Drain-back system and evacuated tubes

[osolemio]

I am researching about drain-back, and how to make the most of it:

Efficiency and simplicity (no heat exchangers, no glycol anti-freeze, no freezing, no boiling, no auto-air-extractors, less lag and so on.)

I don't have any experience yet with drain-back, but I am aware of some of the classic mistakes.

Also, I don't have experience with using or installing evacuated tubes. Are they generally suitable for drain-back? I need to be absolutely sure there are no chance of damage from water NOT draining, thence freezing when it gets cold outside.

I will not have a hot water tank, but a hot water buffer. The difference mainly being that it's the bulk of water in the tank that will be common with the solar, the radiators, and underfloor heating, while the domestic hot water is what will be in a coil, inside the tank. This way, the hot water will be "freshly produced", always flowing (when in use), so there is no "sump of half rotten water". Well, the half rotten water will be that circulated in the system, not what I shower in.

The bit of it about the hot water buffer I am quite confident about, but not so much that of evacuated tubes and drainback. I am going to make an installation of compound parabolic mirrors for the evacuated tubes, and there will a specially designed filter in front, which helps to align diffuse light, for even higher efficiencies in challenging situations.

My main goal is to optimise efficiency, but especially the efficiency when it's 1) cold 2) overcast/foggy and 3) low light. Making usable solar thermal energy, in freezing overcast weather might sound impossible - but it's actually not. Especially not for radiant underfloor heating, all you really need is 30 C / 85 F from the solar panels. You will get a lot more on a sunny day, of course, but that's not the point. The point is - how much can you get on a cold and overcast day? Can you store enough thermal energy to avoid any auxiliary energy source?

My point is, if the technology is right, and the area is sufficiently large - useful solar can be made in almost all situations. Snow storms will probably be the exception, as well as obviously night time (I am not going for a broad spectrum solution, like infrared). But with a combination of diffuse light concentration, evacuated tubes, thermal storage and a few other added tricks, I just need to make it thought the winter.

Again, my main question is about drain-back and evacuated tubes - can I rely on the liquid section of the evacuated tubes manifold to drain properly?


Thank you all for your drainback. I mean, feedback!

[SDMCF]

Originally Posted by osolemio

The point is - how much can you get on a cold and overcast day?

I have 90 tubes and they perform very well on cold days but do not perform at all well on overcast days. The most productive time of year for me is February/March when the temperature can be -30C but the air is clear and the snow on the ground reflects extra light onto the tubes.

Again, my main question is about drain-back and evacuated tubes - can I rely on the liquid section of the evacuated tubes manifold to drain properly?

My system uses antifreeze and I have no experience of drainback systems but I will comment anyway. The water flowing through the manifold is effectively just flowing through a pipe (22mm copper on my manifolds). So long as that pipe is angled so the water will drain back under gravity I don't see a problem.

[osolemio]

Originally Posted by SDMCF

I have 90 tubes and they perform very well on cold days but do not perform at all well on overcast days. The most productive time of year for me is February/March when the temperature can be -30C but the air is clear and the snow on the ground reflects extra light onto the tubes.


My system uses antifreeze and I have no experience of drainback systems but I will comment anyway. The water flowing through the manifold is effectively just flowing through a pipe (22mm copper on my manifolds). So long as that pipe is angled so the water will drain back under gravity I don't see a problem.

Thank you so much for your feedback, SDMCF.

I know about the manifolds being vertical. The issue is if they have a kink in the middle, and leave water standing there. Of it there is any other lock in the system. I could use a compressor to force all air out, but then it gets a bit too complicated ... I also want to recycle the air, so oxygen isn't added to the system (components will start to corrode faster).

About your tubes in overcast weather: THAT is EXACTLY what the main purpose of my system is. I am building a solar panel system which will be optimized for low and diffuse light, even during cold weather. I am combining several known and (so far) unknown principles, to make this possible. There will not be any more energy coming out than what is in the light. But the point is to get something, rather than nothing - when it is needed the most.

[SDMCF]

Originally Posted by osolemio

I know about the manifolds being vertical.

I assume you mean horizontal - or actually a little off horizontal to facilitate drainback. If the manifolds were vertical the tubes would be horizontal and I don't think they would work in that orientation.

The issue is if they have a kink in the middle, and leave water standing there.

You should be able to look through the manifold end-to-end before you install it. I would be very surprised if there were a kink in the pipe, and even more surprised if the kink were big enough to cause a problem. I would return a manifold with such a kink.
Even if there were a minor kink I don't think it would be a problem if you have enough fall on the system. I have heard 1" of fall for every 4 ft of run as being good but don't know fist hand if that is correct. Anyway, I believe freezing any water left standing in the system could only cause a problem if, when frozen, it could be over half way up the pipe at the deepest point. Without that I don't see how freezing could split the pipe, which is surely the real concern?

Of it there is any other lock in the system. I could use a compressor to force all air out, but then it gets a bit too complicated

I think you are over-complicating things and in reality there isn't an issue here.

About your tubes in overcast weather: THAT is EXACTLY what the main purpose of my system is. I am building a solar panel system which will be optimized for low and diffuse light, even during cold weather.

Cold weather is not an issue. You can ignore ambient temperature.

Overcast / low / diffuse light conditions are an issue. You say you are designing around that. Great, I hope you succeed and I will be interested to see your results. Personally I do not know how to do that. In November & December when there is typically heavy cloud cover here my system produces about 2% of the output it does in February & March, even though the temperatures can be 20C or 30C or more higher in November & December.

But the point is to get something, rather than nothing - when it is needed the most.

I agree with that. I guess a lot depends on your local conditions. I need the heat the most when the weather is at its coldest, and around here that does not coincide with when the light is worst. Hence I can get heat when I need it most. I would certainly like to get heat when the skies are overcast but I accept that it has not been possible. If you can show how that can be done I am all ears. Good luck!

[osolemio]

Originally Posted by SDMCF

I assume you mean horizontal - or actually a little off horizontal to facilitate drainback. If the manifolds were vertical the tubes would be horizontal and I don't think they would work in that orientation.


You should be able to look through the manifold end-to-end before you install it. I would be very surprised if there were a kink in the pipe, and even more surprised if the kink were big enough to cause a problem. I would return a manifold with such a kink.
Even if there were a minor kink I don't think it would be a problem if you have enough fall on the system. I have heard 1" of fall for every 4 ft of run as being good but don't know fist hand if that is correct. Anyway, I believe freezing any water left standing in the system could only cause a problem if, when frozen, it could be over half way up the pipe at the deepest point. Without that I don't see how freezing could split the pipe, which is surely the real concern?


I think you are over-complicating things and in reality there isn't an issue here.


Cold weather is not an issue. You can ignore ambient temperature.

Overcast / low / diffuse light conditions are an issue. You say you are designing around that. Great, I hope you succeed and I will be interested to see your results. Personally I do not know how to do that. In November & December when there is typically heavy cloud cover here my system produces about 2% of the output it does in February & March, even though the temperatures can be 20C or 30C or more higher in November & December.


I agree with that. I guess a lot depends on your local conditions. I need the heat the most when the weather is at its coldest, and around here that does not coincide with when the light is worst. Hence I can get heat when I need it most. I would certainly like to get heat when the skies are overcast but I accept that it has not been possible. If you can show how that can be done I am all ears. Good luck!

Yes, of course you are right, SDMCF, I meant horisontal. Where I have installed systems before, people are scared of drainback because there are stories about how pipes or collectors have been cracked open in frost.

Some also mention how an "open" system will keep adding oxygen to the water, and then the installation will corrode from inside.

I know that it does NOT have to be an open system. Water and air can exist in a closed system. In fact, the air means the expansion vessel can be replaced by the drainback-vessel. In theory, you could even leave the top of the hot water buffer with air, although, a separate vessel is probably better.

Rather than making a lot of mistakes and have to redo the system, I am trying to weed out as many traps before hand as I can. I have gotten it down to a system so simple, that the only heat exchanger will be to extract hot tap/shower water from the hot water buffer. And then possibly from an alternative heat source. Where my biggest system is installed (my own house), the shared district heating water is running all the way out to the radiators in each house, in the original installation.

Anyway, I will keep researching to make the best possible system.

ABout the solar panels working in cold-overcast-low light: I will never be claiming to get more energy out of the light that there is. But even with a loss, the techniques I will be using will work together to enrich the energy there is, into something usable. Combined with other technologies, like seasonal heat storage and radiant floor heating, the plan is obviously to minimise or entirely avoid using a back-up heat source.

[jeff5may]

Drain back is not generally recommended with evacuated tube collectors. Mainly two reasons:
First, stagnation temperature can literally melt down gasket and sealing materials. When the sun shines brightly and there's no fluid to absorb the heat, temperatures in the collector can skyrocket to over 150 degC. If there's any liquid pooled in the collector, it expands to steam, which causes more things to cook and melt.

The other big problem is cold shocking. When the collector is under full sun, and fluid pumping stops, you get the previous paragraph. Stagnation and steam generation and pressure and yadda yadda yadda. When the fluid starts pumping again, all of the hot collector plumbing gets a sudden blast of cold fluid. At first, your manifold acts as a large coffee maker boiler, boiling and spewing shots of hot water with steam downstream. If the pump is strong, it pushes the hot froth down the pipe, overcomes the boiling action, then the boiling froth is replaced by lower temperature fluid. This heat shocks the whole plumbing all the way to a mixing vessel (if there is one).

[SDMCF]

Jeff, how do you see such problems happening? I have no experience with drainback systems so I can only guess.

It seems to me there are 2 ways: a failure to pump fluid into the collectors because of pump or power failure, or collector temperature rising rapidly before the system can react and pump fluid. Have I missed something?

Failure to pump is a problem in any system, so not unique to tubes or drainback.

Slow reaction at first seems surprising, but thinking about it I can understand how this might happen. The temperature sensors on mine are at the (outlet) end of the manifold. Without fluid in the system to dampen the temperature rise the end of each tube could conceivably get very hot before the temperature rises is conducted as far as the temperature sensor. If I hold in my hand the end of a tube that is normally inserted into the manifold, with the tube in full sun the end quite quickly becomes too hot to handle.

I had not considered that before, so I was lucky that I did not (could not with my layout) opt for a drainback solution.

[osolemio]

Originally Posted by jeff5may

Drain back is not generally recommended with evacuated tube collectors. Mainly two reasons:
First, stagnation temperature can literally melt down gasket and sealing materials. When the sun shines brightly and there's no fluid to absorb the heat, temperatures in the collector can skyrocket to over 150 degC. If there's any liquid pooled in the collector, it expands to steam, which causes more things to cook and melt.

The other big problem is cold shocking. When the collector is under full sun, and fluid pumping stops, you get the previous paragraph. Stagnation and steam generation and pressure and yadda yadda yadda. When the fluid starts pumping again, all of the hot collector plumbing gets a sudden blast of cold fluid. At first, your manifold acts as a large coffee maker boiler, boiling and spewing shots of hot water with steam downstream. If the pump is strong, it pushes the hot froth down the pipe, overcomes the boiling action, then the boiling froth is replaced by lower temperature fluid. This heat shocks the whole plumbing all the way to a mixing vessel (if there is one).

Thank you for that insight, Jeff5May.

I will have more than one sensor inside a manifold (that is, not on the exit, but built into the panels themselves. I will have plenty of user capacity to sink the thermal energy. If for some reason the system drains back, I won't let the liquid back until the temperature is under control. Ideally, the liquid would be flowing before it heats up at all.

I will even do more, like using compound parabolic mirrors, inside an insulated box. I am looking for the most heat resistant tubes but I do expect that I will blow some tubes, during testing.

The point of these panels are that they will be operating even at very low (and diffuse light), which is the most significant drawback of both electric at thermal solar panel. At least with current technology. I intend to change that, and yes, there isn't any RD without failures and mistakes. I am asking here first, though, to avoid as many mistakes as possible and thus, your feedback is most appreciated.

This is too important to develop that I won't let myself be held back by challenges, but rather figure out how to best overcome them.

[osolemio]

Originally Posted by SDMCF

Jeff, how do you see such problems happening? I have no experience with drainback systems so I can only guess.

It seems to me there are 2 ways: a failure to pump fluid into the collectors because of pump or power failure, or collector temperature rising rapidly before the system can react and pump fluid. Have I missed something?

Failure to pump is a problem in any system, so not unique to tubes or drainback.

Slow reaction at first seems surprising, but thinking about it I can understand how this might happen. The temperature sensors on mine are at the (outlet) end of the manifold. Without fluid in the system to dampen the temperature rise the end of each tube could conceivably get very hot before the temperature rises is conducted as far as the temperature sensor. If I hold in my hand the end of a tube that is normally inserted into the manifold, with the tube in full sun the end quite quickly becomes too hot to handle.

I had not considered that before, so I was lucky that I did not (could not with my layout) opt for a drainback solution.

There are advantages and disadvantages to both systems. If drainback is done right, I sincerely believe it will eventually be simpler in construction, more reliable, more failsafe, cheaper to build and maintan and even more efficient.

I am really tired of watching energy lost in heat exchangers. I need a better system, and that is what I will build.

[WillyP]

I have been contemplating a similar system. My solution was to store hot water for the radiant floor system in underground tanks, buried inside. For the evacuated tube heaters, I am going to use a separate set of tanks, that will heat the water tanks with coils. That way I can just let the evacuated tubes run all day, and heat their tanks as hot as they want. Then when the (underground) water tanks start to cool, they can feed the coils. Of course I am building the whole system inside a "greenhouse" so I don't have to worry about freezing or much in the way of heat loss at night.
I posted a thread about it but here is a diagram that explains it better.