

Except in the winter it's more likely that your tank is surrounded by water if it's at any kind of depth. Placing it vertically should reduce the wall-loss but it's still a lot of surface area. I've never seen a fiberglass or poly milk tanker, always stainless steel. Either way the "head" (where you lose most of your heat) would be the same size if the tank(s) are vertical.

I suspect you'll actually be ahead to use a single tank with vertical stratification due to heat storage in the ground around it tending to help you out rather than fight you.

Now that I've given it more thought, 1 tank would be better. If you had a bunch of air hanging out above your hot water you'd lose heat to the air...convection. The problem I was trying to fix (that wasn't much of a problem) rears its head in my half-baked 2-tank alternative. Way to go, self!

That said, I don't think you want the tank in contact with the ground at all. You'd want to insulate the heck out of it because any loss to the ground is most likely a permanent loss (since it's winter). Sorry for meddling...take it easy

[DamonHD]

Hi,

Yes, the aim of the large seasonal store is to capture heat in summer and withdraw it in winter.

If I go more than (say) 10m down I'm below the surface of the Thames river not so far away as the crow might amble.  Indeed the Thames is still tidal a little way beyond us.

I'm guessing that the ground 10m+ down will therefore be permanently saturated.

My further guess is that if that water doesn't move then the tank(s) could be smaller and uninsulated and the water/ground will add to the thermal mass and stay around to have its heat pumped out in the winter.  Maybe like this if I've understood it correctly:

http://www.building.co.uk/story.asp?sectioncode=747&storycode=3097129&c=2&encCode=000000
00013b6439

However, if there is movement of the groundwater then I think the tanks have to be insulated to the hilt and protected against seepage to prevent the heat being carried away.

Rgds

Damon