Convert thermosyphon to pumped.

[r.bartlett]

Chaps I have a std Chromagen 200L thermosyphon roof top system which although works I believe doesn't work to it's full potential/. The principle means there will always be cool water in the tank which amount will differ depending upon the amount of sunshine.(obviously). It appears on cooler days the panel is very hot but the return is cold water temperature.?


we quickly run out of hot water when I believe the panel still has potential to heat. I'd rather have a tank full of 50c than 1/3rd at 80c. If you get my drift..

My thought was to introduce a small pump in the solar loop regulated by a thermostat to ensure a flow through the panel when the top of the panel reaches a certain temperature (ie 70c) .

I live in Tenerife so we experience 300ish days of sunshine but it's the cooler months coming up when there are a lot cloudier days which is bothering me

TIA

Richard

[J.P.M.]

The performance of thermosiphon type solar water heaters can potentially be improved by adding a circulating pump, but usually not a whole lot. Sometimes such a change can produce the opposite of the desired effect and/or cause unintended consequences and outcomes.

Until the water circulation rate in a forced circulation (pumped) system reaches a point where the flow through the collector risers starts to becomes turbulent, and provided the systems are functioning properly, adding a pump will do little to increase system performance. To get to a flowrate where the flow through the collectors starts to become turbulent requires some information about the collector design that is not immediately apparent from the product spec sheets, but I'd take a SWAG and say maybe a pump that would provide a flowrate of, very approximately about 3 - 5 gal./min (~ 0.19 to ~ 0.32 l/sec) ought to get you there. That's probably an order of magnitude or two more than a thermosiphon system will produce. Once there, expect a performance improvement as measured by increased energy output to be very approximately something like 10 % over the prior thermosiphon performance. Again that's an increase over normal and expected performance. If the current unit is fouled, blocking flow in some parts of the collectors or plumbing, performance of any system, but particularly thermosiphon systems, will suffer as the blockages have more of an effect on the flowrate of gravity induced (thermosiphon) supplied systems.

The increases in fouling will also lower the heat transfer coefficients in the collectors for either type system, but while related, that's a somewhat separate subject to how flowrate affects heat transfer.

If you add a pump, and it's not a relatively big one, that may well be counterproductive to your goal. If your current system is working nominally, and it's not too old (fouling being somewhat a f(time in service)), adding a pump will probably destroy any thermocline (temperature stratification) in the tank that's attainable with thermosiphon systems, and a big part of what contributes to their thermal performance. Unless the pump is of a pretty healthy size to meet a circulation rate as described above, what you'll do is increase output a small amount - but not much, but in the process, destroy any temp. stratification in the storage tank which contributes to the current system's performance, whatever that may be, increase the collectors operating temp., and so increase heat loss to the ambient conditions which will decrease performance.

Before you do anything I'd suggest you get the system checked out for proper operation and design. Sometimes folks unfamiliar w/gravity systems are unaware of some of the finer points of such systems, such as the need and placement of thermosiphon loops or check valves, etc.

BTW: Have you noticed a performance change recently or over time ? Or have your hot water demands increased ? Or ??

I suspect, again, providing the system is functioning nominally, and the idea that, being outside the U.S, you're not a grossly profligate waster of just about everything notwithstanding, that what's going on is the demands on the system are more than the system can meet. I guess that's a sort of no brainer, but the task is to find out why. A 200 l system is OK, but without backup by a relatively quick recovery (read fossil fuel fired) backup system, a long shower or big laundry load or combination loads will quickly bring system shortcoming to the surface.

From the product blubs, I notice that the gas backup available is an instantaneous type with a rate substantial input, but the electric backup is only 2.3 kW - that's not much. What are your backup arrangements to meet hot water demand ?

I'd suggest you consider seeing how/if/where you can cut back on hot water use. (Short) showers with low flow shower heads, flow restrictors in faucets, low water use appliances that sort of thing.

[r.bartlett]

Many thanks

so you understand me.

I live here in San Miguel Tenerife. I moved from the UK 2 years ago as part of my retirement plan I run/own a small air conditioning company. I had one in the Uk for the last 30 years.The solar setup was part of a deal with a customer who had it installed but was made to take it down by the authorities.



I will try to reply within the text below..

The performance of thermosiphon type solar water heaters can potentially be improved by adding a circulating pump, but usually not a whole lot. Sometimes such a change can produce the opposite of the desired effect and/or cause unintended consequences and outcomes.

the issue as I see it is the TS effect relies on a temperature difference. As that temperature difference drops -the water heating up- the effect slows. In hot sunny days this doesn't matter as the sheer amount of heat available means plenty of hot water. in less sunny days this -could be??- a double whammy. less heat available and less flow.

Until the water circulation rate in a forced circulation (pumped) system reaches a point where the flow through the collector risers starts to becomes turbulent, and provided the systems are functioning properly, adding a pump will do little to increase system performance. To get to a flowrate where the flow through the collectors starts to become turbulent requires some information about the collector design that is not immediately apparent from the product spec sheets, but I'd take a SWAG and say maybe a pump that would provide a flowrate of, very approximately about 3 - 5 gal./min (~ 0.19 to ~ 0.32 l/sec) ought to get you there. That's probably an order of magnitude or two more than a thermosiphon system will produce. Once there, expect a performance improvement as measured by increased energy output to be very approximately something like 10 % over the prior thermosiphon performance. Again that's an increase over normal and expected performance. If the current unit is fouled, blocking flow in some parts of the collectors or plumbing, performance of any system, but particularly thermosiphon systems, will suffer as the blockages have more of an effect on the flowrate of gravity induced (thermosiphon) supplied systems.

difficult to get any information from Chromagen. they are notoriously bad at replying and I am no exception. However if the tank was below the tank then a pump becomes a must. I guess I am only trying to replicate that.

The increases in fouling will also lower the heat transfer coefficients in the collectors for either type system, but while related, that's a somewhat separate subject to how flowrate affects heat transfer.

see above

If you add a pump, and it's not a relatively big one, that may well be counterproductive to your goal. If your current system is working nominally, and it's not too old (fouling being somewhat a f(time in service)), adding a pump will probably destroy any thermocline (temperature stratification) in the tank that's attainable with thermosiphon systems, and a big part of what contributes to their thermal performance. Unless the pump is of a pretty healthy size to meet a circulation rate as described above, what you'll do is increase output a small amount - but not much, but in the process, destroy any temp. stratification in the storage tank which contributes to the current system's performance, whatever that may be, increase the collectors operating temp., and so increase heat loss to the ambient conditions which will decrease performance.

Before you do anything I'd suggest you get the system checked out for proper operation and design. Sometimes folks unfamiliar w/gravity systems are unaware of some of the finer points of such systems, such as the need and placement of thermosiphon loops or check valves, etc.

I am eager to learn the finer points but I have yet to find a 'dummies guide to TS solar' . Please can you forward details . I will get some better pictures of my set up for comment.



BTW: Have you noticed a performance change recently or over time ? Or have your hot water demands increased ? Or ??

I suspect, again, providing the system is functioning nominally, and the idea that, being outside the U.S, you're not a grossly profligate waster of just about everything notwithstanding, that what's going on is the demands on the system are more than the system can meet. I guess that's a sort of no brainer, but the task is to find out why. A 200 l system is OK, but without backup by a relatively quick recovery (read fossil fuel fired) backup system, a long shower or big laundry load or combination loads will quickly bring system shortcoming to the surface.

From the product blubs, I notice that the gas backup available is an instantaneous type with a rate substantial input, but the electric backup is only 2.3 kW - that's not much. What are your backup arrangements to meet hot water demand ?

I'd suggest you consider seeing how/if/where you can cut back on hot water use. (Short) showers with low flow shower heads, flow restrictors in faucets, low water use appliances that sort of thing.

not sure it is fully functioning, we have a lot of hot sunshine which maybe disgusing poor set up. as regards back up. sadly the unit never came with a electric heater so atm it's on it's own. I have access to a small swimming pool HP rated at 2.7kw output which I was considering plumbing into the circuit but not sure where. There is a couple of 'drain back' plugs in the side but again I don't really know what they are all about.
IMG_20171204_080343 (3).jpgIMG_20171204_080349.jpg

[LucMan]

Installing a pump is not going to increase your hw production. As mentioned above there may be scale buildup in the heat exchager (most likely). As you mentioned the panel is hot but the supply line is cold= no heat transfer or no ciculation. This can be remedied by circulating a mild acid (vinegar) through the hx until clean.
Production can be increased by tilting the the panel as close to perpendicular to the sun at noon for this time of year.

[r.bartlett]

Installing a pump is not going to increase your hw production. As mentioned above there may be scale buildup in the heat exchager (most likely). As you mentioned the panel is hot but the supply line is cold= no heat transfer or no ciculation. This can be remedied by circulating a mild acid (vinegar) through the hx until clean.
Production can be increased by tilting the the panel as close to perpendicular to the sun at noon for this time of year.

you misunderstand me, the outlet from the heat exchanger (the return to the panel) is always cold. therefore IMHO the full potential of the HE is not being utilised even on the very hottest sun days .

[J.P.M.]

OK, time for a reset.

When you wrote "...std Chromagen 200L...", I assumed you had a direct circulation system. That is, a type that does not employ a heat exchanger, and so the potable H2O is circulated through the solar collector.

It seems that is not the case. My bad for assuming. That said, most of what I wrote previously about collector heat transfer and pressure drop as f(flowrate) still holds true.

However, My focus would now shift to checking out a couple things :

1.) A blockage or flow restriction in the coolant line. Density differences which are what make thermosiphon systems work at all, do not produce a lot of pumping power. An air or gas pocket that forms at the top of a loop in an unvented, closed thermosiphon system, depending on the size of the gas pocket, can stop flow dead. When was the circulating fluid last checked for level and Ph ? Last changeout ?

2.) If it's inside the tank, as it seems, a fouled (crudded up) heat exchanger with the fouling mostly on the outside surface (the tank side) of the heat exchanger.

I'd check the coolant lines for fluid quality and any blockage or air pockets in that part of the system - including the collector(s). Closed thermosiphon systems (the kind with HX's in them) produce more pressure drop. The resulting lower flow means hotter coolant temps. That means coolant will tend to break down sooner.

However, the big unknown is what does the outside ( the tank side ) of the heat exchanger look like ? A common cause of heat exchanger fouling is somewhat counter intuitive: Inverse solubility of some common, but not quite universal periodic table group IIA compounds, mostly of calcium and magnesium. That is, a lot (but not all) of them precipitate (fall) out of water solutions of hard, or semi hard water as the solution gets warmer, and on to the warmer heat exchanger surface. The thus deposited material has a poor thermal conductivity relative to the HX metal, and so impairs performance - lower heat transfer and hotter coolant temps. The longer the deposition process, the more performance is impaired. So, the retrograde solubility of things like calcium carbonate, calcium phosphate, calcium sulfate, magnesium silicate, and a whole bunch of relatively common components of hard water deposit themselves on warm(er) heat exchanger surfaces, acting as thermal insulators precisely where insulation is not needed.

All that said, it's not likely that such fouling would stop all heat transfer. If however, the coolant tank outlet line is close or, reasonably close to the coolant tank inlet line, that may indicate a fouled or scaled up heat exchanger as described above.

A.S.M.E. and most other code pressure vessels above a certain volume have mandatory inspection ports or nozzles for visually inspecting or perhaps getting at tank internals for servicing. I kind of doubt the Chromagen tanks have inspection ports, or arrangement for HX removal, but maybe I'm wrong on that point.

[J.P.M.]

you misunderstand me, the outlet from the heat exchanger (the return to the panel) is always cold. therefore IMHO the full potential of the HE is not being utilised even on the very hottest sun days .

Actually, if the outlet of the coolant line from the HX is cold, that would either indicate that the HX is clean as a whistle (unlikely), or that the flowrate of coolant is very low or zero. I bet you've got an air (gas) lock in the coolant line that's blocking flow.

[r.bartlett]

Actually, if the outlet of the coolant line from the HX is cold, that would either indicate that the HX is clean as a whistle (unlikely), or that the flowrate of coolant is very low or zero. I bet you've got an air (gas) lock in the coolant line that's blocking flow.

ok, so what delta t should I expect?

[Mike90250]

You first have to identify all the pipes and parts.
Is it configured as in the cartoon image in the first post ? Did any parts gets shuffled around, or hooked up in a different configuration ?

Insulated Storage tank
Flat Plate heat collector, aimed at solar noon
Is there glycol fluid in the exchanger loop, and is it purged of air ?

The "exchanger" is inside the round storage tank, and fluid leaving the collector, enters Hot, gives up it's heat
and returns to the collector cold.

I have 3 thermal siphon systems running.
1) backup generator thermalsiphon cooling
2) Masonry Heater has a hot water loop, storage tank has about 80 gal of 95F water
3) rooftop thermal (same style as your cartoon image) 40gal, All totally passive thermosiphon

If your glycol loop was ever filled with plain water and froze, that will ruin the system.
If you have air bubble, that will stop the system

So be sure what point you are measuring, if we are to help you, GI=GO

[LucMan]

Solar HW systems really don't get any simpler. If there is water in the tank and sun is hitting the HX it has to work unless there is an obstruction.

[J.P.M.]

Solar HW systems really don't get any simpler. If there is water in the tank and sun is hitting the HX it has to work unless there is an obstruction.

You mean hitting the solar collector, right ?

To the OP:

If coolant is flowing, the thing will work. However, it's not always a dichotomous will or won't work:

If flow is reduced due to a partial blockage, it will still work, but performance will be reduced approx. in proportion to how much the coolant flowrate is being reduced by the blockage.
And, if the tankside of the HX is fouled, the performance will be reduced approx. as the ratio of fouled to clean performance of the HX.

The temp. of the coolant line from the tank is not much of a smoking gun indicator of flowrate. A better way to think of using temp. as in indicator of flowrate is to think of it in terms of what happens to the temp. difference of the coolant from tank inlet to tank outlet under various conditions :
.
If the flow is unblocked, and the HX shellside is unfouled, the coolant temp. out of the tank will be less than the coolant inlet temp. to the tank and (hopefully) close(r) to the tank water temp., and life is good.

If the flow is completely blocked, and also depending on the state of check valves and piping loops etc. (if any are present), the coolant temp. in the line out of the tank will either be closer to the tank temp. (if reverse flow and coolant line temp. stratification is inhibited ), or closer to the collector temp. (if reverse flow and line temp. stratification is allowed)., but probably closer to the coolant inlet temp. (and again depending on check valve and stratification loop presence)

If flow is partially blocked, and the HX is mostly clean on the tank side, the temp. of the coolant out of the tank will tend to be lower than the temp. of the coolant into the tank. with that coolant inlet to outlet temp. diff. being somewhat inversely proportional to the flowrate - more flow --->>> less inlet/outlet temp. difference.

If the flow is partially blocked, and the tankside of the HX is fouled, the coolant temp.inlet to outlet difference will tend to be a function of both the degree of HX fouling and reduced flowrate; lots of HX fouling will reduce coolant inlet/outlet temp. diff. and lower flowrates tending to increase inlet/outlet temp. differences., with those two driving forces working somewhat against one another witth respect to inlet to outlet coolant temp. differences.

Restricted flow will also make the collector run hotter (and less efficiently), and closer to stagnation temps., but without more information such as collector temp. .histories (probably impossible to get or even hope for) that's an academic consideration. It can however, and FWIW, contribute to faster coolant degradation.

There are other things that need to be checked. Troubleshooting without being on site is a real crapshoot. At this time and from this location, my SWAG money is primarily on an air blockage, but also that the system needs a good blowdown (a thorough cleaning).

One thing the OP might try - WITH A LOT OF CARE ! - MAKE SURE ANY AND ALL THE RELIEF VALVES ARE FULLY FUNCTIONAL AND THE SYSTEM IS PRESSURIZED ! - is to purposefully and positively stop the flow and see what happen to the line temps. at various places. If coolant line temps. don't change (much), the flow is blocked. That's not very qualitative, and while not quite a smoking gun for complete blockage, it will at least indicate a large flowrate reduction tantamount to full restriction. If temps. do change by a full stoppage, the degree of temp. changes may be some indication of the degree of blockage, large temp. changes = not much blockage, and also that the HX probably may not be too badly fouled.

If it turns out that the flow is fully blocked, as I wrote before, my money's on an air lock.

[r.bartlett]

as I say it works and on hot days (today) we have very hot water

but i believe the TS system although simple is not enough . I liken it to a fan v open window. when the wind blows( ie the sun shines brightly) an open window works well. In less windy days (cloudy) the effect drops.

With a forced loop system like when the tank is below the panel ALL the available heat is transferred to the water. If the TS process relies on cool water coming out of the tank -delta T- then that means part of the tank water must be cool too. So to me there will MUST be stratification within the tank. or the TS effect stops from hand touch The water at the top of the panel is over hot especially compared to the panel bottom return..


If the water is forced through the panel & HE then there will be a more even spread of heat give up throughout the HE and the tank water should be a more even temperature. less stratification...???

the system should/does give hot water most days but I believe the TS system although simple is too simple and needs a little boost when the sun isn't as strong. I'd rather have a 200L tank at 55c than 1/3rd at 80 and the rest at room temperature. I believe a pumped system would do this??

Just to clarify in Tenerife we have one of the best climates in the world. In my part San Miguel the lowest temp I have seen was during the night 11c.(51f) the highest was 44c. Therefore we do not use glycol .etc.

https://www.tutiempo.net/san-miguel-de-abona.html

A massive thanks for all the posts so far, I will post pictures and actual temperatures at the weekend.

[J.P.M.]

as I say it works and on hot days (today) we have very hot water

but i believe the TS system although simple is not enough . I liken it to a fan v open window. when the wind blows( ie the sun shines brightly) an open window works well. In less windy days (cloudy) the effect drops.

With a forced loop system like when the tank is below the panel ALL the available heat is transferred to the water. If the TS process relies on cool water coming out of the tank -delta T- then that means part of the tank water must be cool too. So to me there will MUST be stratification within the tank. or the TS effect stops from hand touch The water at the top of the panel is over hot especially compared to the panel bottom return..


If the water is forced through the panel & HE then there will be a more even spread of heat give up throughout the HE and the tank water should be a more even temperature. less stratification...???

the system should/does give hot water most days but I believe the TS system although simple is too simple and needs a little boost when the sun isn't as strong. I'd rather have a 200L tank at 55c than 1/3rd at 80 and the rest at room temperature. I believe a pumped system would do this??

Just to clarify in Tenerife we have one of the best climates in the world. In my part San Miguel the lowest temp I have seen was during the night 11c.(51f) the highest was 44c. Therefore we do not use glycol .etc.

https://www.tutiempo.net/san-miguel-de-abona.html

A massive thanks for all the posts so far, I will post pictures and actual temperatures at the weekend.

For my part, you're most welcome. More later. This is not meant as an insult, but you seem to have a fundamental misunderstanding of how solar thermal collectors, both active and passive, as well as the subtleties of thermosihon systems work. Consult a solar energy text for more info.


I look forward to more information, including photos.

BTW, if you're not running a non freezing fluid through the collector loop (glycol/Water solution) because of non freezing conditions, the heat exchanger is not necessary, and is actually an albatross around the neck of system performance (but remember, you only need 1 cold snap and it's game over, and also keep in mind - things exposed to a clear, dry nite sky can fall way below the amb. air temp.).

[LucMan]

as I say it works and on hot days (today) we have very hot water

but i believe the TS system although simple is not enough . I liken it to a fan v open window. when the wind blows( ie the sun shines brightly) an open window works well. In less windy days (cloudy) the effect drops.

With a forced loop system like when the tank is below the panel ALL the available heat is transferred to the water. If the TS process relies on cool water coming out of the tank -delta T- then that means part of the tank water must be cool too. So to me there will MUST be stratification within the tank. or the TS effect stops from hand touch The water at the top of the panel is over hot especially compared to the panel bottom return..


If the water is forced through the panel & HE then there will be a more even spread of heat give up throughout the HE and the tank water should be a more even temperature. less stratification...???

the system should/does give hot water most days but I believe the TS system although simple is too simple and needs a little boost when the sun isn't as strong. I'd rather have a 200L tank at 55c than 1/3rd at 80 and the rest at room temperature. I believe a pumped system would do this??

Just to clarify in Tenerife we have one of the best climates in the world. In my part San Miguel the lowest temp I have seen was during the night 11c.(51f) the highest was 44c. Therefore we do not use glycol .etc.

https://www.tutiempo.net/san-miguel-de-abona.html

A massive thanks for all the posts so far, I will post pictures and actual temperatures at the weekend.

Could it be that you are expecting to much from your system for this time of year?
For a better understanding of solar thermal performance for each month try downloading the free trial solar thermal assisant software from Solarpathfinder.com.
Enter a system close to the one at your site, and you can then see how performance varies for each month.
Good luck.

[Mike90250]

I find that's very true at my site, my SHW really cooks along in summer, but does little in the winter, because the glazing is almost at an angle for a mirror. But winters, my wood heat is going and that gives me oodles of hot water