RyanF

Hi BritishRacingGreen, I have them (also DS18B20) straight on the copper pipes - at the geyser inlet, at the geyser outlet and one on the solar panel inlet. Previously I had the solar thermometer in the pocket provided, but found out after some time that it was over-reading significantly. I guess it must have been in contact with metal inside the pocket that was heated by the sun, but wasn't influenced by the water circulation. Now it is straight on the copper pipe and readings are more beleivable. I just tape them on with electrical tape and cover with elastomeric insulation.

Hi Derek. Some thoughts: I know the SunPays swears by their suggestion, but my testing showed that heat traps do not work, you have to have NRV's on both lines. Keep in mind this is a pumped (not thermo-syphon) system. If you feed from the hot side, at night the temp differential between top of geyser and solar panel is greater and therefore reverse thermo-syphoning is more likely to overcome the spring type NRV. I don't have any problems with my system now, as shown in the diagram - running for over a year and no thermo-syphoning up to now. I do agree though that having the pump on the 'cold' side is better and I plan to move my pump to that side. The main reason for this is that if your panel overheats due to a circulation failure, the steam is forced thru the pump. And a friend had his pump burn out in this situation. So I plan to move my pump to the cold side and put another flap-type NRV between it and the panel to further protect it. As @P1000 says, the system is closed and under equal pressure throughout, until a tap is opened. So when the circulation pump starts, it pulls from whatever side of the geyser you have chosen (cold in my case) and pushes to the other side. Regarding control of the system, that is where I like the SunPays approach. No controls = no electronics = less things that can go wrong! It is true that a controller might introduce some small efficiencies, for example allowing the geyser to settle for the hot and cold water to separate as P1000 describes. But from observation of my solar panel temps in a Joburg environment with ample sunshine, the pump would switch off for 2 or 3 minutes at most and then go straight back on. The last couple of weeks I have been experimenting with only starting the circulation pump in the morning when the panel temp gets to 40 and switching off just before sunset. I have noted very little overall improvement in efficiency. BTW I have a Wemos D1 running Tasmota which controls a contactor to switch the element on if the geyser temperature requires it. This same little controller reads the temperatures and lets me keep record - screenshot below of what I see on my network for this controller. This is fun for me because I like tinkering with electronics and data. But if I do an installation commercially like this, unless the Client wants the wifi control, I would just use a standard DIN rail mount timer with a 2-pole contactor to switch the geyser element on at the time(s) the client wants. This way components are simple and easily replaced - doesn't require an electronics technician or programmer.

@PietpowerAgree on this point - and that is what the graph in the June 25 post shows very clearly. Once convection flow was stopped with two non-return valves on supply and return legs from the geyser, the panel temp drops below the geyser temp early in the evening. The panel temp starts going down at 5pm and was cooler than the geyser by 8pm. I guess the panel doesn't reach ambient temperature immediately on this installation because the pipe and manifold are very well insulated.

Sorry for the long delay @conradl Below is a simplified diagram of what I have. Hope this helps.

I think it is possible for the panel to be hotter - because it is higher than the geyser. Simple convection law is that the hottest fluid rises to the top and the coolest to the bottom. The pipes in the panel only hold a few litres of water, so if there is no NRV to stop the water flowing up there at night, the hottest water rises there. That is what the data seems to prove - temperature probes at 3 different levels: Highest probe is in the panel manifold which is approximately 2.5m above the bottom of the geyser - hottest temperature. Middle probe is at the top of the geyser, hot water outlet. About 2m below the panel manifold - middle temperature, but closer to the panel temperature as the bulk of the hot water is in this 'zone'. Bottom probe at the cold inlet of the geyser - coolest temperature. So with incorrect or no NRVs, the panel and the geyser are effectively one system, joined by the piping - hottest water goes to the top, coldest sinks to the bottom.

Because for that data set/graph, I didn't have the spring type NRV in yet on the cold side - flow direction from geyser to panel. Once I installed that NRV, this graph looked much better - see below. As you can see, panel temp dropped quickly between 17:00 & 21:00 to below the geyser temp - and I was a much happier man.

@AntonySo after trying two depths of heat trap loops (150mm first, then 600mm on both inlet and outlet), they really don't seem to work much (if at all). After the heat traps, the next iteration was to install a flap type NRV (horizontally positioned so the flap would close under reverse flow conditions) on the pipe which flows from the panel to the hot side of the geyser. I expected that ALL of the thermo-syphoning was going out the top of the geyser and the flap-type NRV would sort the problem. To my surprise the thermo-syphoning continued and when I had a close look at the data it was clear that it was going out of the cold side of the geyser up to the panel during the night (by this stage I had installed a temperature probe on both the inlet and outlet of the geyser). Some posts suggested a spring type NRV on the cold side, but I was not convinced because during the day the water has to flow in the same direction as the thermo-syphoning would go at night. But to my surprise it stopped the thermo-syphoning - I guess the spring is just strong enough to stop the little pressure 'bursts' which characterize thermo-syphoning - but not strong enough to prevent the pump flow during the day. For the fist time the panel temperature dropped well below the geyser temp for the whole night! So my experience is that you need two NRVs - one on each line, but the cold side one has to be the spring type. Probably worth trying before re-piping 😰. Just be careful to get the NRV directions correct so you don't end up with a closed loop 🥵 FYI - Below is a graph of the data when the thermo-syphoning was happening out of the 'cold' side of the geyser. The blue line is panel temp, the grey line is geyser hot side temp and the orange line is geyser cold side temp. The more regular 'up-and-down' of the orange line from around 23:00 is the small bursts of pressure that thermo-syphoning produces as it forces the hot water along and it is replaced by cold water. Overall you can see that the panel is hotter than the geyser during the night and they all have the same downward trend indicating the heat is being bled out of the panel.

Sorry for the long delay in replying @phil.g00. I have data from a few more iterations which I will post as I have time. Regarding the main conclusion of the 150mm heat traps not working, that seems evident from the panel and geyser declining in temperature at exactly the same rate during the night hours. The two have almost the exact same temperature and downward trend when the ambient temp it is warmer - April 24 to 27. But this becomes even more apparent when the ambient temperature dropped on May 2 to 4. For those nights the differential between the panel and geyser increases to more than 10°C (with the panel being warmer!!!) but the downward trend is exactly the same. But with the correct setup (minimal to no thermo-syphoning), once night comes, the panel temp should drop below the geyser temp fairly quickly - the time for that temp drop depending on the ambient temp. But as you can see from the graph this is not the case - the energy from the geyser is bleeding to the panel - keeping it nice and cosy on the roof! 😝 Regarding the secondary conclusion about the pump head reducing the flow rate and thereby constricting the energy harvesting, that also is most evident once the ambient temperature fell on May 2 to 4. During the day when both the panel and geyser are trending up, you would expect to see a much closer correlation between the panel temp and geyser, not a differential of >25°C . The pump data sheets shows a flow rate of 3L/min with a 2m head - that therefore seems woefully insufficient flow to harvest the available energy. Obviously the panel heats up and cools down much quicker than the geyser, given that it is a much smaller amount of water. However the faster the flow rate the more energy is transferred from the hotter manifold to the colder water. Since the geyser is much better insulated and a larger volume, you obviously want to harvest the energy as soon as possible to get it into the geyser where it should cool slower. Hope that makes sense? But data can be interpreted different ways, so looking forward to any other insights. The next sets of data also seem to debunk another couple of myths floating around, so more coming soon...

Thank you @phil.g00 Further to the questions above, I am trying a few modifications on this system and recording the data, which I will post for those who are interested. So first graph below is the following: 24 Evacuated tubes working with a 150L geyser. 12v DC Solar pump with head of about 2m. Reverse flow as recommended by supplier - Water from geyser outlet (hot side) to solar panel, return comes in cold side. Heat traps of 150mm deep, on both supply and return lines. No non-return valves. No electrical heating during this period, with hot water being used by occupants in the late afternoon. May 2, 3 & 4 was the first cold snap in Joburg, sunny days, but much colder incoming water, hence the larger differentials. Data has been scrubbed - some days the system was being worked on and data was removed. Here is what I could pick out the data: Average over the 7 days of heating gain during the day in the geyser was 22.7°C. Average for heat loss in the geyser overnight was 18.4°C. From the above it was clear the heat traps were not working. Noticed too that the geyser would only rise about 25°C each day, even if the panel was considerably hotter. So it seemed that, despite the pump constantly working during daylight hours, the head was too much, resulting in a low flow rate.

Really enjoy reading all the feedback on this forum, very educational and nice that it is specific to SA conditions. I have many years of construction experience on larger commercial projects. Doing some smaller jobs at the moment and installed a few solar water heating systems, all purchased from TheSunPays.co.za. Their standard kit has no controller, only a 'dumb' 12v solar powered circulation pump. Of course it makes sense to at least install a timer relay on the geyser element to ensure you get a reasonable amount of solar energy harvested during the day. A timer is uncomplicated, much cheaper than a differential temperature controller and can be serviced by your average electrician. So overall this seems like a good approach for your average customer who has no interest or time to maintain a controller. On the most recent installation (for a family member), I built an arduino controller to record temp data, and it also has two relays for geyser element and solar pump control. Got the first few weeks of data now and seen a couple of things where I need to make adjustments for a more efficient system. So after all that blah, blah history, my questions: The guys at TheSunPays provide both a typical SANS installation diagram, plus their own recommendation which is reverse flow i.e. water from hot side of the geyser goes to the evacuated tubes, and the return water goes in the cold side of the geyser. I spoke to the one guy and they swear from experience that the reverse flow is better. See their diagram attached below. Anybody that has this 'reverse' installation and some longer term experience with it? Good, bad, ugly? Their diagram also wisely shows the use of heat traps to prevent reverse thermosyphon. Seems a heat trap is a better long-term maintenance option as opposed to non-return valves. Does anyone have experience on how deep the heat traps on both sides need to be to be effective?