Thoughts On "Solar Water Heating: A Comprehensive Guide..." by Ramlow & Nusz

JUGHNE

This is our version of a passive solar heated house, we wanted something to look in place in the area but still take advantage of solar heat.

This picture was taken June 21, the summer solstice, you can see all the major south windows are shaded from sun. On Dec 21, the winter solstice, all glass is exposed to the sun. The sun reached about 17' into the 20' sun room.

That room is a 6" pour with ceramic tile installed, also about 1000' of tubing in the concrete. There is about 4' of sand under that. The outside foundation walls are insulated down to the footings.

The only difference in the structure is that the trusses are designed so that the inside 8' ceiling and the outside soffit ceiling are at the same level. There is a 30" overhang that provides the shade in the summer but allows wintertime penetration.

This 30" overhang was calculated and drawn out on graph paper based on our latitude, 42-43 degrees, and also the bottom of the windows. The first floor windows and patio Frenchwood door were placed under the standard 12" headers as usual.

If you look closely there is about 12-14" of wall above the windows, the result of truss design.
Most ranch houses here have only 2-3" of wall above the windows on the outside.

We started planning this in the early 90's and constructed in 1995.

We studied "The Passive Solar Energy Book" by Edward Mazria, published by Rodale Press in 1979. (this really dates me, doesn't it),
687 pages, packed with info.......old book but the sun is still working as it was then.

For passive cooling all rooms have small north windows that allow the summer breeze to flow thru the house. We did not get the AC until 2000. Something about aging changes things for the fairer sex.

Jamie Hall

Nice house! I like it!

jumper

MIT Technology Review has article about storing heat in bricks.
Nothing new in my opinion. Decades ago English used electricity during low demand to store heat in ceramic room heaters. When one desires warmth she'd switch on a fan I presume. Nowadays we can use insulated louvers maybe.

This approach sounds like the simplest for OP. She can buy electric stone heaters made for sauna and add fans.

desert_sasquatch

hot_rod said:

Really no need to measure your self, NASA has been doing that for you, probably with more suitable meters, and over 20 years or more.
The SRCC shows collector performance under various conditions.

OK, good to know that the SRCC categories are accurate. The thing that confuses me is that when I plug it all into the spreadsheet it functions much better than it did in the TSOL modeling software.

In other words if the SRCC "cloudy/partly cloudy/sunny" categories are roughly accurate then I can plug that into the weather data I have which also tells me how many days of cloudy/partly cloudy/sunny weather I should expect each month. The thing is, when I do that it winds up being vastly different from the TSOL results. Here's an example:

Where I am the average January high is 35 F and the average low is 7 F. Since the panels are used during the 5 or so warmest hours of the day (in January, obviously they're used for more hours in the summer) I'm going to assume that the average temperature when the panels are in use is about 32 degrees. Assuming I'm heating water to 122 degrees F that's a delta T of 90 F.

Looking at the SRCC specs for the SunEarth Empire 40-1.5 at delta T of 90 F I should get 24.8/13.0/2.8 kBtu/day on a sunny/partly cloudy/cloudy day, of which in January we get 13/10/8 respectively.

For the sake of making this simpler, I'll assume I put this heat in a big tank of water that not only is insulated itself, but is within the building envelope so that a large portion of the heat lost from the tank is given to the building and it effectively acts as a kind of space heating.

This is a ballpark because if I made it more complex I'd just be pointing to "my spreadsheet" and everyone would be wondering--legitimately--whether I'd just screwed up the math somewhere. I say this now because obviously the temperature of the storage tank--which changes hour-to-hour--would impact the efficiency of the panels. But I think 120 is a fine middle ground. If the tank is big enough that one winter day's heat is stored in a 10 F temperature change then a decent portion of the time the tank will be below 120 and on a properly sized system it'll only get above that occasionally in the winter. DHW would be somewhere near 120 so...close enough?

So the total Btu per average January from one panel should be about 24.8*13+13*10+2.8*8=474.8 kBtu. That needs to be adjusted for the altitude (adjusts it upwards because I'm two miles above sea level) and for the time of year (adjusts it down because there fewer hours of daylight than most of the year--and I'm assuming SRCC trials use an "average" amount of sunlight).

The altitude adjustment, I think, is +8% irradiance for each 1000 meters above sea level. [https://www.sciencedirect.com/science/article/abs/pii/S1011134496000188]. So that's +20%.

There are about ten hours of daylight in the winter and 12 hours near the equinox. I know not all of those hours are usable so if it's six hours of usable daylight in the winter and eight hours near the equinox then that's 6/8= 75%. In other words -25%.

0.75*1.2=0.9. So in January the output of a panel would be more like 427 kBtu I think.

As per BEopt my heating needs (space & hot water) in January should be about 8,184 kBtu. Looking at the chart from BEopt, I probably need about 80-85% of that to achieve a year-round solar fraction of 90% (it's obviously a bit subjective when doing back-of-the-napkin because I don't know with precision how well my storage tank will smooth out the extra heat from the sunny days).

So lets say I need enough panels to produce about 6956 kBtu in January to achieve the 90% solar fraction. It *seems* like I should be able to do that with about 16 panels. And for what it's worth my more detailed spreadsheet said much the same.

So here's where I get confused: That's not what TSOL tells me. Here's the graph from TSOL:



As you can see, TSOL disagrees. It thinks that 16 panels would give me maybe a 62% solar fraction over the whole year. And what's more, the curve of the graph appears to be a...I don't know the exactly word, but basically an equation in the form of something like y=x^1/2. The slope decreases as x increases, indicating--I think--that the problem they perceive might be some very cold or perhaps snowy days when the panels produce no heat. Of course, that's what the heat bank is supposed to help with (and the TSOL model did have a 2000 gallon tank in it) so I'm left scratching my head...

Was their heat bank just not insulated well, and thus unable to store heat for several days? I can't think of another reason why I'd be so far off.

I know I've asked something like this before but I come back to it because if the above calculations are right then solar thermal costs about the same as solar PV for off-grid heat, but takes up less room. But if TSOL is right then I think solar PV+resistance heaters wind up being cheaper, particularly for producing winter heat.

Sorry for the long post. But I had thought that the lux meter explained why my results were different than TSOL...but it seems that's not it. So I'm curious if anyone can explain why TSOL is giving me such a different result than the one I expect based on SRCC numbers.

hot_rod

I don’t know that T Sol uses the SRCC data in their program? That could be some of the difference. They might use performance data from a similar size collector tested in Germany?

Looks to me like the graph shows fraction going up as you add collectors?

I see you have been to the T Sol support forum already, that is probably the best place to have your inputs checked. You need someone familiar with the program to help with the inputs.
Basically the collector type, orientation, location, number of collectors and the load, In your case the storage tank.
T Sol had a support gal in So Cal that you could send your design to and have her confirm the inputs were correct, then chat on a phone call. I think the factory support comes along with the purchase of a license. If you are using the demo version, the online forum may be you best place to get assistance.


RET Screen is a free simulation program out of Resources Canada, they had good, free tech support. Its a large energy modeling program, takes some computer space when you download it 😗
Might watch the tutorial at that site.

desert_sasquatch

Thanks @jumper interesting article. Here it is for anyone else who wants to read it

https://www.technologyreview.com/2023/04/10/1071208/the-hottest-new-climate-technology-is-bricks/?truid=*|LINKID|*&utm_source=the_download&utm_medium=email&utm_campaign=the_download.unpaid.engagement&utm_term=*|SUBCLASS|*&utm_content=*|DATE:m-d-Y|*

It's an interesting idea to directly heat up bricks and have an insulated, louvered opening with which to let the heat out into the air. Rather like a stove but without the burning of wood and the pollution and work that implies.

That said...just thinking out loud, I can see some issues with using this on smaller (household) scale: They were talking about heating the bricks to 1,800 F sometimes which seems like it would be a fire risk (what happens if there's an earthquake?). I'd be afraid of heating anything past about 400 F (ie keeping it just below the temperature that wood will combust). Plus there's the issue of insulation.

Just some back-of-the-napkin math...I was thinking of using a cylindrical gravel pit heat bank with a diameter of 16'. The gravel would be 5' tall, the insulation would make the whole thing taller. The idea there was that it might be heated to 155 degrees F at most--something under 180 F, which I gather would melt the PEX tubing. So lets call that 75 F of useful heat. Increasing the max temperature up to 400 would add another 245 F of useful heat--a more than 4X increase. Which would mean I could use about 1/4 of the space for the gravel...however I'd also have to double the insulation four times (in other words, increase the thickness by a factor of 2^4=16) in order to keep the heat loss the same. Honestly I could probably get away with half that. So I'd just have to increase the depth of insulation by 8X. As I was already counting on using R-30 this would be R-240, which if my insulation is rockwool with R-3.65 per inch would be 5.5 feet thick.

Aaaanyhow. I can see why folks are talking about using something like this on an industrial scale. The bigger your heat bank the less you pay for insulation per unit heat storage. I hope they succeed in getting this to work for some of the heavy industries that currently require coal to do their high-temperature metallurgy (or whatever they're doing).

But given the price of space within a home, when I crunched the numbers I just didn't see a way to make this the most economical option...

High temperatures would also require more insulation, or possibly some kind of specialized vacuum enclosure (I'm talking out of my ****, no idea if anyone does that). All of that would be easier on a massive scale.

Jamie Hall

Heating up bricks? Or any other mass? Sure. That's one of the things that makes passive solar work. If you want to make it explicit, ever heard of a Trombe wall? Maybe look it up?

It's hardly a novel concept. Using mass in one way or another to reduce temperature swings in an occupied space has been done -- with varying degrees of success -- for at least the last four or five thousand years.

jumper

If you're worried about high temperature or insulation you can put heat storage media in ground. The problems I see are how to heat rock and how to extract heat from rock.
How does electric sauna heater work? To extract heat I guess you can use a blower with high temperature safety. Now that photovoltaic is affordable you can finally store sufficient heat. Decision is between high temperature water or very high temperature solid.
Solar thermal especially with storage will be a pita. Regard the issues in this forum.

desert_sasquatch

Thanks @hot_rod

OK, I spent some time familiarizing myself with RETscreen. It's definitely giving me a different result than TSOL. In fact, it seems to be giving me better results than my spreadsheet or back-of-the-napkin math had anticipated. As far as I can tell it's saying that one panel in January produces about 23 kBTU per day--which according to the SRCC sheet for this panel would be about the production on a cold, sunny day. Except I know that more than half the days in January are partly or fully cloudy. Which according to the SRCC should cut the output on those days by 50% or more.

So I'm a bit puzzled. Perhaps my ideal-for-winter angle of the panel did this? Or perhaps there's user error somewhere. I'd like to ask the RETscreen folks about it. Any idea where I can find their tech support? I've sniffed around their website but didn't see any forum or any phone number more specific than the general one for national resources of Canada...

Speaking of user error...I have a new guess for why TSOL gave me weird results. I may have set the azimuth incorrectly, pointing the panel due north rather than due south. I can't say for sure since as I said I'm past the 30 day trial but I think it could result in the error I saw, and I did it on RETscreen before I caught it.

desert_sasquatch

@jumper I admit that I'm taking another look at solar PV. But for the sake of keeping this on topic(ish) I'll address that in another thread when I've organized my thoughts.

Can you say more about the "storing heat in the ground" thing? I'm not sure I understand.

hot_rod

Maybe just fill out the “contact” form. It’s not alway easy getting in touch with a government person.
I took a couple of their online courses and had a contact for solar modeling help. Years ago🥸

To load any storage in the cold of winter the array has to be sized large enough to cover the load that day and have excess to load into the tank. Unless you had a large enough storage system to load in non heating months.

If that 8 unit apt in Switzerland required 54,000 gallons, 1/8 th of that is 6700 gallons, and as you can see the collectors cover the entire roof.

I visited a couple solar thermal district systems in Italy once, same massive tank and collectors on a large energy building. PV on all the various apt buildings. A couple oil fired Viessmann boilers for backup.

Jamie Hall

We used to have a saying: KISS. It's still applicable. You need enough power input to satisfy current needs and your storage. That's your collector area, straight if it's solar thermal, multiply by five for PV. Then you need to figure out how much energy you need to store. There are simple ways to store energy. There are complicated ones. Then you need to get the energy to where you are using it. There are simple ways to do that and there are complicated ones.

Then you need to sell the client. Since solar is regarded as new and whizzy (never mind that it's been being done for a few thousand years) that means much arm waving and magical incantations, and fancy machinery and visible collectors.

KISS.

desert_sasquatch

@hot_rod hah wow that was just an 8-unit apartment? I somehow though it was much larger...

desert_sasquatch

OK, I think I figured out the mystery with RETScreen. Barring the presence of some portion of the software that I've not seen--not an impossible thing and I have contacted them--I think the problem was that the program isn't set up to look at solar thermal for space heating. Just for water heating.

The problem there is that with domestic hot water the solar thermal panel gets cold water coming in at maybe 40 F. Whereas with a closed loop solar thermal that's used for space heating the water would usually be above 95 F--often tens of degrees more.

Point is, using cold water as the input makes for a lower delta T between the water in the panel and the ambient temperature outside than would actually occur during much of the time that water was being heated for space heating. And indeed, when I raised the minimum input water temperature from about 40 to about 85, the output of the panels in the program dropped by about half...at least in the winter. Which was much closer to what I thought the panels should be putting out based on the SRCC data.

I could be wrong about this but it's the best guess I've got.

hot_rod

So the key to solar thermal loss is the ambient. Wind also rips away at that energy.

Hard to believe an uninsulated pool collector on a roof can approach 90% efficiency! But look at the operating condition on this graph comparing 3 types of collectors. Ambient and operating temperature are close on the top of the pool collector slope.

Also surprising is how flat panels can out perform evac tubes, down to that cross over. This has to do with the double layer of glass around this evac tube. It works against you at warm conditions, but as ambient drops, the double wall and vacuum becomes more important.

The data and formula to compare efficient at operating vs ambient.

You will never beat mother nature, at some point, given the right condition, she will grab all your energy, pressure, humidity, etc.

desert_sasquatch

DC Zone Valves:

Ramlow and Nusz say that zone valves, which are used to regulate the flow of hot water to whatever heating zone requires it, generally run on 24V AC power.

First of all...it seems like manifolds with valve actuators are more popular when there are more than two zones, right?

But second of all: 24V AC seems like a weird configuration. Is that just what the controllers put out?

In any case, since I'd like my heating system to work even when grid power is down, I'd really like the zone valves or manifold valve actuators to be able to operate off of a battery. Ideally I'd like them to operate on DC power--just to avoid needing an inverter. Is there such a thing? Or am I chasing a unicorn here?

It's not the biggest deal in the world and in all likelihood I'm missing a great deal of context but I thought I'd ask.

Jamie Hall

24 Volts AC is an odd sort of choice, but it's been the standard for most controls for decades, and 24 Volts AC or DC is standard on most large trucks and aircraft -- and now some cars.

The latter being the case, odds are you can find some solenoid valves intended for trucks which operate on 24 volts DC. You'd have to fabricate your own manifolds, but that's not hard. If you are planning on using conventional two terminal battery powered thermostats, no problem. If you are looking to run a "smart" thermostat -- then you are going to need an inverter for 24 VAC.

hot_rod

AutoMag still offers DC zone valves, the AA series.
. They may be high current draw, check that out.
The new 24VAC zone actuators are like Caleffi 6564 are 250mA in rush 125mA hold current

https://www.automagzonevalves.com/_files/ugd/39dc32_b762338b58e940979c33b30fc2f1c2fe.pdf

jumper

Methinks OP is over thinking. Solar is problematic enough without extra complications.

hot_rod

Thermostatic radiator valves work without any power requirements 

Oventrop Uni Box thermostatic valves can be used to control radiant zones or manifolds without electricity

desert_sasquatch

Interesting idea @hot_rod. Do you think something like that would provide a fairly even temperature control if it was mounted in an appropriate place? In other words would you expect them to operate about a smoothly as the zone valve plus thermometer you mentioned earlier?

desert_sasquatch

What do people think (or is there a place I can read about) the relative costs and benefits of using a manifold with actuators vs a manifold with zone valves vs a separate circulator for each zone?

(Er, and while I'm at it, is "actuator" just another name for "zone valve"?)

I ran into a youtube video of a guy saying that having a separate circulator pump (plus a check valve) for each zone was better than using zone valves. What do yall think? That would also avoid the 24 VAC issue...at least if I got DC pumps so they can run off the battery (which I Intend to).

Jamie Hall

Properly plumbed and connected, either zone valves or zone pumps work about as well as the other. Note: Properly plumbed and connected.

hot_rod

desert_sasquatch said:

Interesting idea @hot_rod. Do you think something like that would provide a fairly even temperature control if it was mounted in an appropriate place? In other words would you expect them to operate about a smoothly as the zone valve plus thermometer you mentioned earlier?

Thermostatic valves provide excellent control and comfort. They are proportional valves compared to zone valves or actuators that are on/ off controls.

hot_rod

desert_sasquatch said:

What do people think (or is there a place I can read about) the relative costs and benefits of using a manifold with actuators vs a manifold with zone valves vs a separate circulator for each zone?

(Er, and while I'm at it, is "actuator" just another name for "zone valve"?)

I ran into a youtube video of a guy saying that having a separate circulator pump (plus a check valve) for each zone was better than using zone valves. What do yall think? That would also avoid the 24 VAC issue...at least if I got DC pumps so they can run off the battery (which I Intend to).

I’d venture to guess the majority of zone pumped systems are over pumped, rare to see a balance valve on them, so who knows exactly what flow they are providing. 80w pumping moving 2-3 gpm isn’t great design in my opinion, they are running maybe 10-20% wire to water efficiency!
Newer ECM have more adjustability, so a better zone flow match up, sometimes.

Is the total flow is 12 gpm or less you are well within the range of a single pump with TRV or zone valves

An ECM circ running 30-40 watts with thermostatic valves would be the most energy efficient system, and the best comfort control.

Anyone telling you that zone circs are better than zone valves, without knowing the application , or explaining why??

Maybe you are doing too much internet surfing? Getting too much expert advise😳

desert_sasquatch

OK, so I finally made it to the part about a thermal sand bed. They say that they can usually store several months worth heat in their mass. But I've done some math and I just don't see it. The gravel bed I'm envisioning would store maybe four or five days worth of heat, the equivalent of 2000 gallons of water. And it's not that much smaller than the sand bed they're talking about.

For anyone interested, here's the math I did:

They give an example of a 1,200 square foot home. They figure the sandbed is two feet deep--about what I'm envisioning for my gravel. So here's the math on that as I understand it, and I'll also do the math for water so we can get a feel for the "gallon equivalent" of their sandbed.

They say that compacted sand weights 105 lbs per cubic foot, which is in line with the 80-100 lbs per cubic foot estimate in engineer's toolbox for non-compacted sand. One cubic foot of sand weighs 105 lbs which is 47.6 kg.

One cubic foot of water is 6.22 gallons. One gallon of water weighs 3.785411784 kg. So one cubic foot of water weighs 23.545 kg.

The specific heat of sand is...probably around 800 joules per kg degree K (J/(kg*K)). The specific heat of water is 4182 J/(kg*K).

So one cubic foot of sand whose temperature is raised 1 degree Kelvin has absorbed 47.6*800=38,080 joules of heat. One cubic foot of water whose temperature is raised 1 degree Kelvin has absorbed 23.545*4182=98,465 joules of heat.

2000 gallons of water, then, would hold 2000/6.22*98,465=31.66 megajoules per degree kelvin.

2,400 cubic feet of sand would hold 2,400*38,080=91.392 megajoules per degree kelvin. In other words, his proposed sandbed is the equivalent of about 6,000 gallons of water. Just for comparison's sake.

Now on to just looking at sand:

BEOpt told me that my total yearly space and DHW heating needs were about 57 MBTU's. 1 btu is about 1055 joules so that's about 60 gigajoules. The space heating part is less, perhaps 35 gigajoules. Most of that happens in the six coldest months, perhaps 30 gigajoules. I know that averages here are deceiving because the coldest months require a lot more heat than the other months but three months of winter heat would be over 15 gigajoules for sure. Assuming I heat the sand to 150 degrees F and can use the heat until it drops to 85 degrees F I'd need 15,000,000,000/(38,080*((150-85)*5/9)=10,908 cubic feet of sand. If I've got a 1,200 square foot home with 2 feet of sand underneath that's 2,400 cubic feet. And that's not taking into account thermal losses, which WILL occur over the floor area.

So...huh. While I've made some generous assumptions about which months he's talking about I can see how, in a climate with a mild winter, his assertion could be accurate. But that said...I thought he was in Minnesota.

Plus of course there's the issue that his design doesn't have any insulation on the top of it--the sand is in direct contact with the cement slab. If he heated his slab up to 150 his home would get hot. The heat wouldn't keep.

In summary, I think I agree with the folks thus far who have said that seasonal heat storage is just not practical. Maybe 5 days or a week or more but not months. If someone wants to defend Ramlow and Nusz I'd be curious to hear their thoughts, though.

desert_sasquatch

Thanks again @hot_rod. We'll see what my guy is comfortable doing but I'll suggest the single ECM pump with TRV's or zone valves.

hot_rod said:

Maybe you are doing too much internet surfing? Getting too much expert advise😳

No such thing if you can figure out what experts to listen to

hot_rod

desert_sasquatch said:

OK, so I finally made it to the part about a thermal sand bed. They say that they can usually store several months worth heat in their mass. But I've done some math and I just don't see it. The gravel bed I'm envisioning would store maybe four or five days worth of heat, the equivalent of 2000 gallons of water. And it's not that much smaller than the sand bed they're talking about.

For anyone interested, here's the math I did:

They give an example of a 1,200 square foot home. They figure the sandbed is two feet deep--about what I'm envisioning for my gravel. So here's the math on that as I understand it, and I'll also do the math for water so we can get a feel for the "gallon equivalent" of their sandbed.

They say that compacted sand weights 105 lbs per cubic foot, which is in line with the 80-100 lbs per cubic foot estimate in engineer's toolbox for non-compacted sand. One cubic foot of sand weighs 105 lbs which is 47.6 kg.

One cubic foot of water is 6.22 gallons. One gallon of water weighs 3.785411784 kg. So one cubic foot of water weighs 23.545 kg.

The specific heat of sand is...probably around 800 joules per kg degree K (J/(kg*K)). The specific heat of water is 4182 J/(kg*K).

So one cubic foot of sand whose temperature is raised 1 degree Kelvin has absorbed 47.6*800=38,080 joules of heat. One cubic foot of water whose temperature is raised 1 degree Kelvin has absorbed 23.545*4182=98,465 joules of heat.

2000 gallons of water, then, would hold 2000/6.22*98,465=31.66 megajoules per degree kelvin.

2,400 cubic feet of sand would hold 2,400*38,080=91.392 megajoules per degree kelvin. In other words, his proposed sandbed is the equivalent of about 6,000 gallons of water. Just for comparison's sake.

Now on to just looking at sand:

BEOpt told me that my total yearly space and DHW heating needs were about 57 MBTU's. 1 btu is about 1055 joules so that's about 60 gigajoules. The space heating part is less, perhaps 35 gigajoules. Most of that happens in the six coldest months, perhaps 30 gigajoules. I know that averages here are deceiving because the coldest months require a lot more heat than the other months but three months of winter heat would be over 15 gigajoules for sure. Assuming I heat the sand to 150 degrees F and can use the heat until it drops to 85 degrees F I'd need 15,000,000,000/(38,080*((150-85)*5/9)=10,908 cubic feet of sand. If I've got a 1,200 square foot home with 2 feet of sand underneath that's 2,400 cubic feet. And that's not taking into account thermal losses, which WILL occur over the floor area.

So...huh. While I've made some generous assumptions about which months he's talking about I can see how, in a climate with a mild winter, his assertion could be accurate. But that said...I thought he was in Minnesota.

Plus of course there's the issue that his design doesn't have any insulation on the top of it--the sand is in direct contact with the cement slab. If he heated his slab up to 150 his home would get hot. The heat wouldn't keep.

In summary, I think I agree with the folks thus far who have said that seasonal heat storage is just not practical. Maybe 5 days or a week or more but not months. If someone wants to defend Ramlow and Nusz I'd be curious to hear their thoughts, though.

Others have come to that conclusion, without all joules thrown in😁

Ramlow lost me with the “judicious opening and closing of windows” to help manage late fall overheating issues. I suppose if you are a stay at home mom or dad, that may work.
I applauded his approach to simple affordable thermal storage, it’s just not for the masses.

https://digginginthedriftless.com/2011/12/20/why-we-are-not-using-a-sand-bed-to-store-thermal-heat/

jumper

Unfortunately most sunshine energy is available when not needed. Regarding your calculations PV can heat sand or ceramic to 700°.
You really should contact Indeeco who worked on this stuff decades ago. Perhaps they still have the info.