So, as started in another thread, a bit more delving into what has happened to this solar thermal collector that has seen its ability to harvest energy drop off quite suddenly.

This evening has involved building a 'climbing frame' (calling it a scaffold would be stretching it a bit...) up the roof and around the collector so I can get a right look at it and decide where to go next. Something definitely seems to be awry inside it as there are a couple of things that look like they are meant to be tubing / pipe clips, made of wire, loose and visible around the edge of the aluminium sheet with the selective coating on it. That sheet also seems to have slipped down a bit, as there is a gap at the top of the panel, where you can see by the sheet to the insulation behind it, and this gap is visible at the top, but not the bottom. And in one corner of the sheet, some of the selective coating has come off in what I guess must be flakes, revealing the aluminium underneath

As its an inroof install, deciding what to do next needs some careful consideration. Options;

• Reverse the installation process. This involves taking off the cemented on clay ridge tiles, about 3 rows of Welsh slate and then the top flashing / hood to get to the top pipe connection and brackets. This will be reasonably painful, as those top 3 rows of slates are cut into some unusual shapes to fit round the flashing, and being the last courses, will only likely have one full course, the rest being cut down ones, with a reasonable amount of "creative fixing". Once released, lower the complete panel to the ground for further investigation.


• Cut access holes in the top flashing / hood to gain enough access to disconnect the top pipe and brackets, then lower the complete panel to the ground. Saves disturbing ridge tiles and slates, but then a dilemma on how to rebuild the flashing on completion again. Its made of aluminium (alloy) sheet and has quite a complex set of folds and then folded / rolled seams. Trying to release it and getting it to bend up will likely just result in creases all over the place. Cutting access holes will probably keep it flatter as a base for a subsequent repair.
  That could be done with lead, suitably patination oiled, but difficult to beat into shape if its over aluminium.
  Else get some more of that brown coloured sheet and make a panel that covers the whole existing flashing, holes and all


• Try and split the panel up there - Lift off the glass and frame, and do the investigation in situ. Tricky as there is no obvious and accessible fixings that hold the glass frame onto the body of the panel. There is also the fun then of handling the glass and frame on a pitched roof, and finding somewhere to set it so that it remains put and doesn't get broken.


• Take the glass only out of the frame in situ, after popping out the bead seal that is holding it in. In theory, this is straightforward, as 100% of that seal is accessible without disturbing anything. However, there is then the fun of handling a large sheet of glass on a pitched roof, with no frame to protect it. Chances of breakage probably about 1000%!

I think I am leaning more towards option 2 - Cut access holes in the existing flashing - then get a sheet of aluminium flashing material and cut a piece that can still slip under the slates above, and be suitably fixed to the existing flashing, but in a way that can be readily removed again in the future. That then becomes a full panel down to the ground job, which is slightly challenging given the 6 Velux windows in a bank of 2 x 3 underneath it. But the panel went up after the windows went in, so I just need to reverse that process. My 'climbing frame' already involves 2 x 6m long lengths of 6 x 2 up the sides of the panel / Veluxes, so these can act as a railway track to slide the panel down on.

Another scorching hot day today here in Orkney, and the store temperature is still sitting at just 52 degrees. So something needs to be done!

Posted By: Bryan_RendallCutting access holes will probably keep it flatter as a base for a subsequent repair.
That could be done with lead, suitably patination oiled, but difficult to beat into shape if its over aluminium.
Else get some more of that brown coloured sheet and make a panel that covers the whole existing flashing, holes and all

Option 2 sounds like the way to go.
Cutting access holes is probably the best option, over here repairs to ali. flashing is done with the same ali. flashing formed to as near as possible then pop riveted (use ali. rivets) in place with silicon sealer to provide the water tightness. Apply the silicon sealer between the flashing, then the pop rivets.

Another possibility might be to drill access holes with a, say, 100mm holesaw, then plug later with a blind grommet style plug. A quick search shows such a plug might be available, and that could be a possibly tidier repair than aluminium sheet over the top, though the plugs will be visible from the ground. I can't quite recall the build up of the flashing behind this hood, and if that point needs to be watertight to stop water getting in and behind the tray that the solar panel sits on. I guess once holes are made and the panel is out of the way, it will be possible to look up under the flashing to see how well the holes needs to be sealed. If plugs are not viable, then sheet over it with ali as suggested by Peter.

From your description of a hot sunny day up there it seems to me that even a poor collector should be working reasonably well in those conditions. Are the clips you mention intended to hold the pipes to the copper backplane? Seems a poor idea if yes. Soldered or brazed I would have thought

Is the circulation OK? Is there an easy way to check the amount of coolant or water passing through the system?

How big is this panel? Removing the glass sounds easy enough,unless it is very large, with the difficulty of moving it being the main hazard. Could you arrange a temporary resting place on the roof, rather than moving it too far? I use a couple of suction handle to move large panels - mine are cheapo suction handles for shower and bathroom use. If the worst happens and you break the glass, would it be very expensive or difficult to replace it (in Orkney!)

I wish you luck with whatever you decide. My days of climbing around on my roof are behind me!

OK, after testing to see how well adding images works - Not very well, seems you need to add one at a time with a post edit each time....

Anyway....

Well, its down. And yes, still in one piece, as am I after something that resembled a mountain rescue operation. I knew that harness and working at height rescue equipment would come in handy somewhere!

So went for the cutting the holes in the flashing to gain access to the top pipe and the brackets, built a wooden carriage for it to sit in, and run down over the Veluxes on 2 wooden rails that kept the panel and its carrier higher than the Veluxes. Then another couple of battens to skid it down from the eave to the ground, where it now lies, pending investigation.

But first, some pics for any glazing gurus who may be able to suggest how to get the glass out. I had hoped that maybe the glass and the frame would detach from the box body of the panel, but alas no. The aluminium extrusions are such that that is all 1 piece. First pic below.

So the next hope was to pop out the rubber bead that appears to be holding the glass in, and lift the glass out and set to the side, 2nd and 3rd image below. However, this is easier said than done. I am gathering the rubber bead has a barb on it that is pressed into the aluminium extrusion. Trying to pull this back out is having a tendency to want to split the rubber, even with the most careful of prising with plastic vehicle trim removal tools. As the corners appear to be bonded / welded to form a single square bead, my hope was to be able to re-use this, rather than try and source some generic bead on a roll that needs corner finishing. So the appeal is to see if anyone has any useful tips on extracting this rubber intact. Have tried every which way but loose - start on a corner, start on a straight - The issue seems that the barb has such a tight grip, the rubber is minded to separate first.

Anyway, onto the remaining pictures. Some show these rogue clips - A bit fuxxy due to the optical characteristics of the glass, and the displaced aluminium plate withthe selective coating. That plate is lying in the bottom and to the right of the box, leaving a gap at the top and the left.

My other possible option is to come in through the back of the box. There is a large rectangular hole in the back, behind which the rear insualtion panel is visible. But that insulation panel is bigger than the hole, not flexible, and could only come out in bits if attempted. My guess is that the panel is built up from the back to the front, with this rubbed bead being tapped in last. Anyway, if it needs to come out in bits and be replaced with new, that;s what it will have to be. Not sure how easy it will be, if at all possible to get component parts, i.e. rubber beads or aluminium plates from Grant - suspect they will just want to sell a whole new panel. Till then, any suggestions on bead removal much appreciated!

OK, so the conclusion drawn after an invasive post-mortem is that the Selective Coating on the aluminium plate has indeed degraded, and this is the cause of the loss of performance. This issue mainly seems to be discussed online academically, rather than in a more practical real world. An example here....

https://www.mdpi.com/2076-3417/14/18/8438

The mottling and speckles on the selective coating viewed through the glass do indeed seem to be oxidisation. The speckles where the coating has flaked off the aluminium sheet revealing the silver bare metal beneath. The mottling I presume where its in the process of transitioning to something that can flake off.

Very disappointing on a solar panel that is only 10 years old, and when fitted, I thought was for life. Looks like my days of clambering roofs are not over yet, and this will become a 10 year ritual.

Given the constrution of the panel, its is not intended to be disassembled, so cleaning the existing coating off and re-applying it is probably not practical. So looks like its a whole new panel. Will make some enquiries on availability tomorrow.

I could try and contact Grant, but when I tried before, to enquire on possible causes, all they could repeat was "blockage, blockage, blockage" Warranty terms will also likely be trotted out, viz, 5 years extended to 10 if you use a Grant Accredited installer.

Aside from that, it becomes debatable if the product has failed. It still produces hot (warm....) water. Just not as efficiently as it did. Unless deterioration rather than failure is specified in warranty, then no joy there, and it is of course easy to state that "everything deteriorates over time" so it should be expected - Warranty, merchantable goods or otherwise.

And, notwithstanding all that, I have a Grant shaped "docking station" up on the roof, that can only really be filled by an identical panel. Sure, I could go another manufacturer route, but will give myself a reasonable task on re-jigging the roof covering to marry up to the revised panel. If this is a generic coating issue rather than a Grant issue, is that worth it I ask myself?

With hindsight, maybe I should have gone in roof rather than on roof. And with further hindsight, ground mounted rather than on roof. Then all of this would have been so much easier to deal with! And even more hindsight (great stuff this hindsight....) a manufacturer where you can flip open the panel in situ like a cold frame and swap out the flat plate, coating and all, as easy as changing a cassette on a 1970's recorder!

A salutary tale. This kind of story was part of what influenced me to go fully PV, and to make sure both the panels and the microinverters had 25 year warranties!

Posted By: Bryan_Rendall

Posted By: Peter_in_HungaryHave you confirmed with the manufacturer that the panels produced today are identical with those of 10 years ago and so a new panel will be a drop-in replacement ?

Yes, dimensionally they are the same, but wholesaler is presently following it up with them to get availability and price of another collector.

Don't forget to ask about the connection points location and type as well

Posted By: Bryan_RendallSelective Coating degradation is caused, among other things, by exposure to sun..........

Err.........this is a solar panel is it not ???

Selective absorption thermal coatings seems to be an active research area. And in typical bigger 'solar furnace' type applications the sunlight tends to be stronger and the temperature is much higher than in domestic systems in Scotland. So it might be a 'value engineering' choice and it might be worth looking for something better.

Posted By: Bryan_RendallThe trouble is, asking suppliers / manufacturers how good their selective coating is on a scale of 1 to 10, and trying to pick out the truth when they all say 10!

Yes, I think you'd have to ask them what their coating is, and perform the value judgment yourself

Yes, I imagine the lack of oxygen will be a big help in preventing oxidisation!

Progress today so far....

Have sealed up the panel as far as post manufacture allows - Taped the seams between the back of the insulation panel and the aluminium frame. Then applied "Heat Mate" high temperature (275 degrees C) silicon sealant to all possible breathing paths. Then, drilled a hole and fitted a 3/8 stud coupling with a push fit female connection for plastic tubing. This tubing will then be the breather and will be run into the loft, with the decision on how it breathes to worry about later. Either;

• Free breathing from the loft - At least its not direct outside air, a mile and a bit from the sea.


• Free breathing through a silica gel dessicant pot, in the loft. As above, but only drawing in dry air.


• Very very slightly positively pressurised with a low voltage fan, just to make sure nothing gets into the panel from outside. This is possibly overkill....

Anyway, panel is now back up on the roof, but not fully in position. To finish off tomorrow. Supposed to be a nice day here tomorrow till about 3, so the proof will be in the pudding when the sun comes round onto it..... Early start just to make sure I have it ready for then!

You appear to know me to well Peter! With the effort that has gone into this replacement, I will need to do something to preserve it. Not going to be minded to do this again in 2035! But it has been worth it. Its baking hot here again today, new panel running with just water at the mo. Domestic Hot Water is already up to 59 degrees since the panel came on line at about Midday. Pump running continuously on speed control, which the old panel just wouldn't do. After a week of cooking hot sun, that one only managed to eventually get the water up to 50 something. Today, after a few hours, that's been exceeded.


So definitely been the panel and I think definitely the selective coating. But note to self - Never, ever put a solar thermal panel up near the ridge on a Weslh slate roof, above a bank of 6 Veluves. Unless I in the future acquire a 12 metre loadall, and a means of driving it around to the front of the house without tearing up all that is planted there!