We have a fairly standard solar thermal system commissioned this time last year as part of a building project. The system is evacuated tubes (SSE-facing pitched roof, no shading) feeding to the bottom coil of a 250L HWT, controlled by a Resol Deltasol BS controller. But it doesn't seem to be working as it should, and I am looking for suggestions as to what might be the problem.

The specific problem is that I don't get water as hot as I think I should. Even in the hot and very sunny weather we've had recently (this is Surrey), the temp of the collector on the roof never exceeded 53C, and the water at the bottom of the HWT was closer to 40C. I have experimented with reducing the circulation flowrate from 6 L/min to 2 L/min, which got us to 55C on the roof, and then I increased the temperature differential at which the pump kicks in to 20C (ie collector 20C hotter than HWT before heat is removed from collector) allowed the collector to 60C today... but the HWT was still close to 40C. All this means that except on the very sunniest of days in midsummer, I have needed to use the boiler to get the HWT up to temperature. That can't be right: my tubes simply don't seem to be collecting enough heat. So does anyone have any suggestions as to what might be the problem?

[Regarding installation, builder employed installer, we then fell out with builder, as did installer (he didn't get paid). This has consequences for come-back against installer whom I did not contract directly - I have asked them to come and have a look, but am not confident of being able to force them. The main practical consequence of this dispute was that the v long pipe run from tubes on roof to HWT (c15m) apparently wasn't fully insulated (installer stopped work, builder boarded over wall). That obviously doesn't help, but may or may not be relevant to this problem.]

Thanks in advance for any ideas...

Brian
Collector is 30 tubes, but I don't have any info about it otherwise. We often use virtually no hot water during the day.

The main thing is that I have absolutely no idea what sort of temps I should be seeing in the collector on a sunny day. If I had been seeing very high temps combined with low temps in the HWT I would immediately have assumed that excessive heat loss in the ong pipe run between the two was to blame, but the collector never gets that hot (except once after a power cut, where with no pump running the temp reached over 130C).
j

Thanks Megacycles. Will do some measurements tonight, although the pipe between the collector and control unit (which is next to the HWT) is at least 20mm.

Regarding the tank being too large, I was under the impression that most solar thermal systems produce masses of excess heat in the summer, so that at this time of year it wouldn't be an issue.

It looks there are two possible faults, 1, the tubes are not transferring their heat to the water or 2, the water is loosing the heat in the extended pipe run.
So as for tube faults, is it possible to mount the tubes upside down? or not tighten them properly, as I understand it the heat from an individual tube is conducted via a copper thread into the water filled manifold? Should there be some sort of thermal paste applied to these threads before assembly.
Loosing the heat fault, this looks more likely. You have some temperature measurements are these built in to your controller or are they what you have measured with your own bit of kit, because if the water is leaving the collector at 55 degs and arriving at the tank at 40 then you truly have some pipe losses. Sounds to me that you should be able to find warm spots on the walls where there should not be any! That long length of insulated piping looks like the problem to me. So you have 15m out and 15m back ! so to my mind you could be loosing 10 deg on the return and 15 degrees on the flow to the tank which, if fixed could give you a nice extra 25 degrees or so.
This "boarded over" pipework, can you not get to it anywhere? If its run in the cavity of plaster board walls, are they decorated?
Again thinking out a loud, I wonder if you locate the pipe run with a metal detector/stud locator, you could drill a series of 1/4" holes into the plasterboard cavity and inject expanding foam. Its a very dodgy thing to do as its VERY likely to bulge the plasterboard as it expands. A better way would be to fill the cavity from the top, by drilling a 1" diameter hole into the cavity and filling it with polystyrene beads, but some very long drills will be required to get through the noggins.
Frank

Power = 4186 * flow * Temperature uplift

Power is in Watts
Flow in Liters per second
Temperature in C

If the uplift is 20C
flow rate is 2L/M = 0.033L/S = 0.033Kg/S approx
then the power collected is

= 4186 * 0.033 * 20
= 2.7KW

But is the flow temperature 60C or is that just the "collector temperature"? Ideally you want some way to measure the actual water flow temperatures rather than pipe or collector temperatures.

You can also work out how much power is going into the tank the same way, just measure the temperatures at the flow and return near the tank. The difference between what's collected and what's going into the tank should be the loss in the pipe runs.

Thanks Megacycles and everyone else. My pipe run is actually 22mm copper, so presumably heat losses will be greater (curious - how do you calculate?). It sounds more and more as if that lack of insulation is the cause.

Not sure if this is an issue but with 2L/min down 15m of 22mm pipe it appears the water takes just over two mins to get from the collector to the water tank.

I do think you probably have at least 3kW to start with. And losing half of it. If you fix that you'll have a surplus on hot days.

Hi,

I’d be concerned about the volume in that pipe. At 15m and 22mm pipe you have just under 6lit in each pipe. When the panel is hotter than the tank and pump comes on a large slug of cooler water (tank to panel leg) is going to go in. This is exacerbated by being cooler than necessary due to lack of insulation. Thus, the panel may not now be sufficiently hotter than the tank and switch the pump off again. Likewise the slug of cooled water in the panel to tank leg will go into the tank coil leaving the hot water from the panel in the leg, where it cools. So forth and so forth.

It sounds like an complex pipe run so perhaps that’s why it was done on ordinary domestic heating pipe rather than the dedicated (and very expensive) smaller bore and pre-insulated pipe more usually used for panel.

Don’t forget the coil in the tank would be a 22mm (finned) pipe and could be 15+m in length. Therefore, you have a circuit total of about 18lit.

Following on from a comment above how long does the pump run each time?

Cheers
Mike up North

Turning the boiler off for 4 months is a myth, you can easily turn if off for 6 months with occasional immersion top-up provided the solar system is installed correctly and the house is well insulated/draught proofed. I'm in Perthshire 200m ASL and a sunny day in February will produce a tank of 50'C hot water although my system is deliberately oversized to cope with the variable conditions here - 60 x 47mm tubes initially heats 180litres and then a 120litre secondary tank.

It so disappointing to see a professional tradesman make such a fundamental mistake of using 22mm pipework,
along with dodgy ex DG salesman selling overpriced and overhyped systems, it makes it an uphill struggle convincing the 'man in the street' that this technology really does work when done correctly.

Thinking positively, the hard work has been done ie the panel and tank are in place, and they are well matched so theres nothing to stop this system working well (you easily have an extra 10-20% insolation). It easy for me to say as I can do the work myself, but if your going through the hassle of exposing the pipework I would remove it and replace with 10mm which is flexible so easy to get round the various corners. I think your always going to see poorer performance because of larger losses and the large system volume, and if you intend to stay in the house the pleasure of a 'solar-bath/shower' in March would soon erode the memories of the hassle. If you can't do the work yourself expose the pipework that needs replacing and get a decent plumber to do the work in December or find a nocturnal one, it shouldn't take more than a day (and send the bill to the installer/builder).

Yes, the uninsulated stuff will limit the peak temperature noticeably.
I just remembered that I took off a small section of insulation from mine last year to stop it boiling the tank. Like a heat-leak radiator. Still seems to work fine though.