Convert thermosyphon to pumped.

[J.P.M.]

When I say "not rocket science" let me put that into perspective.. when I started intensive retraining into air conditioning -after a 4 year apprenticeship at a UK atomic weapons base- I thought to myself "what's in a fridge, it sits in the corner and gets cold?"

That was in 1982. 35 years later I'm still getting my arse kicked by them.. So I don't mean to undermine the science here or those involved . I agree I was naive to begin with and have spent some time reading (often conflicting) advice but I'm getting there slowly

I will sit and mull over your fantastic reponse

OK, While you mull over my response, I'll mull over three questions:
1.) What are you doing that is pissing off refrigerators and causing them to kick your ass ? Does Tenerife have gangs of refrigerators with bad attitudes ?
1.) Just how does one get their ass kicked by a refrigerator ? I'm having trouble conjuring up a mental image of that one.
2.) After 35 years, can't they be reasoned with ?

Seriously:

I've contended and written for a long time that, with the possible exception of the direct application of the photoelectric effect, solar energy applications are not rocket science.

Most things solar can be readily understood by taking an entropic view of the basics of what's known about what's going on, and then seeing were it leads. That approach seems to work particularly well, for me anyway, with respect to thermosiphon type systems. That and a basic understanding of Thermodynamics, heat transfer and fluid mechanics.

So, if not meant sarcastically, I'd question fantastic. Just attempting to do the best I can in this limited venue, and no more than a surface scratch.

I've found the more I learn, the less conflicting various aspects of a subject become, and also the more aware I become of what is, at the end of the day, my profound ignorance.

Best practice KISS:
- Keep the tank above the collector.
- Keep the piping as straight as possible, running up only once and down only once with no local high/low spots that can and will form pockets.
- Use generous pipe sizes and as few elbows as possible, and use long rather than short radius elbows.
- Insulate the crap out of everything except the glazing on the collector.

With some caution, see "builditsolar.com". Lots of ideas and pretty pictures.The caution being that some of those ideas are good example of a little knowledge sometimes being a dangerous thing.Caveat Lector.

[r.bartlett]

When I say "not rocket science" let me put that into perspective.. when I started intensive retraining into air conditioning -after a 4 year apprenticeship at a UK atomic weapons base- I thought to myself "what's in a fridge, it sits in the corner and gets cold?"

That was in 1982. 35 years later I'm still getting my arse kicked by them.. So I don't mean to undermine the science here or those involved . I agree I was naive to begin with and have spent some time reading (often conflicting) advice but I'm getting there slowly

I will sit and mull over your fantastic reponse

[J.P.M.]

just to say I got the heater kit installed and had a good feel around inside the tank. there was no evidence of any scale build up . all seemed clean as a whistle. still looking at upgrading to larger loop pipe and cleaner flow lines.with large radi elbows etc.

n the meantime i dropped the return pipe so it sags below the level of the bottom of the panel. I read this means this return pipe is the bottom of the cold loop and the panel is all in the upflow..if that makes sense

. I set the stat at 50c for the electric heater. However the water seems very hot . but we do have a clear bright sunshine days atm.

I know and understand the principles of TS. It's not rocket science. There still is no clear explanation of why a tank situated higher wont reverse . apart from 'because' which isn't really an answer

If the TS effect increases with tank height due to I presume 'head' why the reverse effect isn't made stronger too...

Putting a loop or more tubing to locally change elevation in the line from the tank out to the collector inlet won't get you much.

Like most aspects of solar energy, it certainly is not rocket science. If you truly know and understand the governing principle driving thermosiphon flow (that is, thermally induced density differences), that's as much explanation as needed to understand why warmer (actually less dense) fluid will tend to stay at higher elevation in any closed fluid loop, and why cooler (actually denser) or additionally cooled fluid will tend to stay at or move to lower elevations. Less dense fluids rise to the top of fluid loops and tend to stay there.

Think of a pipe loop consisting of a solar flat plate collector with a simple line connecting the outlet from the collector back to the inlet of the collector with no positive changes in elevation between collector outlet at the top and collector inlet at the bottom. It's all downhill, or all uphill, and one size line with no intentional storage.

Now, imagine what happens when the collector sits in the sun. The collector and the fluid in the collector portion of the loop heats up. As the fluid in the collector heats up, because of its (now) lower density it rises, just like the air in a hot air balloon, displacing some of the fluid above it and moving that displaced fluid upward and "forward", out of the collector and around the loop. As more heat is added, this circulation continues in one direction ("upward"), and all's well.

Now, thin what happens to fluid temps. in the loop after a sunny day. The sun goes down, the collector and the fluid in it become cooler than the rest of the fluid in the loop, and the fluid density in the collector increases, with the loop volume that was occupied by the formerly cooler (and less dense) fluid now filled by fluid from "above", inducing reverse circulation with nothing to prevent it. As long as the condition exists where the collector and the fluid in it are warmer than the surroundings, and so heat is being removed from the fluid, and the fluid density increases as fluid temp. decreases, the fluid flows "backwards", top to bottom.

Now, imagine a wide spot in the loop and call it a tank, and imagine what happens when the tank is put in two different locations. First, put the tank directly and entirely below the collector. During a sunny day, the collector loop and the much larger quantity of contained fluid will have been operating just like the situation with no wide spot/tank in the line. At the end of the day, the line, and the tank, and the contained fluid will be relatively warm compared to the surroundings. As the collector cools after sundown, the fluid in it will also cool down and so become denser than the rest of the fluid in the loop. Just like the no storage case, the now denser fluid in the collector will then "drop" into the now lower tank, with the specific volume difference induced in the cooler fluid by the temp. drop being replaced by fluid from the "upper" collector connection (the "hot" connection). While all this is going on, the warmer fluid in the tank at a lower elevation really wants to get to a higher elevation (just like the hot air in a balloon) than the cooler fluid in other portions of the line (like the collector) - and it has just the path to do so - "backward" through the line from the tank "inlet", and up to the collector "hot" "outlet". This reverse circulation will continue as long as the temp. of the fluid in the collector remains cooler than the fluid in line "above" it.

Lastly, envision a similar setup, but this time with the wide spot/tank directly above the collector. Same sunny day. Fluid in the collector heats up and flows upward through the collector into the tank just like before, same flow direction and everything. Only difference this time is the location of the tank. Throughout the day, life is good and fluid temps. at day's end are probably close to or a bit more than when the tank was below collector. Now, sun goes down, the collector and the fluid in it cools off same as before and, just like before, because the collector and the fluid in it are getting cooler than the rest of the fluid in the loop (including the tank fluid), the collector fluid gets denser and wants to sink. In the mean time, the warmer (and less dense) fluid in the tank, now at the top of the loop is quite content to stay right where it is, and thus acting as a counter to the cooler fluid wanting to drop and initiate a reverse flow. This push and pull quickly results in a stalemate: Circulation stops, or almost stops. Cooler, more dense fluid goes to the lower portions of the loop and wants to stay there. Warmer fluid, already at the top of the loop doesn't want to go anywhere, and for the most part, it doesn't.

Themosiphon effects or fluid density driving forces increase with relative temperature differences in the working fluid, as well as by greater specific volume (units of volume/mass, as in ft.^3/lbm., or m^3/kg.) changes as f(temp.). For heating applications where fluid density decreases with increasing fluid temp. Storage needs to be above the heating surface to prevent unwanted (reverse) thermosiphoning. For fluids where density increases with increasing temp. (usually not a consideration) tanks need to be below the heating surface. Wwhere fluid density decreases with increased temp. (as in most practical cases, the tank needs to be above the level of the highest heating surface do not necessarily increase with tank elevation changes once the tank is above the level of the highest heating (or cooling) surface. As long as a fluid exhibits a density decrease with temp. increase, a, say, 50 deg. C. temp. diff, in the working fluid temp. in a thermosiphon loop, will produce more flowrate than, say, a 10 deg. C. temp. diff., regardless of the relative diff, in heights between the heating (or cooling) surface and a , i.e. a flat plate solar collector and the fluid height above the collector.

In a twist of nature, and pretty much useless for this discussion unless considering thermosiphon activity below that temp., but perhaps worth noting, water's greatest density occurs at ~ 3.98 C. or so and becomes less dense both above and below that temp.

[r.bartlett]

just to say I got the heater kit installed and had a good feel around inside the tank. there was no evidence of any scale build up . all seemed clean as a whistle. still looking at upgrading to larger loop pipe and cleaner flow lines.with large radi elbows etc.

n the meantime i dropped the return pipe so it sags below the level of the bottom of the panel. I read this means this return pipe is the bottom of the cold loop and the panel is all in the upflow..if that makes sense)

. I set the stat at 50c for the electric heater. However the water seems very hot . but we do have a clear bright sunshine days atm.

I know and understand the principles of TS. It's not rocket science. There still is no clear explanation of why a tank situated higher wont reverse . apart from 'because' which isn't really an answer

If the TS effect increases with tank height due to I presume 'head' why the reverse effect isn't made stronger too...

[J.P.M.]

I have concerns regarding check valves as it seems pointless smoothing out the water paths and having a naturally highly restrictive component in the circuit

as part of my daily advancement:
Can you educate me as why a tank at panel height can reverse TS but a tank higher can't
thanks

I'll help folks some - as I believe I have on this thread, and as I see part of the mandate of my former profession (engineering), but in general, and not directed at you personnaly, I consider spoonfeeding folks things they can readily find and learn by their own efforts to be no more than enabling behavior on my part - a behavior I do not engage in too much , too often or too deeply, and which does no one any good.

But, you can easily teach yourself which is what I've been suggesting all along in this thread.

I won't feed you, and I won't teach you to fish, but I will point in the direction where you can take control of your situation, be self actualizing and learn to fish on your own.

I'll get you started: www.appropedia.org/Thermosiphon will explain a lot of the basics. Otherwise , google "Why do thermosiphon systems work ?", learn the basics, and then use your critical thinking skills and logic to figure out why more height will lead to higher flowrates in thermosiphon systems, and for your situation and question, why the minimum necessary height of storage to completely prevent reverse thermosiphoning is above the highest heating (or - hint - cooling) surface elevation ( the collector).

My prior posts to this thread also contain some nuggets. Reread them after knowledge quest about the basics. You may gain some further insights.

[r.bartlett]

I have concerns regarding check valves as it seems pointless smoothing out the water paths and having a naturally highly restrictive component in the circuit

as part of my daily advancement:
Can you educate me as why a tank at panel height can reverse TS but a tank higher can't
thanks

[J.P.M.]

I am off to buy a new NRV (I have a 1 hr drive) and an electric heater kit I think the nrv is a major factor on why I don't have enough hot water esp in the morning.

The rest of the upgrading as per your post will be done over Christmas.

I'd respectfully suggest you consider learning why any suggested improvements, including mine, be helpful before you do the upgrade. You have the advantage over everyone here in that you are on site and can see all that is going on better than anyone. More education will lift the scales from your eyes.

Also, and to reiterate, increasing the tank elevation is a better solution to the problem of reverse thermosiphon flow than a check valve. The check valve is a necessary evil brought on by the low tank elevation.

Good luck.

[r.bartlett]

I am off to buy a new NRV (I have a 1 hr drive) and an electric heater kit I think the nrv is a major factor on why I don't have enough hot water esp in the morning.

The rest of the upgrading as per your post will be done over Christmas.

[J.P.M.]

Couple things:

1.) When/if you add a check valve back into the system: DO NOT use a spring loaded check valve as is commonly available. As I wrote, common spring loaded valves need more velocity head against them to open (commonly called "cracking pressure") than any thermosiphon system can generate. A spring loaded check valve may well stop operating flow altogether. Swing type or gravity check valves are better, but will still reduce flow to what's probably/likely an unacceptable degree, with little in the way to tell if/how much it reduces flow. If you do use a swing check valve, get one that's a pipe size larger than the line it's in and use 2 pipe reducers, one on either side of the valve to fit the line.

2.) The big deal/reason for using a check valve is to prevent reverse flow that can reduce tank temps as much as or more than they increase during the previous day. A check valve is necessary in thermosiphon systems when the tank is not completely above the collector as is your situation. One sure fire way to get around a check valve altogether is to elevate the tank so that it is completely above the collector (I know, a real PITA and a whole lot easier to say than do solution). But, in doing so you will eliminate the need for a check valve altogether. The way it is now, and after it gets serviced, you'll lose some collected heat at night. So, you'll need a check valve or a tank raising. Between those two options, finding a suitable thermostatic check valve with low pressure drop may not be possible. That leaves raising the tank, or taking a chance on a swing or gravity check valve having low opening/operating pressure drop characteristics. I'd raise the tank.

3.) Thermosiphon systems require VERY low system operating pressure drop. That means large(r) pipe diameters than for pumped systems, as few bends/elbows as possible with all those bends/elbows of a wide radius type, and as I wrote, only two changes in elevation for the whole system, one up, and one down.

4.) You mentioned something about reducing a pipe diameter - Don't do it - Make it bigger. Looks to me like your system was put together by someone who was trying to save money on piping at the expense of system performance, or ignorance of the importance of keeping thermosiphon system pressure drop low, or both. The piping on the collector loop is simple and pretty straightforward. I'd raise the tank about 60 or 75 cm. or so, plumb the collector loop with 3/4", or better, 1" copper - No plastic crap - and use reducers at the tank and collector connections. I'd then reinsulate with 1" or greater thickness closed cell foam, either jacketed or (as I did) wrapped with aluminum duct sealing tape (NOT cloth type "duct tape"). Do it right and you might get away with as few as 4 to 6 long radius elbows if you play heads' up ball with the tank elevation.

5.) While you're at it, check and service the expansion tank. They can fill up with water after a few years as the bladder's usually will leak after a period of time.

[r.bartlett]

Once upon a time, some horizontal tanks came with cold feed line inlets and collector returns that ran the length of the tank with holes along the length to help avoid destroying the thermal stratification. Others techniques such as diffusers of various designs were sometimes tried. I toyed with a few such designs, but came to the conclusion it wasn't worth the time/hassle/effort in most cases, and that reducing the draw was a good solution for preventing most thermocline disturbance by the makeup water, in conjunction with perhaps a larger collector return preserved most of the thermal stratification.

I thought about suggesting a higher slope of, say, 50 deg. or so for the reasons you cite as well as a bit more head for driving density differences, but at 28 deg. N. latitude, and the goldilocks ambient seasonal temps., the high year round clearness index, and with everything else the system needs, I figured there was already enough going on. FWIW, I agree in principle with a higher tilt.

When I repipe I will reduce the outlet pipe diameter . I am starting to think I need a second system. One of the problems you mention is high draw there is a long distance between the solar tank and the main house. the solar is sited on a small second house to feed two dwellings.
tenerifehouse plan.JPG
the tank is sited on the grey roof and feeds the small finca and under the path to the main house ....note solar located finca roof is now red so not to confuse you!

[r.bartlett]

Thanx for the photos. Looks to me like the collector outlet is going directly to the tank, and the tank outlet is going to the collector inlet.

So, if I'm seeing things correctly, while you may have a HX inside that tank, the way the system is plumbed, it's not utilizing the heat exchanger.

chromagen diag.JPGchrmoagen.JPG

it's plumbed in as per the above (espanol)
Not sure what you mean exactly though tbh.

Although your tank is not at an elevation that will prevent reverse flow entirely, that lower elevation is probably contributing some, but not too much to your flow problems. A higher tank elevation would help some, but I don't believe that's your biggest worry.

I did not see one, but in systems where the tank is not completely above the topmost collector elevation, a check valve is necessary in the collector/tank circuit to prevent reverse thermosiphoning. If there is a check valve present, three things: first, conventional check valves cannot be used - they will kill the flow because they need more pressure head to open than thermosiphon systems can possibly generate, 2d, a thermostatic check valve similar in principal to those used on liquid cooled ICE's is required. Third, check valves fail. Because it's an obstruction in the line, like any valve, it will get crudded up. That means it may get stuck partially closed(or open - same thing).

there was a check valve on the bottom of the tank return to the collector. I removed the inners to ensure no restriction

I believe your problem is a mix of air blockage(s), smaller than usual pipe diameters (for thermosiphon systems ), and any likely mineral deposits on the piping and in the valves and other line devices that reduce diameters and increase pressure drop = reduce flowrate further. That lower flowrate means a larger temp. difference between coolant inlet and outlet, and that means lower performance.

The convoluted flex joints add to the pressure drop. They're handy and useful, but need to be kept short. Short radius elbows as you have aren't helping either.

I can re-pipe it

Thermosiphon systems do not easily (or at all sometimes) tolerate more than two changes in elevation: one going up and one going down. Local, that is small, changes in elevation can cause problems or at least reductions in performance. Even if that air vent at the collector is working (FWIW, I get about 2 years out of mine and it fails closed), that slight elevation drop and then rise in the line at the collector outlet can slow the flow.

Between the tank elevation, what looks like a perfect spot for an air lock in the collector outlet (if what looks to be an air vent is non operable from scaling), and what's probably some rather substantial mineral scale buildup in the system in general, the added flow resistances are decreasing the flow.

I'd get the system serviced and checked out.

the AAV is working correctly. I will look at repiping with better smoother lines and no undue rise/falls

Some good news: Without a HX present, you can certainly add a pump to the system, but a properly designed and reasonably maintained and cleaned thermosiphon system has substantially the same performance as a pumped system would have. Also, and hard to say without being there, but adding a pump to the existing system without addressing the other things going on may not solve too many problems.

I'd get the system serviced, meaning cleaned, checked for equipment operating normally, and maybe clean up the piping layout (shorten and straighten the runs and get rid of that at least partial air lock at the top), and then see what happens before I added a pump.

I will repipe.

Some other things:

Looks like plastic tubing on the tank out to the house. Bad idea, particularly if/when the system starts producing really hot water.

I didn't notice any safety relief valves in the system. That's a big deal. Seriously. Are there municipal building/plumbing codes in force ?

yes there is a PRV on the cold inlet. no PRV on the solar loop. Plastic is used here all over the island. It's rated above 100c.

The closed cell foam insulation such as you have in place is good stuff, but it needs to be protected from sunlight. Your stuff is deteriorating. I've found aluminum duct sealing tape (NOT cloth duct tape) lasts ~ 5 - 7 years outside under direct sun and does a good job of protecting the Armaflex type insulation. Your insulation is also shrinking. Insulation gaps will cause a larger decrease in overall insulation effectiveness than apparences may make you believe - a 5% area uncoverge will probably decrease overall insulation performance by a lot more than 5%. 20+ % or so is more likely.

upgrading the insulation can be done at the same time as the repiping.

big thanks.

[J.P.M.]

Looking at your pictures the collector looks to be approximately 30 degrees a good compromise for year round use. Increasing the angle to 50+ degrees would optimize the collector for winter operation with out losing to much for the summer season.

Once upon a time, some horizontal tanks came with cold feed line inlets and collector returns that ran the length of the tank with holes along the length to help avoid destroying the thermal stratification. Others techniques such as diffusers of various designs were sometimes tried. I toyed with a few such designs, but came to the conclusion it wasn't worth the time/hassle/effort in most cases, and that reducing the draw was a good solution for preventing most thermocline disturbance by the makeup water, in conjunction with perhaps a larger collector return preserved most of the thermal stratification.

I thought about suggesting a higher slope of, say, 50 deg. or so for the reasons you cite as well as a bit more head for driving density differences, but at 28 deg. N. latitude, and the goldilocks ambient seasonal temps., the high year round clearness index, and with everything else the system needs, I figured there was already enough going on. FWIW, I agree in principle with a higher tilt.

[LucMan]

possibly which is why I am trying to squeeze more out of what I have

Do you have flow restrictors on all of your faucets and shower heads? You want to keep your terminal flow rates below 1.5 gpm if possible, especially with a horizontal storage tank. With high flow rates it becomes and issue with the entering cold water mixing with the hot water effectively reducing the amount of HW available. Vertical tanks are are less prone to this issue because of the additional distance between the cold water inlet & HW outlet.
Looking at your pictures the collector looks to be approximately 30 degrees a good compromise for year round use. Increasing the angle to 50+ degrees would optimize the collector for winter operation with out losing to much for the summer season.

[J.P.M.]

https://www.dropbox.com/sh/w7lumy77h...CEuvP2Bfa?dl=0

I've added a dropbox link as I am having trouble uploading pics.

the temps were taken around 11.30 today.

Thanx for the photos. Looks to me like the collector outlet is going directly to the tank, and the tank outlet is going to the collector inlet.

So, if I'm seeing things correctly, while you may have a HX inside that tank, the way the system is plumbed, it's not utilizing the heat exchanger.

Although your tank is not at an elevation that will prevent reverse flow entirely, that lower elevation is probably contributing some, but not too much to your flow problems. A higher tank elevation would help some, but I don't believe that's your biggest worry.

I did not see one, but in systems where the tank is not completely above the topmost collector elevation, a check valve is necessary in the collector/tank circuit to prevent reverse thermosiphoning. If there is a check valve present, three things: first, conventional check valves cannot be used - they will kill the flow because they need more pressure head to open than thermosiphon systems can possibly generate, 2d, a thermostatic check valve similar in principal to those used on liquid cooled ICE's is required. Third, check valves fail. Because it's an obstruction in the line, like any valve, it will get crudded up. That means it may get stuck partially closed(or open - same thing).

I believe your problem is a mix of air blockage(s), smaller than usual pipe diameters (for thermosiphon systems ), and any likely mineral deposits on the piping and in the valves and other line devices that reduce diameters and increase pressure drop = reduce flowrate further. That lower flowrate means a larger temp. difference between coolant inlet and outlet, and that means lower performance.

The convoluted flex joints add to the pressure drop. They're handy and useful, but need to be kept short. Short radius elbows as you have aren't helping either.

Thermosiphon systems do not easily (or at all sometimes) tolerate more than two changes in elevation: one going up and one going down. Local, that is small, changes in elevation can cause problems or at least reductions in performance. Even if that air vent at the collector is working (FWIW, I get about 2 years out of mine and it fails closed), that slight elevation drop and then rise in the line at the collector outlet can slow the flow.

Between the tank elevation, what looks like a perfect spot for an air lock in the collector outlet (if what looks to be an air vent is non operable from scaling), and what's probably some rather substantial mineral scale buildup in the system in general, the added flow resistances are decreasing the flow.

I'd get the system serviced and checked out.

Some good news: Without a HX present, you can certainly add a pump to the system, but a properly designed and reasonably maintained and cleaned thermosiphon system has substantially the same performance as a pumped system would have. Also, and hard to say without being there, but adding a pump to the existing system without addressing the other things going on may not solve too many problems.

I'd get the system serviced, meaning cleaned, checked for equipment operating normally, and maybe clean up the piping layout (shorten and straighten the runs and get rid of that at least partial air lock at the top), and then see what happens before I added a pump.

Some other things:

Looks like plastic tubing on the tank out to the house. Bad idea, particularly if/when the system starts producing really hot water.

I didn't notice any safety relief valves in the system. That's a big deal. Seriously. Are there municipal building/plumbing codes in force ?

The closed cell foam insulation such as you have in place is good stuff, but it needs to be protected from sunlight. Your stuff is deteriorating. I've found aluminum duct sealing tape (NOT cloth duct tape) lasts ~ 5 - 7 years outside under direct sun and does a good job of protecting the Armaflex type insulation. Your insulation is also shrinking. Insulation gaps will cause a larger decrease in overall insulation effectiveness than apparences may make you believe - a 5% area uncoverge will probably decrease overall insulation performance by a lot more than 5%. 20+ % or so is more likely.

[r.bartlett]

13:30

been up on the roof again. the outlet temp is 60c and return 29c tank up around 28c