ya that what this house has. Stone floors with radiant heating underneath. I cant help thinking there is a huge difference between the energy needed to change the temp and the energy needed to maintain the temp. for example here when I change the thermostat its 2 or 3 days before I notice a difference. At the ranch I wonder if I used the electric basseboards to get the house to temp (with some thermal mass like I'll put stone tile under the wood-stove and in the bathroom and maybe some baskets of rocks in the basement) can I design a water system just to maintain that temp? with less demanding specs?

What you're dealing with here is the thermal mass of the stone/concrete floor, which is unrelated to the average load, but will reduce daily peak loads significantly. The thermal mass of the floor is storing and releasing heat. When it's cold out I'm sure the water temp in your floor needs to be at least ~30-35C to keep up, but there's a lag time of hours before you see it in the room temp as the boiler's output charges up the thermal mass of the stone.  PID thermostatic controllers are usually required to keep room temps stable when managing large thermal masses. 

But the energy required to keep the room to a specific temp will still be the same- it just has this thermal-mass buffer to deal with.  But when heated by a boiler the FUEL required might be lower, since having the thermal mass load to work on increases the burn length on the boiler and reduces the number of cycles, which brings it's average efficiency closer to it's steady-state thermal efficiency.

At the ranch, the average heat load issue will be the same, but you can reduce the peak loads by adding thermal mass, which means you may be able to run your setpoint a bit closer to the freeze-limit.  Here again, per unit volume water will provide slighly more than 2x the heat storage per degree than concrete/stone, and per unit weight about 2.5x that of concrete/stone, making it the preferred storage medium. Water is a very convenient storage medium.  (The specific-heat of concrete is ~ 0.18x that of water, and it's density is only ~2.5x that of water.) The more thermal mass you put in the space the lower the daily temperature swings will be, but the overall amount of energy you need to pump into the system during the heating season will be the same. for less-well insulated structures the fuel used can be quite a bit lower on  during the early fall & late spring shoulder seasons, when the daily outdoor temperature and solar gain swings go between heating & cooling.  For superinsulated structures the additional benefits of high thermal mass isn't all that great from a heating perspective, but can still be a factor during the cooling season to moderate the temperature of daily solar gains, allowing one to use night-ventilation as a primary cooling method. 

The amount of heat stored at moderate temps with slabs/rock/tanks inside the house will still be less than your daily heat requirements on a day when it's -25 out. And if the walls/ceilings/foundation are all R30+,  despite the much smaller area, the R2-ish windows are dominating your heat load.

Seal and insulate the place as best you can, including removable insulation over the windows when you shut it down for the winter and see how you make out.  By the time you've insulated the foundation down to the frost line and sealed the crawlspace, with tight-fitting insulating shutters over the windows your average heat load to keep it at 10C may already have been cut in half or more, and the freeze-up risk for setting it to only 5C may be going away.  Then by the following season if you can set up 100 square feet of thermal air panel it may well be enough to cut the remaining load by half or more. 

R12 panels to fit the windows from the inside are dead-easy to build out of 2" iso board, are light, easy to handle & store, etc.See:  http://www.builditsolar.com/Projects/Conservation/Slider/SlidingDoor.htm

The next-most cost effective measure would then likely be 5-10 square meters of thermal air panel, which may be enough to reduce the seasonal duty-cycle on your electric baseboards down to inconsequential levels.  There are many variations on the theme, passive thermosiphoning, active PV/grid powered, etc.  and you may want to make them building-integrated & permanent as the primary heating source for the place, leaving the wood stove & baseboards for backup. (You'd have to shade them in the warmer season to keep from cooking yourself though.)  Designs are well evolved since the 1970s, and are readily available on the web.

Then by the following season if you can set up 100 square feet of thermal air panel it may well be enough to cut the remaining load by half or more.


I think of the house as insulation powered. I mean if I do a good enough job on that side I might just heat it with baseboards if it's like $50 a year rather than 300.

You've got it! 

When you drop the heat load low enough the size & efficiency of the systems need to support it fall into the "don't care" category. That's the PassivHaus approach.

See:  http://www.e-colab.org/ecolab/SmithHouse_files/EnergyDesignUpdate%20Article1.pdf

The size and efficiency of a massive dirt or water storage system to support a less-well-insulated house gets out of hand pretty quickly.  Applying the insulation you would have used on the tank/dirt to the house works- you're collecting & storing the heat where you actually want to use it. 

But if it's unoccupied in winter it doesn't have the 150-180watts/person + whatever other heat being dissipated inside, so a bit of lo-tech thermal air isn't a bad idea to think about to truly get that down to $50/year.  You can build-in some pretty substantial thermal-air for less than $500 in materials, eg:

http://www.builditsolar.com/Projects/SpaceHeating/SolarBarn.pdf

For you an active system with a small fan and a differential thermostat rather than a passive thermosiphon like the one above probably makes more sense, since it'll have much smaller breaches in the thermal envelope.

the basic Idea now becomes to throw away all the heat in summer and get heat from the sun in winter. This isn't quite the idea I have long term. In a continental climate the summer is so hot and the winter so cold that my goal is to do a system for annual storage. There is a tradition of ice caves around those parts where ice is stored from the winter for the following summer. It's worked great for hundreds of years. While common sense and economics lead me to the efficiency minor/heating solution, curiosity leads me to annual storage. Gottat get all that sun from summer somehow and use it in winter.

reading up here, I think the solar air system is cheap and good but I want to work with my well water, and just heat it up a bit from a panel. Maybe I'm reading wrong but this air system and all its vents seems to demand too many holes in my nice tight structure. Just a half inch pipe should be all I need...?

Posted By greendreams on 09 Feb 2010 05:53 PM
reading up here, I think the solar air system is cheap and good but I want to work with my well water, and just heat it up a bit from a panel. Maybe I'm reading wrong but this air system and all its vents seems to demand too many holes in my nice tight structure. Just a half inch pipe should be all I need...?

With an active thermal air panel  you're looking at  maybe 2-4  round holes, 10-20 cm in diameter and one or two 20-35 watt fans. Thermosiphon version have typically 10-20x as much cross sectional area for the vents since it's relying on the much lower force of the buoyancy of the heated air to move the heat.  Efficiencies of active thermal air panels are comparatively higher due to the lower loss from the vent holes and higher turbulence on the panel's heat exhcanger, but they do require power to turn run the blowers.  Some use photovoltaic powered blowers to simplify control (the stronger the sun, the faster the DC motor turns.)

This ~4-5 square meter homebuilt has only two 160mm holes:

http://www.builditsolar.com/Projects/ActiveWallCollector/activewallcoldet.htm

 

All thermal air panel designs need  backflow prevention to keep it from thermosiphoning out heat. But active versions with blowers just work better. Designed right you get more heat capture per unit area, lower (= more efficient) operating temps.  Designers who brag about high delta-Ts or high output temps on their homebuilt solar collectors are missing the all important fact that high temp = high loss = lower collection efficiency. 

Some really rough order of magnitude numbers (not to be considered gospel without a lot of more specific analysis & design):

A really decent active thermal air panel design running at low temp will suck in ~0.2-0.3kw/square meter during the middle part of a sunny day- more if there's snow on the ground or when outdoor temps are a bit warmer resulting in lower loss at the panel.  So a 10 meter panel running 5 hours/day dumps ~10-15kwh of heat/day into the building.

Assuming 20 sunny days/month, that's 200-300 kwh/month you didn't spend on running baseboards.  At 7 cents/kwh that's  $14-21 in savings.

If your wintertime bills were $50, and you 've cut the heat load in half with your insulation & insulating shutters you might have spent $25 heating the place with baseboards.  Odds are good that 8-10 meters of air panel will deliver something like 90% of the load, avoiding most of the power costs, but not all::

If it costs you 100W of blower power run it, that's 0.5kwh/day, or 100kwh/month or $7 @ 7cents/kwh.  If it turns out to be only 90% of the load you'll have to spend another ~$3 in baseboard backup, but you're at around $10/ month for mid-winter freeze-up avoidance instead of $25.

And in the fall/spring it can be run to avoid needing the wood stove.  You may have to shade or turn them off (or inhibit the blowers with a line-voltage thermostat) to keep the place from overheating the place when the daily temperature averages are running near freezing.

But there are many factors that could limit the output to merely half that- a lot depends on snow cover winter insolation, outdoor temps during the active collection period, etc. But it doesn't cost a lot to experiment and scale it accordingly.  Start out with 4-5 meters of panel, measure it's performance with the power bill, add more panel as-needed.

This guy's ~5 meter design looks clean enough, and could be used to pump the heat into the crawlspace after you insulate the crawlspace walls and foundation.  The approach doesn't disrupt  anything on the main floor, and won't creat a chillling draft on the humans when running at low/very-low temp:

http://solarairheater.lampresource.com/

 

He's running it at higher temp/lower efficiency than you would be, using a single 20W blower  (powered with photovoltaic panels he added later, if you read his web page.)

Being a lateral-flow design it minimizes the potential for back-thermosiphoning losses.

still I cant help thinking these ideas are from a different climate. My place is in the great white north, where it's asking a lot to believe you can heat a house with solar at 20below and cloudy days all winter. I need to figure out a way to store the extreme heat of summer for winter. At one point it was suggested that a 16 foot box wouldn't store 10% of what I need. What about 10 16 ft boxes? or a 3*2 array of them that's 2 high so 12 boxes?

Posted By greendreams on 16 Feb 2010 08:38 AM
still I cant help thinking these ideas are from a different climate. My place is in the great white north, where it's asking a lot to believe you can heat a house with solar at 20below and cloudy days all winter. I need to figure out a way to store the extreme heat of summer for winter. At one point it was suggested that a 16 foot box wouldn't store 10% of what I need. What about 10 16 ft boxes? or a 3*2 array of them that's 2 high so 12 boxes?

We're not talking about "heating" exactly, only keeping it above freezing.

You're in the Okanogan, not Yukon.  Gary Reysa (the guy with the thermosiping air panel on his barn/shop) lives at 8k' in Montana- probably colder than you. Without knowing your exact location & altituded I'd hazard you're at or under 3000 base 18C heating degree days. (Jasper Alberta is ~2975 HDD is your place truly colder than that by a signficant degree?)  Your peak loads are high, but no higher than parts of Maine or New Brunswick, and your average isn't all that high (might be lower than mine). Find yourself on this map and give us a guesstimate.  Peak loads of 20below is only a factor how big the heat exchanger needs to be, not the total amount of energy storage.  Does it stay 20F below for days/weeks/months on end? (Days maybe, weeks/months, I doubt it.) For seasonal storage it's all about the average, not the peak.

And that map uses 18C as a base temp, whereas for mere freeze control purposes you can use 5-10C.

And, you are super-insulating the place.

And you're east of the coastal range, with easily twice the winter sun as foggy-dew Vancouver.

Bagging significant amounts of low-temp solar is much easier than storing a full season's heat. (The former is easy, the latter, not-so much.)

Can I buy you a calculator? It isn't hard math to come up with an order of magnitude.

Assume you'll get something on the order of 0.15BTU/lb/degree-F of storage out of random dirt,  but use the best case of compacted clay if you like, which is ~0.22BTU/lb/degree-F and assume a best-case density of ~110lbs/cubic foot.

A 16' cube then contains 4096cubic feet of clay, which is 450, 560lbs. Ten of them contains 4,505,600lbs.

So for every degree-F above any arbitrary setpoint you get 4,505,600 x 0.22 =   991,232BTUs of storage per degree F

If it's been taking $50/month x 3 months=$150 to keep the place above freezing for the season, at 7cents/kwh that's about 2143kwh/season which is 3412btus/kwh x 2143kwh=  8,235,569 BTUs/season

so 10 16' cubes would only have to be:

 8235,569/991232 =8F above the arbritrary setpoint.

But the temp it really needs to be depends on what delta-T on the heat exchangers in the dirt needs to be to retrieve the peak loads,  and how hot the home's radiation need to be to to deliver the load to the house. Even if we make the dubious assumption that it's only needs to be 5F delta for each heat exchange, that means you need to keep the dirt hot enough that at the end of the peak heating season the temp it 10F above your room air setpoint.   If you're trying to keep it at 5C/40F that means at the end (at depletion) the clay needs to be over must be 40+5+5=50F/10C, and at the beginning of the season it needs to be another 8Fwarmer, or 58F.

Then there's the small matter of the amount of insulation it would take to keep from losing it all to the surrounding dirt that's less than 10C. Which is a function of R value surface area and delta-T. To get the equivalent volume of 10 of your 16' cubes you're looking at a ~34.5' cube, which has ~414 square feet of surface area in contact with the (let's call it 10C) soil.  Three months is about 2200 hours, and if you want to keep the losses under 20% over those three months.  To tolerate a 20% loss you need to start with 2143/10= 214kwh or (214 x 3413 =) 730,382 more heat stored.

As a crude model lets assume an average storage temp of 15C/59F, and a 5C/9Fdelta against the surrounding soil.  You can only lose 730,382BTU/414sq.ft.=1764BTU per square foot over the season, which is a U-value of 1764/2000= 0.882BTU/hr/ft^2.

The R value that delivers that performance is 1/0.882= R1.1

That's a best-case scenario, with R1.1 or sides, bottom and top, buried deep enough that the average temp of the surrounding soil is close to the deep-well numbers.  5F cooler at the top of a shallowly buried cube results in a higher loss (= higher insulation requirement), but you could get there.  If you used crummier soil than densely packed clay you'd need roughly 100% more delta-T, between beginning and depletion temp, so call it a storage temp of 40F+10F+ (2x8F)= 76F, which gives you a beginning delta of 25F against the surrounding subsoil instead of 9F, nearly 4x the heat loss for a fixed R, so figure on at least R4 to have any hope of not running out of heat before the freeze risk is gone.


If you used your stacked boxes you'd have more surface area==y more loss==higher insulation requirements.

But storage looks do-able at 10x the volume in the box- got a good plan on how to stuff that much heat INTO the box, and retrieve it?

If instead of a 35' cube or boxes full of packed clay and heat exchangers (we haven't calculated heat exchanger sizes or estimated distribution losses yet mind you) and all of that other crap, by the time you've insulated the house to R40ish you wouldn't need to store ANY heat for longer than few days, and a very modest amount of solar would support the bulk of the load through all but the cloudiest of winter weeks. (It really would! DO THE MATH! )  Using water or concrete as thermal mass storage within an R40 structure buys you quite a bit of peak load buffering. 

And the heat load between 40F (indoor freeze control) and a -20F peak is only 60F delta- less than what it takes to heat a place for human comfort when it's 0F outside, easily doable with a single meter of thermal air panel in a modest sized R40 house when the sun is shining.

Seriously- insulate those windows with R10-R15 panels and seal up the place for winter- your electric bill will plummet.  Then we'll know how much storage or solar you REALLY need.

It kind of seems like it's going around in circles now. that's cause I keep moving between real world where I improve the house a lot and it just gets cheaper (bout 5 months the heat kicks in) and theory...where I want to be able to build things like greenhouses and keep them abouve 55f all winter. My idea right now is an underground box framed in 2 by 12's so r44 insulation. You're insistance that I do the math is going to make me build one box the first years and times it out. The soil can be good and hot like 300 degrees on the first day of fall. The box would be deep enough that water taken from the top could still go to the house at 5 feet below ground so we're talking 5+16 ft deep to start. Dunno your insistance is making me think that a better thing to test the first year is a single thermal solar panel. Let say this "if you were going to explore the underground heat option what would you do?"

Sitting outside yesterday looking at my freshly shoveled driveway I had a thought along these same lines. It was 25 & partly sunny with light snow as well. It got me to thinking, why not put in a geothermal field and lines under the driveway and simply recirculate the antifreeze between the two. My field is at 36, plenty warm enough to melt / sublimate snow. There would have to be some automation to not let it run 24x7 so when it's -10F and clear it wouldn't keep cooling the field for no reason.

I am guessing this is done somewhere because it just makes sense.

Posted By greendreams on 16 Feb 2010 02:13 PM
It kind of seems like it's going around in circles now. that's cause I keep moving between real world where I improve the house a lot and it just gets cheaper (bout 5 months the heat kicks in) and theory...where I want to be able to build things like greenhouses and keep them abouve 55f all winter. My idea right now is an underground box framed in 2 by 12's so r44 insulation. You're insistance that I do the math is going to make me build one box the first years and times it out. The soil can be good and hot like 300 degrees on the first day of fall. The box would be deep enough that water taken from the top could still go to the house at 5 feet below ground so we're talking 5+16 ft deep to start. Dunno your insistance is making me think that a better thing to test the first year is a single thermal solar panel. Let say this "if you were going to explore the underground heat option what would you do?"

Do a manual-J type heat loss calc on the house in it's finished state at 55F indoors, and -20F outdoors (a 75F delta-T) to size your heat exchangers.  If you know what temp you were keeping the place when burning up that 2000kwh for the winter, with some nearby-city weather data you can probably come up with a reasonable upper bound on how much energy it takes to keep it at 55F. 

When you're talking about smaller boxes, bear in mind that volume goes down with the cube of linear dimension, but the surface area goes down with the square, so at your 16' cube has 1536 feet of area to insulate, whereas your 35' cube with 10x the volume has ~7500 square feet to insulate, only 5x as much. So shrinking it will increase your R value requirements in two ways- higher temps, and more insulated area per unit volume.

I don't know how you expect to heat the dirt to 300F with any sort of efficiency (got any spent-fuel rods from the nukes down-river to bury in there? ) With solar technology you'll be crapping out  on summertime efficiency before 200F (and above 200F you have to have to run it at high pressure to keep the steam-explosion potential under control.  Heat pumps? Maybe... but that would be an expensive hardware proposition fraught with technical difficulties as well.

The kind of  "...underground heat option..." that makes sense to ME is natures-own: Skip the insulated box, drill a well and install a heat pump (geothermal).  It's just too hard to make the numbers work in any practical sort of way on dirt/water buried storage at modest temps- the size just has to be gia-NORMOUS to get the temps low enough not lose it all by January.  But super-insulating the building is proven, well-modeled, and designable , and can be made to work in southern B.C. without spending a quarter-million on the project or anything like it.  Do a careful analysis of project like this, and see what parts can be factored into your place at reasonable cost.  Properly applied it'll be a lot less insulation than you'd need for a seasonal storage concept, lowering the load to where even the weaker mid-winter solar can carry a significant fraction.

If the heating season is currently 5 months, a single 5 meter thermal panel will likely reduce that by roughly a month, maybe more at the shoulder seasons with the house in it's current state. If you super-insulate the house to PassiveHouse standard  the heating "season" can be reduced to pretty much "cloudy mid-winter weeks only".  It won't provide enough heat to heat a greenhouse to 55F all winter, but that's just part of the "real world" problem, eh? ;-)   Greenhouses are inherently lossy, and it'll always take a lot to heat them.

You may find this climate data for Pendicton useful, assuming you're at a similar altitude.  For base 18C it looks like you get about a 3400 heating degree-day climate, but if you use base 10C (reasonable, for freeze control) you're at only 1477 HDD. 

Using base 0C (absolute min freeze control) it's only 239 HDD, 170 of which occur in December-January- fully 2/3 of your real freeze risk in a better insulated house occurs during these two months, and the rest of the time earth coupling alone no solar, just R40 down to the footings and R15 over the windows) would carry you through, except for the random 3-day cold-snap.

Assuming a whole bunch of stuff like:

Your $50/month power bills occur during December and January, and that you have it set to 10C inside,

That works out to about 350HDD (base 10C) during those months,

So assuming again...

Your rates are 7cents/kwh,

it means your freeze-control heat load is currently adding up to ~700kwh/350HDD, or 2kwh/HDD(Celcius) which is 1.1kwh/HDD (Fahrenheit).

That's about 0.046kwh/heating degree-HOUR, so

assuming (again) the peak load is at -20F, with an interior temp of +10C (50F) that works out to (50+20) x 0.046= 3.25kw (=11KBTU/hr), which isn't bad (not a humongous heat exchanger)

But if you cut that in half or by 2/3s with insulation alone it's a whole lot smaller as is the average bill.

From the climate data, the average daily temp in January is -1.7C(29F), and in December it's -1.1C(30F), which is a considerably warmer average than I get here in central MA (like 3-6F warmer!), so while your peak loads are higher than mine, they're not often very long in duration. (Clear nights go down to -20F, but it comes up to 0-10F by the afternoon of the same day during a cold snap?)  This is where a bunch of thermal mass inside the thermal envelope of the house &/or earth coupling saves the day- the peak loads are supported by the thermal mass, which averages out the daily load for you.

Unfortunately December & January are also your cloudiest months, with well under half the possible daylight hours in bright sun. You may need a bit more thermal air panel to get there, but they do work in bright clouds (albeit at much reduced output.)  Spend the money on insulated window panels and insulating the crawlspace walls for earth-coupling first- your solar output when it counts the most is going to be small.

It would appear that -20C (-4F) is a more typical cold-snap low than -20F, if you look at the number of days of -30C (-22F), it's coming up zero, and only fractional days for -20C during the months of December & January.  February & November are nearly 2x as sunny, and March & October get 4x as much sunlight. This means a panel sized to handle freeze-control load in January would put out enough to be the primary heat source for human comfort levels in March & November and parts of February & December.

the ranch is close to lumby east of vernon. It's much higher than places on the lakes like kelowna and penticton and doesn't have the moderating effect of a big body of water close by. So ya 20 below celsius is normal and the cloudy days are real but so far if it's like $100 a year to heat there is no payback in my battery designs.

Posted By greendreams on 16 Feb 2010 09:52 PM
the ranch is close to lumby east of vernon. It's much higher than places on the lakes like kelowna and penticton and doesn't have the moderating effect of a big body of water close by. So ya 20 below celsius is normal and the cloudy days are real but so far if it's like $100 a year to heat there is no payback in my battery designs.

Payback? 

If it all came down to the raw net-present-value of the money invested, homes would barely be built to code. But superinsulating to the point where the up-front cost and operating costs of heating/cooling systems drops to near-zero, the reduced utility costs can sometimes more than offset the increased mortgage.

Something as exotic as a storage system designed to squirrel away thermal energy as heat to be used months later will never pay back.  Storage for a week, maybe, but even then the payback would be in decades not years.  Storage as fuel, well... (biofuels have serious solar conversion efficiency problems, even under best-case scenarios.)   But insulating panels for the windows would have a very rapid return on investment.  DIY thermal air panels may have low output during peak load periods for you, but probably also have a reasonable payback (sub 5-years.)  Super-insulating to R-40, even longer.  The fact that you have some of the lowest electricity pricing in N. America (!) makes the payback on any heating related investment far longer than say, someone who is heating with propane with a 70% efficient wall-furnace, or oil boiler, etc.

But tighter, more insulated buildings also have "payback" in comfort, not just dollars.  Sealing & insulating the crawlspace will make the floor warmer (especially if you then use a thermal air panel to heat the crawlspace) insulating the walls to high-R makes the interior surface of the walls warmer, resulting in lower radiative absorption, and you FEEL warmer at the same indoor air temps than you would otherwise, etc.

Lumby's elevation is  ~500m, compared to Pendicton & Kelowna's ~350m, so you can probably add 100-150HDD celcius to Pendicton's numbers and hit pretty close to Lumby's seasonal heat load.  (Just subtract 1-1.5C from the daily averages for 100 days of heating season. 1 degree for elevation shift, 0.5 degrees for the lake's moderating effect.)  The -20C peak heating loads will tend to occur on clear nights, but those nights are also more likely to be preceded &/or followed by the sunnier days, so a solar thermal panel may moderate the peak day kwh usage, but it's the average load that shows up on the electric bill.

I'm up in the hills at 750 metres.Great exposure, not too much heat. Good wind. I need to explore storing summer heat for winter. If ice houses work, and the ice gains extra staying power because it's ice not water, perhaps storing heat in something like molten salt is the answer..

btw: If the first thing I do is say 2000 gallons of water in the basement is there a real benefit to storing in a tank under pressure or can it be a hot tub..?

check

hmm its impossible to uncheck he box to receive email notifications for this topic.

Posted By greendreams on 22 Feb 2010 11:10 AM
I'm up in the hills at 750 metres.Great exposure, not too much heat. Good wind. I need to explore storing summer heat for winter. If ice houses work, and the ice gains extra staying power because it's ice not water, perhaps storing heat in something like molten salt is the answer..

btw: If the first thing I do is say 2000 gallons of water in the basement is there a real benefit to storing in a tank under pressure or can it be a hot tub..?

Phase change taking advantage of materials with high heat-of fusions makes for much more compact storage.  Unfortunately most phase change materials that with useful heat-of-fusion capacity at the temps you require are lot more expensive than water, and you'll still need lots of it.  The most commonly used in space heating apps is sodium sulfate decahydrate, (aka "Glauber's Salt") which has a heat of fusion of ~250kiloJoules/kilogram, and melts at around 32C, making it very efficient charging with low-temp solar, and getting into the room with normal amounts of low-temp radiation (like a radiant floor or something.) A kiloJoule is about 0.94BTU, so you get about 235BTU/kg, so it takes about a half-ton of it to store a 1 therm==29.3/kwh.  With 100 tons of it you can do a lot, but at what price? 

But you'd still need significant insulation to store 32C Glauber's Salt for months when surrounded by 10C soil. A primary problem with months-long storage issue for anything but very modest storage temp is that R value is denominated in energy loss per degree difference per unit area per HOUR.  A 720 hours/month it takes takes a lot of R to hold up against a big delta-T for even one month, let alone months of coasting time between September when your solar inputs are falling fast and late-November into December & January when you need to start retrieving significant amounts of heat and the solar gain is fraction of the load.

There are no energy benefits for keeping your thermal mass storage water under pressure.  A hot tub or large septic tank at atmospheric pressure would be fine for thermal storage. Only if you're pumping that water with oxygen-susceptible pumps (iron impeller circulation pumps) would pressurization be called for.