Drainback solar water heat exchanger design options

[semipro]

I'm starting to scrounge parts for a solar water heating system and need some help thinking through my options. I plan to use an open tank design for heat storage and exchange and came up with a 3 basic alternatives (see diagram). All tanks are non-pressurized and use pressurized tubing for domestic water and heat transfer. The preheated water would then flow to our GeoSpring heat pump water heater (HPWH).

Option A - Single tank design - Bigger tanks are harder to find and move. Simpler overall design and plumbing would be minimized. Tank heat loss would be minimized.

Option B - Multiple tanks in series with hot water from collector entering near where cold supply water enters. It looks to me like this option is pretty much equivalent to Option A and both tanks would be roughly the same temperature. More plumbing, more heat loss.

Option C - Multiple tanks in series with hot water from the solar collector entering near where preheated water flows out to the HPWH. It looks to me like option would result in different tank temps and possibly less overall efficiency since temperature exchange rates are directly dependent upon temperature differential. However, having storage tanks at different temps might be an advantage for other applications such as hydronic heating. The bad: more plumbing, more heat loss, potential heat transfer efficiency loss.

I guess another option is multiple tanks in parallel. But that seems pretty straight forward to me. The only difference between that and a single tank would be increased heat loss and more plumbing to do, neither an advantage.

Thoughts?

 

[pdf27]

Option C - Multiple tanks in series with hot water from the solar collector entering near where preheated water flows out to the HPWH. It looks to me like option would result in different tank temps and possibly less overall efficiency since temperature exchange rates are directly dependent upon temperature differential.

Other way around - when the hot water from the collector leaves level with the heated water leaving the tank, it can never be cooler than the water leaving the tank, even with 100% efficient heat exchange. When it goes in reverse flow like this, then you will still be getting some heat transfer from it to the hottest water, heating it still further, while as it goes down into the cooler water in the tank it will still transfer some heat, leaving at a little over the feed temperature.
What you're thinking of is the rate of heat flow per unit area, which is indeed dependent on temperature difference. Since you aren't area limited in this sort of heat exchanger, then you're better off going with a reverse flow heat exchanger like you describe here, which will extract far more heat from your coolant, thus extracting more heat from your solar system.

 

[woodgeek]

Go with one big tank and 'use' stratification....this is the way commercial units work. Put the heat input coil and cold water in near the bottom, and take the DHW off the top. B/c of stratification, it will act like a first-in first out buffer, rather than a mixed homogeneous tank. As far as heat xfer is concerned, make the coil big enough for it not to be a problem.

 

[semipro]

Thanks for the responses.

WoodGeek,
Interesting, I would have thought that less stratification and more movement of water within the tank would be more efficient due to the breakdown of boundary layers at the tubing surface allowing better heat transfer. So you're saying just add more tubing to compensate for that?

 

[GaryGary]

Hi,
This is one option that is pretty close to your option A:
http://www.builditsolar.com/Projects/SpaceHeating/DHWplusSpace/Main.htm
You can just leave the space heating parts out if you don't want to do any space heating.

Typically the systems take the colder water from the bottom of the tank for the supply to the collectors -- this makes the collectors operate more efficiently as there average temperature is lower and they lose less heat out the collector glazing. The return from the the collector goes to the top of the tank -- it actually has to be returned to the airspace above the tank water level, or the system won't drain correctly -- air has to be able to make its way up the return line for drain backs to work.

You might get some stratification with this setup, and that's good. But, a lot of drain backs that are set up this way lose all their stratification not long after the collector pump starts up in the morning. The fluid flowing out of and into the tank just mixes it up. You may be able to encourage some stratification with a baffle or screen where the return comes back in that keep the return water from stirring things up as much. Personally, I think the benefit stratification is over rated.

The heat exchanger in this scheme is just a 300 ft coil of 1 inch diameter pex that sits in the top third of the tank. The street water makes on pass through it on the way to the hot water heater. Even though pex does not conduct heat so well, this scheme works fine because there is so much surface area, and because the coil itself stores about 10 gallons of 100% preheated water -- most hot water draws come from the 10 gallons.

I'm wondering if you will get a good return from doing both the heat pump water heater and solar water heating? Our solar water heater is a bit oversized by conventional guidelines, but it supplies nearly 100% of our hot water needs. I guess it might depend on how heavy your hot water use is.

Gary

 

[semipro]

...This is one option that is pretty close to your option A:
http://www.builditsolar.com/Projects/SpaceHeating/DHWplusSpace/Main.htm
Thanks Gary. I reviewed your system in detail before starting to design mine.

You can just leave the space heating parts out if you don't want to do any space heating.
I may incorporate some space heating later if I'm successful with water heating.


Typically the systems take the colder water from the bottom of the tank for the supply to the collectors -- this makes the collectors operate more efficiently as there average temperature is lower and they lose less heat out the collector glazing. Good point. I hadn't thought of looking at it from the collector side.

The heat exchanger in this scheme is just a 300 ft coil of 1 inch diameter pex that sits in the top third of the tank. The street water makes on pass through it on the way to the hot water heater. Even though pex does not conduct heat so well, this scheme works fine because there is so much surface area, and because the coil itself stores about 10 gallons of 100% preheated water -- most hot water draws come from the 10 gallons. At present I plan to build mine with PEX this way also possilby with additional PEX or an added submerged tank for additional preheat capacity

I'm wondering if you will get a good return from doing both the heat pump water heater and solar water heating? Our solar water heater is a bit oversized by conventional guidelines, but it supplies nearly 100% of our hot water needs. I guess it might depend on how heavy your hot water use is. We have the heat pump water heater already. Its located in a basement that is pretty cool in the winter so we don't get a lot of heat pump benefit then. It still uses too much power in my mind. Hot water production is our number one energy draw. Space heating is also attractive via hydronics but I want to address domestic water heating first and then scale the system up as needed.

Do you or anyone else, see any benefit in Option C assuming I would end up with tanks at 2 different temps? Will I end up with tanks at two different temps? I have already collected old glass lined electric water heaters that I can cut the top off of and use as tanks. Problem is that they are only 40 gallons in size so I'd need more than one.

Gary

 

[GaryGary]

...This is one option that is pretty close to your option A:
http://www.builditsolar.com/Projects/SpaceHeating/DHWplusSpace/Main.htm
Thanks Gary. I reviewed your system in detail before starting to design mine.

You can just leave the space heating parts out if you don't want to do any space heating.
I may incorporate some space heating later if I'm successful with water heating.


Typically the systems take the colder water from the bottom of the tank for the supply to the collectors -- this makes the collectors operate more efficiently as there average temperature is lower and they lose less heat out the collector glazing. Good point. I hadn't thought of looking at it from the collector side.

The heat exchanger in this scheme is just a 300 ft coil of 1 inch diameter pex that sits in the top third of the tank. The street water makes on pass through it on the way to the hot water heater. Even though pex does not conduct heat so well, this scheme works fine because there is so much surface area, and because the coil itself stores about 10 gallons of 100% preheated water -- most hot water draws come from the 10 gallons. At present I plan to build mine with PEX this way also possilby with additional PEX or an added submerged tank for additional preheat capacity

I'm wondering if you will get a good return from doing both the heat pump water heater and solar water heating? Our solar water heater is a bit oversized by conventional guidelines, but it supplies nearly 100% of our hot water needs. I guess it might depend on how heavy your hot water use is. We have the heat pump water heater already. Its located in a basement that is pretty cool in the winter so we don't get a lot of heat pump benefit then. It still uses too much power in my mind. Hot water production is our number one energy draw. Space heating is also attractive via hydronics but I want to address domestic water heating first and then scale the system up as needed.

Do you or anyone else, see any benefit in Option C assuming I would end up with tanks at 2 different temps? Will I end up with tanks at two different temps? I have already collected old glass lined electric water heaters that I can cut the top off of and use as tanks. Problem is that they are only 40 gallons in size so I'd need more than one.

Gary

Hi,
You might have a look at John Canivan's multi tank setup at http://www.jc-solarhomes.com/ -- he uses a series of poly barrels and claims some advantages in thermal progression for this. To be honest, I've never read it carefully, and I don't really have an opinion-- but, I think John gives a good detailed description.

Gary

 

[semipro]

Option C - Multiple tanks in series with hot water from the solar collector entering near where preheated water flows out to the HPWH. It looks to me like option would result in different tank temps and possibly less overall efficiency since temperature exchange rates are directly dependent upon temperature differential.

Other way around - when the hot water from the collector leaves level with the heated water leaving the tank, it can never be cooler than the water leaving the tank, even with 100% efficient heat exchange. When it goes in reverse flow like this, then you will still be getting some heat transfer from it to the hottest water, heating it still further, while as it goes down into the cooler water in the tank it will still transfer some heat, leaving at a little over the feed temperature.
What you're thinking of is the rate of heat flow per unit area, which is indeed dependent on temperature difference. Since you aren't area limited in this sort of heat exchanger, then you're better off going with a reverse flow heat exchanger like you describe here, which will extract far more heat from your coolant, thus extracting more heat from your solar system.

By reverse flow you mean my option C correct? The more I ponder this the more I think you're right.

 

[pdf27]

By reverse flow I mean the hottest temperature fluid on one side being next to the coldest on the other. It's standard practice in just about every commercial heat exchanger I can think of short of the most basic radiators.
The diagram below (flagrantly stolen from Wiki) should make it a bit clearer.

http://upload.wikimedia.org/wikiped...Exchangerflow.svg/600px-Exchangerflow.svg.png

 

[semipro]

By reverse flow I mean the hottest temperature fluid on one side being next to the coldest on the other. It's standard practice in just about every commercial heat exchanger I can think of short of the most basic radiators.
The diagram below (flagrantly stolen from Wiki) should make it a bit clearer.

http://upload.wikimedia.org/wikiped...Exchangerflow.svg/600px-Exchangerflow.svg.png

Thanks. I wish I'd seen that earlier.

 

[semipro]

I'm not sure what I was thinking before as the I had the domestic water supply and solar collector flows transposed. Sorry for the confusion.

Attached is a revised diagram showing 3 options.

A. A single larger tank (which I believe is close to what Gary uses in the system he referenced on his builditsolar website)
B. Multiple tanks in series (my most likely option as I already have multiple steel insulated tanks)
C. A single large tank with overflow/underflow baffles (basically replicates option B but with one tank (just an exercise in design as I"m unlikely to find or build this).

I hope these make more sense than previous options.

Now I've got to figure out how to coil PEX tightly enough to fit (without kinking it) inside 40 gal. water heaters with tops removed.

Thanks for the input so far.

 

[pring7]

I have option B with a pressurized system and two evacuated tube collectors and I am pretty happy with it. Tank 1 is 200 gal. and tank two is my regular 50 gal. electric hot water heater (hwh). The only problem that I have is occasionally thee solar water will come in so hot that it will trip the safety off on the hwh. I think that this sometimes happens in April and we don’t figure it out until late November or early March after several days of overcast. It’s always in the morning when you want to get a shower before going to work too. Anyway, the fix is easy, three Phillips head screws and you flip the panel off and reset the hwh.

 

[pdf27]

Options B and C I wouldn't bother with the double coil - the amount it'll save you per year will be minimal, whereas it looks like a bit of a nightmare to plumb up. The only time it's worth considering is if you're really having trouble fitting more than say 50ft of tube in the cylinder. See http://solar-components.com/solartankdrawing.jpg for a fairly typical twin-coil solar cylinder over here. The lower coil is for the solar, upper a hot water loop to normally a gas boiler, stored water goes to the domestic supply. The diameter of the cylinder in the drawing is typically ~2ft, so that gives roughly 50ft of tubing inside the cylinder.

The main advantage of that setup, as opposed to the version you're using, is that you need a much smaller quantity of antifreeze for the solar circuit. It also appears to deal better with limescale from hard water (common over here). The version you're using does exist, but is much rarer.

Drainback systems are pretty rare over here, with most people using the evacuated tubes. Looking into it, I think the major advantage is that they split down into much lighter components, making it a LOT simpler and safer to install on pitched roofs. That's what I'm planning for this summer - trip over to Germany to visit my uncle and pick up one of these http://www.ebay.de/itm/Solaranlage-...978484609?pt=Solaranlagen&hash=item3a6f7bc181 - including installation the whole system will be ~$1500 and do ~70% of my hot water for the year.

 

[pdf27]

I have option B with a pressurized system and two evacuated tube collectors and I am pretty happy with it. Tank 1 is 200 gal. and tank two is my regular 50 gal. electric hot water heater (hwh). The only problem that I have is occasionally thee solar water will come in so hot that it will trip the safety off on the hwh. I think that this sometimes happens in April and we don’t figure it out until late November or early March after several days of overcast. It’s always in the morning when you want to get a shower before going to work too. Anyway, the fix is easy, three Phillips head screws and you flip the panel off and reset the hwh.

Not sure if you have them over here, but for us it's a legal requirement to fit thermostatic mixing valves where there's a risk the water temperature will exceed 140F. Makes sense to me since there is quite a severe risk of scalding if the temperature exceeds that. In the end I just decided to use the solar controller to cut the flow off when the cylinder reaches that temperature however - limescale buildup is a major problem at higher temperatures, and we're in a relatively hard water area.

 

[semipro]

I have option B with a pressurized system and two evacuated tube collectors and I am pretty happy with it. Tank 1 is 200 gal. and tank two is my regular 50 gal. electric hot water heater (hwh). The only problem that I have is occasionally thee solar water will come in so hot that it will trip the safety off on the hwh. I think that this sometimes happens in April and we don’t figure it out until late November or early March after several days of overcast. It’s always in the morning when you want to get a shower before going to work too. Anyway, the fix is easy, three Phillips head screws and you flip the panel off and reset the hwh.

Not sure if you have them over here, but for us it's a legal requirement to fit thermostatic mixing valves where there's a risk the water temperature will exceed 140F. Makes sense to me since there is quite a severe risk of scalding if the temperature exceeds that. In the end I just decided to use the solar controller to cut the flow off when the cylinder reaches that temperature however - limescale buildup is a major problem at higher temperatures, and we're in a relatively hard water area.

They're commonly referred to as "tempering" valves here. I plan to include one in my system.

 

[woodgeek]

Given that your payback is BTUs in the winter (since the HPWH covers the summer cheaply) you need to consider your resource....all solar DHW crank heat in the summer, and most can struggle in the winter. Look up the monthly solar resource tables for your area, or just use PVWATTS to look at trends.

For my example in PA, I have a great solar resource overall, but 90% of the output hours (IIRC) is March-November. In other words, winter is kinda cloudy where I am. I had lots of visions of DIY solar, kicking in space heat, storage, you name it, until I saw that.

New England was a bit worse for this. Rockies and Southwest rocked in the winter (which is why, ahem, solar heating is big there) Don't know about SWVA. Look it up before you go any further.

 

[pdf27]

Given that your payback is BTUs in the winter (since the HPWH covers the summer cheaply) you need to consider your resource....all solar DHW crank heat in the summer, and most can struggle in the winter. Look up the monthly solar resource tables for your area, or just use PVWATTS to look at trends.

Just for fun (!) I've been playing with that this week, trying to see what the most you can get out of a solar thermal system is without a massive summer heat dump (i.e. swimming pool). Assuming a 2000 sq ft Passivhaus in Boston, MA with the panels orientated vertically to maximise winter gain at the expense of summer, they can manage all the hot water for an average year and provide all the heating from about April-October. That's pretty much the best you'll ever get out of a solar thermal system for a winter/summer climate, and requires ~160 square feet of high efficiency tubes as well as a superbly insulated house. Based on the rough numbers (I don't have a copy of PHPP) you'd need about 600 square feet (i.e. a third of the floor area of the house) to provide all the heating needs for the year, and will be dumping a LOT of heat in all but December and January.

In most cases it only makes sense to size the system to cover summer DHW use and maybe a bit to either side (depending on how cheaply you can get the collectors). It's only when you are pushing close to Passivhaus levels that it can start to make sense to use some of it for space heating, and even then only if it works with the chosen heating system...

 

[GaryGary]

Hi,
I guess I'd like to hear more about the details of the calculation.

Its been my experience that no heat dump is needed for vertical panels no matter how large the collection area is. On my system, they get the tank up to 165F, the controller is set to turn the pump off at this temperature so the tank water does not get too hot. Then the collectors drain back, and stagnate. The vertical panels get such a high incidence angle in the summer that the stagnation temperatures are not that high. This condition does not happen as often as you might think because summer sunny day solar input to the collector is less than half the winter sunny day input. But, even if it happened every day, I don't see any problem?

One thought on adding collector area for space heating is that the more poorly the house is insulated and the more cold weather you get, the faster any collectors you put up for space heating will payback their costs in saved fuel. That is, on a high heat demand house, the solar collector space heating output will be used more of the time than on a very well insulated home that might heat itself with just internal gains in mildly cold weather. You may not achieve a high solar fraction, but the payback period should be shorter?

http://www.builditsolar.com/References/HowMuch/HowMuch.htm

Gary

 

[woodgeek]

yeah, i'm not too worried about heat dumping or adjusting the angle of the array to minimize stagnation temps....it just seems that in a lot of locations the solar resource in the dead of winter is lousy. Not in Montana or Colorado or New Mexico, but def in the Northeast. I used to be down on PV versus solar thermal; the whole low tech and DIY thing really attracted me. But after looking at the winter resource, I can see that the quantity and value of the PV electricity all year long is huge compared to a pulling a few BTUs on the occasional sunny hour in December.

 

[pdf27]

Hi,
I guess I'd like to hear more about the details of the calculation.

It's relatively crude - based on the main Passivhaus requirement being an annual heating demand of 15 kWh/m2/year. Take the number of degree-days the site experiences per month and assume each day is the same in the month (told you it was crude!). That then gives you the average daily space heating requirement per month for a given size of house. Various calculations for hot water requirements are also available. Off the top of my heat (not having the spreadsheet to hand) it was something like 13 kWh/day for water heating year round and up to 30 kWh/day for space heating in December on a 2000 sq ft house.
Next comes the output from the solar panels. I used evacuated tubes since that's the more common technology over here and I could find performance curves (from the Solar Keymark testing done by the Frauenhofer institute) quite easily. A typical high-performance 5m2 evacuated tube collector will produce 2kW of energy at 1kW/m2 insolation with water and air at the same temperature, dropping to 1kW at 100 deg C temperature difference. Assuming the heating and hot water are driven from the same accumulator tank to get the temperature difference (DHW at 55 deg C) and taking the average temperature from the degree-days already mentioned gives the power output in kW per panel at 1kW/m2. Then using the PVWatts data for kWh/m2/day gets you the panel output in kWh/day - or a reasonable approximation thereto.

Its been my experience that no heat dump is needed for vertical panels no matter how large the collection area is. On my system, they get the tank up to 165F, the controller is set to turn the pump off at this temperature so the tank water does not get too hot. Then the collectors drain back, and stagnate. The vertical panels get such a high incidence angle in the summer that the stagnation temperatures are not that high. This condition does not happen as often as you might think because summer sunny day solar input to the collector is less than half the winter sunny day input. But, even if it happened every day, I don't see any problem?

For flat panel, no, it isn't a problem. I was doing the calculation for evacuated tubes as that's what I'm familiar with and can get the data for. When they stagnate, they cook the antifreeze, which is a bad thing.

One thought on adding collector area for space heating is that the more poorly the house is insulated and the more cold weather you get, the faster any collectors you put up for space heating will payback their costs in saved fuel. That is, on a high heat demand house, the solar collector space heating output will be used more of the time than on a very well insulated home that might heat itself with just internal gains in mildly cold weather. You may not achieve a high solar fraction, but the payback period should be shorter?

Makes sense, with a couple of major caveats:
1) The system I described is already 6'6'' x 27ft for a superbly insulated, 2000 sq ft house. Houses are more spread out in the US, but even then a lot won't have that much unobstructed south facing wall space/garden available for a solar thermal system. To a large extent I picked that size as it's the largest I could realistically get away with putting on the side of an external garage (can't put windows behind it so it's a poor fit with the side of a house).
2) Getting twice as much energy for a vertical panel in midwinter as midsummer is likely to be heavily dependent on location. For Boston (the city I picked as an example) the yield in December is only something like 10% or so higher than that in June. I'd guess this is down to cloud cover, which may not affect your site.
3) If your house is poorly insulated, the payback time for improving the insulation will also be fast too - odds are, faster than the payback for anything but a pure DIY solar thermal install. That's a means of justifying a decision that you've already taken for other reasons, rather than making it.

 

[pdf27]

yeah, i'm not too worried about heat dumping or adjusting the angle of the array to minimize stagnation temps....it just seems that in a lot of locations the solar resource in the dead of winter is lousy. Not in Montana or Colorado or New Mexico, but def in the Northeast. I used to be down on PV versus solar thermal; the whole low tech and DIY thing really attracted me. But after looking at the winter resource, I can see that the quantity and value of the PV electricity all year long is huge compared to a pulling a few BTUs on the occasional sunny hour in December.

Depends on the subsidies you get, and how much it costs you to heat your water normally. Electricity over here is ~$0.22/kwh and Natural Gas is ~$0.08/kwh. I'm also a LONG way further north (roughly the same as the southern tip of Alaska or Newfoundland), and probably get a lot more cloud in winter. Even given all that, with an east facing roof, heating water by Gas and doing it by myself so not qualifying for any subsidies, the payback time is about 6 or 7 years.
A similarly sized PV system would be about 5 times the price. Since there's a huge subsidy on them over here right now I'd be very tempted if I had the money, but that I don't and the payback time is probably actually a bit slower with electricity (we don't use net metering, and there are a few other differences) I decided against it.

But overall, remember that for retrofit applications solar thermal is well suited for summer hot water, less well for anything else. Work out the paybacks on that rather than heating in December, and it all looks more sensible.

 

[woodgeek]

thanks for the details...I think we agree. I was relating that prior to learning about the technology and thinking about the resource, I really didn't see how PV ever achieved paybacks comparable to solar thermal. I also assumed that b/c it was low tech, solar DIY thermal would give someone with the technical skills effectively 'free heat' in the winter. For me, the key realization was that a solar thermal space heating system only 'works' or pays you back 3-4 mos of the year, and is idle the rest. This really hurts payback times or BTU/DIY effort, etc. OF course, solar DHW works year round, but it seldom/never has extra BTUs in the winter for space heating. The poor winter resources in the Eastern US kicks this problem while it is down....your large array to heat a (conservative) Boston passive haus model is not surprising. I bet the same model moved to Boulder CO or Flagstaff AZ would show lots of solar bang for the buck. I suspect that Gary also has a point that your model is likely not the 'optimal' solution in its details.

Incentives can be an imperfect solution....I got a professional solar DHW quote that was nearly as much as a small PV system, and which had a 10 year payback AFTER a 70% tax rebate incentive (relative to conventional electric). And it had an operating cost and First Hour rating comparable to a HP water heater that cost 3x less upfront. Its clear that when possible installers jack their price to take the incentives.

For Semipro's case...if his HP water heater is satisfactory on a cost performance basis in the warm months, then the 'payback' DIY or otherwise to a solar system will depend on the winter resource....and may be disappointing. OF course, there is more to life than 'payback', he should go for it if he wants to try it out.

 

[pdf27]

Flagstaff AZ Boston MA
January 5.39 3.27
February 5.62 4.19
March 4.51 3.6
April 3.31 3.03
May 2.58 2.77
June 2.2 2.6
July 2.26 2.76
August 2.68 3.27
September 4.14 3.59
October 4.99 3.92
November 5.32 2.96
December 5.28 3.06

Values are in kWh/m2/day

Net result is you need a bigger area for your summer DHW needs as you're a long way further south so the panels see the sun at a more oblique angle in summer. Applying that to winter means you get about twice the heat in winter, which is **probably** just about enough for a Passivhaus to heat it with no assistance. It's complicated somewhat because you're probably in a cooling climate in Arizona, which means you'll get less solar gain in winter, and the insulation might be driven by cooling needs as well - meaning your requirements might be below the 15 kWh/m2/year I based it on.

If you're heating anything else, you're still going to need to dump heat in summer by one means or another though...

 

[woodgeek]

starting to feel like a hijack, but I need to comment.....the above is for a vertical panel? I get the aversion to dumping heat or high stagnation temps, but there seem to be better solutions. Why not tilt the panel at the altitude of the winter sun if you want space heating? Seems like you really want to minimize the panel area before the summer heat overload....the simplest heat dump (for a panel, not a tube) is just to circulate at night if the store is above some high setpoint....seems pretty easy to rig, even DIY.

Flagstaff look like it has more heating degree days, fewer cooling degree days and more snowfall than I do in PA. I guess its the altitude.

 

[GaryGary]

Hi,
Just a couple comments on all the comments

Vertical collectors work well in cold snowy climates. The reflection off the snow in front of the collectors makes vertical as good or better than something like a tilt of latitude + 15 degrees (the usual winter recommendation). Vertical often works out nicely as the collector fit flat against a south wall and are easier to install and look better. But, steeply tilted panels also work fine for winter collection and are resistant to summer overheating -- they also perform somewhat better in the late spring when the sun is getting higher and its cold enough that you still need space heating.

The point I made vertical panels gaining more in the winter than in the summer was for sunny days, and was just addressing the overheat issue. Just to check, I ran Rad On Col -- at 45 deg lat, sunny day, Dec 21 gives 1440 BTU/sf of collector, and June 21 gives 780 BTU/sf. At 30 deg lat, Dec 21 gives 1800 BTU/sf and June 21 gives 320 BTU/sf. So, mid winter, sunny day output is much higher, which helps with the overheating issue.

You can do drain back evac tubes with no anitfreeze cooking problems: http://www.builditsolar.com/Projects/WaterHeating/LargeDB/LargeDrainBack.htm

On the relative payback for DIY PV vs DIY solar thermal, I have one of each -- just by coincidence they both produce about 3300 KWH a year. The PV system is a 2150 watt, micro inverter system. The solar hot water system is this one: http://www.builditsolar.com/Experimental/PEXColDHW/Overview.htm

The before rebates out of pocket cost of the PV system was $10K
The before rebates out of pocket cost of the solar hot water wad $1K

Now PV prices have come down a bit since I did the PV system, but this is a factor of 10 difference!

I've since upgraded the solar thermal system to do both space and water heating:
http://www.builditsolar.com/Projects/SpaceHeating/DHWplusSpace/Main.htm
but, I still only have $2K into it (still 5X less than the PV system), and it produces quite a bit more energy on a yearly basis than the earlier system that did domestic water heating only. So, its hard for me to see how the payback for a DIY PV system comes close to a DIY solar water and/or space heating system. The exception might be if you live in a VERY cloudy winter climate the space heating part may not pay well enough to make it worthwhile.

This solar thermal system pays back its full cost of materials each heating season: http://www.builditsolar.com/Projects/SpaceHeating/solar_barn_project.htm
We do have descent sun here in MT, but its not nearly as good as a place like Denver -- but, it still pays quite well.

To me, DIY solar thermal is just an exceptionally good payback opportunity.
I'd encourage people to run the numbers for their situation and see what really comes out the best.

Gary