Ok, so the pump is the lowest part in the complete system.

The pump is also pushing the water up into the system, not sucking the water down, so I think I am pretty much coverd wrt the airlocks.

Even if air does collect, it will just go out via the hot water taps.

I will keep an eye out for that problem

 

I did have a look at putting the non return in the cold line, but could not find a good location to do so.

It is the flap type, so no spring return. This means it must be mounted vertically for the flap to operate correctly.

 

Ok, so seems like SuperDIY nailed my coffin!!!

Trying to figure out why my pipes would melt and burst over the week-end, I saw that the pump was running, but the collector temp kept increasing.

I sped the pump up to max, but still no luck. Eventually the pipes got so hot they just popped!

 

Anyway, as I did not have much time, I removed the non return valve completely and moved the air release valve to the last section.

Yesterday all seemed fine again.

 

For no I have covered the 2 flat plates with 80% shadenetting, as the vacuum tubes supply MORE than enough water for both geysers .

Right, so I did notice that the solar collector temp stays very high during the night now, since removing the non return valve.

So I installed the non return now just before the water inlet to the solar collectors.

So far all looking ok.

 

I must say I was extremely impressed on Sunday. As you know we had storms and rain all day Sunday.

Water temp still increased from 45 to 54 for the day. That is 9 degrees increase !!!

Right,

 

So I have been playing around with my own solar water heating system now for a couple of years and this is what I have learnt to date:

 

1. Vacuum Tubes is WAY better than Flat Plate collectors.

Very true. VT tech is by far superior

2. VT orientation is not that important compared to FP.

Interesting I didn't know that

3. Tilt angle IS important. The more square you can get the sun in cold months, the better. It will also prevent overheating in summer.

Should look into this. Struggling with overheating in summer and not enough in winter

4. Thermo siphon works when you follow all recommendations.

It also works in reverse, so when it gets cooler at night, expect your water to cool down rapidly unless you put in a heat trap or non return valve

5. Problem with thermo siphon is the location of the geyser and the frame to mount it on. Also the circulation effect only starts working at a high temperature.

True

6. With the geyser located outside, heat loss is a problem.

True

7. With the geyser mounted outside, hot water piping runs can be very long, resulting in long wait times for hot water.

Likely

8. Work on 300L hot water storage for 2 persons.

We have about 350lt for 4, works fairly well

9. 24 Vacuum tubes is ok for 200L geyser.

Yup

10. 2,5m2 Flat Plate collector is not really enough for 200L geyser.

Flat plate is a waste of time and money

11. Direct systems works way better than an indirect system ie geyserwise controller with the VT not connected to the geyser.

12. When your geyser is outside, insulate your pipes against frost or you will have no hot water in winter and have the added benefit of a Bomb on your roof when your element turns on and there is no room for expansion. Make very sure you don't have a pipe connected to the safety valve. It can run straight onto the roof and you don't want the last resort to also freeze up.

13. I have my geysers in series with the one geyser pre-heating the water and the final geyser receiving hot water. This works very well, as you eliminate thermosyphon between the two geysers, if your temperature drops below 55 degrees and the element switches on, you have less water to heat back up and therefore use less electricity. 

14. Also add a safetyvalve or solar relief valve to your tube array if it is separated from your geyser. Pipes takes much longer to defrost that the small amount of water in the VT tanks to boil and if it has no where to go... Boom. And even insulated pipes freezes up from time to time.

Regarding the water freezing in pipes....

The European type controllers have an antifreeze function built-in.

Basically if your outside collector temp reach 4 degrees, it will start the pump.

 

I must say, I always believed thermoshypen to be the best for solar heating,

but now I know using a pump and correct controler gives so much better results!!!

I have a VT system with 200 lister geyser. The geyserwize controller I have also has a setting to allow water from freezing. Its stupid for our climate as in the last 8 years I have never had water freezing in the collector on the roof so in winter it stats to cycle at night and then starts to pinch the hot water.

Yeah, that is one of the most frustrating settings. Nothing worse than the geyserwise stealing your hot water every 30 mins to keep stuff from freezing, except the tank bursting lol.

So THAT is why I lose temp in winter on geyser!

I have a prefeed system. XStream fibreglass tank with a flat panel (SolarMax, I believe the factory is in Kleinmond), pointing slightly north-east (not ideal, but that's the way the house was built). From the sound of things, switching to evac tube is going to make a massive difference. I wonder what it would cost to retrofit that.

 

Alternatively, I could leave that system as is, put up evac-tubes on the North-west side (stupid low-pitch roof with very little space), and pump that into the main geyser (that's fed from the XStream). This will obviously cost a bit more.

 

Finally... I wish I could get hold of the temperature probe of a geyserwise (or something like it) to stick into the XStream tank and wire it up to an Arduino for plotting some charts. Because other than seeing a huge difference in electricity use between winter and summer, I actually have no idea how it is performing. Anyone know of such a thing, something that looks like a geyser thermostat and fits in the hole, but is really a thermistor? :-)

Anyone know of such a thing, something that looks like a geyser thermostat and fits in the hole, but is really a thermistor? :-)

 

I'm using LM35's or you can use LM335z's or even this: http://www.ebay.com/itm/180970372664

I think this is the same as the ebay one. I've used the company - just had delivery of two Arduino Mega's today. Building my own Differential Controller.

https://www.robotics.org.za/index.php?route=product/product&path=57_131&product_id=320

 

OKay, I have to admit I have never (had to) replace the thermostat on a geyser, so I actually have no idea how it works. I know what they look like, they have this long peg that's responsible for sensing the water temperature, and then on the back the have the small adjustable pot where you set the temperature. Now I don't know if that long peg actually sits inside the water, or if it sits in some kind of pocket, or what (basically, the answer to the question "do you have to empty the geyser to replace the thermostat" would answer all my questions).

 

So in theory, if I took a bog-standard thermostat, stripped the guts out of it and screw an LM35 to it, I have a temperature sensor :-)

 

Right? :-)

Yes, the standard thermostat slides into a, as you call it, "pocket" - it is not in direct contact with the water and thus you don't have to empty the geyser before you pull it out. The thermostat is about 6mm diameter, so I doubt if you will get a TO92 LM35 inside the exsiting probe, unless to "trim" the LM35 to a smaller version , but you can perhaps cut the probe's tip off and run the wiring inside the probe and add the LM35 to the tip of the probe, maybe use some Pratley steel and shape it to the same diameter as the probe.

Yes, the standard thermostat slides into a, as you call it, "pocket" - it is not in direct contact with the water and thus you don't have to empty the geyser before you pull it out. The thermostat is about 6mm diameter, so I doubt if you will get a TO92 LM35 inside the exsiting probe, unless to "trim" the LM35 to a smaller version , but you can perhaps cut the probe's tip off and run the wiring inside the probe and add the LM35 to the tip of the probe, maybe use some Pratley steel and shape it to the same diameter as the probe.

 

Okay, that is good news. I wasn't in the mood for getting all the way up onto the second-story roof to investigate. I would expect that the shaft of the thermostat serves as a conductor to conduct heat to some kind of bi-metal strip that acts as the actual switch. Placing an LM35 in the same area should do the trick then. I have no idea, I'll have to check. Weekend project (as if I don't have enough of them already! I have a garage door that is halfway sanded down that's probably going to consume this weekend...)

You can also try to buy tempreature probes that is used with new solar installations.

These probes is the correct size required.

Normally they either PT1000 or PT100.

I got some extra probes from Sunscan some time back.

Thanks Wetkit. That might just be the way to go.

 

In other news, I finally relented, bought the cheapest belt sander I could lay my hands on, and finished the garage door project on the weekend. (Orbital is nice and all, but so much slower). It needs a last coat of Rystic Timbacare... but then I am officially done with one weekend project.

 

Next up... the well point pump needs to be re-fitted and I need to finish the opto+Triac circuit that's going switch it on from the "master valve" output on the ancient rainbird irrigation computer.

 

And there's the old Microcare MPPT lying in a box that now works as long as you stay under 180 Watts. Above that it resets... which I suspect is because it attempts to turn on the fan. However, the BD139 transistor seems fine and there's no shorts... so that one is still a mystery. Maybe I'll do the solar water heater temp probe before I get back to this.

 

(There's also the long LONG range idea of ripping out the PIC-based controller and making something with an arduino... because the MC is neatly split into two boards, one that does switching and current sensing, and a separate control board... hehehe... if I had time for all my hare-brained schemes).

Right, some more feedback regarding my system.

The flipping weather is playing havok. I have been running the full summer with the 80% shade netting over the 2 flat plate collectors.

Now since April 2016 I have removed the netting as we frequently had overcast days. Problem is the water still came out too hot and have been causing airlocks!!!

So for now I will look at cleaning the pump and installing a 2nd steam release valve.

 

Also been thinking about a heat exchanger so I can use the excess hot water to heat the pool water. Either that or to shed the extra heat in my irrigation water tank.

 

So far I have only switched on the electrical element to test my setup. To date never needed it

So removed the pump yesterday and it was full of gunk!!!

Used an old bottle brush and all is looking fine now again.

Also installed a 2nd steam release valve and relocated my water bypass line. Will see if that works better.

The guy an SunScan where I got the steam release valve from said that the Wilo pumps does have this problem of collecting gunk and should ideally be cleaned every 6 months.

I only installed my solar geyser towards the end of the summer, so still need to see the impact of overheating

I know the Geyserwise controller can open/close a solenoid. I am playing around with the idea of linking it to high temp solenoid value, that can fill up the pool. So in summer when the geyser hits 80C, it opens the valve, fills up the pool with hot water and cools the geyser at the same time.

 

 

Nice Wetkit, my situation is a 12 tube on a 150l geyser (replaced a flat panel), but unless absolute perfectly positioned, a evacuated tube setup battles to replace a thermosiphon system by itself. When ours was first installed the temperatures in the collector where reaching horrific highs (nearly boiling even on cloudy days), but geyser was below 45... It may cost a bit extra, but install a pump (even if you think you don't need it), install a safety mixer (prevents water over certain temp from getting to taps/shower), and a PROPER escape for overheat (last year the geyserwise measured over 90 at one point - deadly to people and to your geyser! I don't even make coffee that hot!).
I have a small 12v pump on the Cold side, connected to the same solar panel I use to vent my roof (1 x 20W panel, runs the fans and the pump). It just needs to move the water a bit, even slowly, because thermosiphon is a great idea when it works, but small things can stop it working and make it useless. The advantage is that it auto stops when there is no sun on the tubes... I think it's a win/win.

Right, so this past Sunday was a nice sunny day, but not very hot.

I took the following readings at 08:00 and then again at 18:00, so 10 hours.

My top tank increased from 49 to 66 degrees, 17 degree increase.

My bottom tank increased from 22 to 59 degrees, 37 degree increase.

So if each tank is 200L, how much energy did I put into my water????

3.95kwh and 8.5kwh respectively.

Sent from my GT-I9195 using Tapatalk

Had to switch on the electrical element today for the first time.

Electrical system made a total of 4,8kwh

On 11/6/2015 at 0:26 PM, superdiy said:

Yes, the standard thermostat slides into a, as you call it, "pocket" - it is not in direct contact with the water and thus you don't have to empty the geyser before you pull it out. The thermostat is about 6mm diameter, so I doubt if you will get a TO92 LM35 inside the exsiting probe, unless to "trim" the LM35 to a smaller version , but you can perhaps cut the probe's tip off and run the wiring inside the probe and add the LM35 to the tip of the probe, maybe use some Pratley steel and shape it to the same diameter as the probe.

I know this is an old topic, but thought I'd reply in-case someone else needs to know the answer to this riddle, and since you didn't give the answer

A waterproof DS18B20 [*1] fit inside the thermostat pocket of standard heater elements, so any TO92 probe, like an LM35 will work as well. There's enough room to heatshrink the pins with some "4 core comms cable" as well if you want to use an analoge temp sensor. 

[*1] http://www.netram.co.za/1624-waterproof-temperature-sensor-ds18b20.html

 

Ok, so my system been performing very well to date.

But as luck would have it, the day after I left to go work in Nigeria, the house power starts tripping. To make a very looooong story short, the fault was traced to the Wilo pump. Basically the pump was end of life and had worn through the housing. This caused water and electrics to mix and trip the house E/L.

So once the pump was disconnected, what happened next? Well, the pipes overheated and burst

By the time I got back the plumber had to replace the pump and my plastic pipes with copper. $$$

#Schoolfees paid!!!

Right, just some more feedback to date.

System running very well, but I have to keep cleaning the flap type non-return valve and pump every 3 months. Seems like the calcium is depositing too quickly.

For now the copper pipes is working very well and I can run the water temp up to 75Degrees now.

 

To try and use less water, I have installed a hot water loop and pump for the house. It works, but I am unable to stop the water circulation.

Anybody done anything like this before?