Could I use mirrors to improve Winter output in United Kingdom?

[GodfreysGhost]

Hi, I am about to set up a 22 panel ground mounted solar system, using Canadian Solar 595w panels. I have bought one already to try out, and I have an Elejoy 800w solar panel multimeter/tester, which I am having great fun with, lifting up the panel (with help from my girlfriend, because it's huge, 2172mm x 1303mm x 35mm) and adjusting its angle to the sun, and taking down the readings from the tester. This is in June 2022, in the U.K., Gloucestershire - I have got (allegedly) 680w from the panel, on a clear sunny day, when completely perpendicular to the sun, at 12 noon - I don't know if my tester is incorrect or not, but it's incredibly useful for finding out what difference the angle between the panel and the sun makes. I got 200w from the panel when the panel was almost at right angles to the sun at 10am in the morning, even though only a small amount of the panel was then being presented to the sun.
I am trying to decide how to mount my panels on the ground - the easiest and cheapest way for me to begin with, is to use bricks or breeze blocks on the ground, on top of weed control fabric, at each corner of the panels, and also in the middle of the long edges of the panels. Obviously this is isn't going to give me the best output during the Winter months, so I was wondering if it would be possible to put a cheap, and light mirror (probably plastic sheet that has been given a mirror finish) upright behind the panels (on their North side) so that the sunlight coming from low in the sky during the Winter months will be increased, as the taller the mirror I have, the more effective sunlight I will get to hit the panel. It will be much easier for me to create an adjustable mounting for a lightweight plastic mirror, than for a 31kg solar panel.
I am NOT going to be using the mirrors in the Summer months, and probably not even Autumn or Spring, it's just for the worst months of output, November, December, January and February, or thereabouts, so there will be no risk of overheating the panels (I presume). I would completely remove the mirrors outside of those times.
I wondered if anybody else has done anything like this (I am specifically interested in increasing output in the Winter only), or if anybody knows of any research papers that have been done about this.
I tend to look at solar farms for guidance on what works best, because I presume they are out to maximise production at all times, to make as much money as possible, but the solar farm near where I live has its panels angled at something like 30 degrees from the horizontal, so their Winter output must be much lower than their Summer output. This then makes me think that mirrors are not financially viable, because solar farms don't use them, and nor is changing the angle of the panels, because most (if not all) of the solar farms I have seen photos of in the U.K. have fixed ground mounts. I'd be very interested to hear what everybody hear thinks.

[J.P.M.]

Your description of testing a single panel brings back lots of fond memories of my early days of experimenting with solar energy, including booster mirrors and portable testing rigs on hand trucks and hand held pyranometers. Thank you

On the idea of using booster mirrors: The concept looks attractive at first glance but for several reasons it's usually impractical for most residential installations with the practicality sort of inversely proportional to an array's size. That usually doesn't bode well for the efficacy of booster mirrors in residential PV applications.

One important consideration, particularly for PV applications is that irradiance enhancement is best used when the enhancement is equally distributed over an array especially if string inverters are used.

So, for example, that means to get equal irradiance on an array, the reflector field must usually be larger than the array so that light from the reflectors covers the entire array at say, for example,1000 hrs. solar time as it does at 1400 hrs. solar time, earlier or later depending on the application.

That also means that a reflector field must usually be close to the array, probably attached to it. The smaller the array, the greater the ratio of reflector area to array area required. .
The structure to do any reflector scheme will most likely need to be about as robust as the array supports and so probably about the same price per square area of reflector as the price for the panel supports.

Also, the any increase in daily, monthly or annual output that results from a reflector field is probably not as great as you may think.
- there are reflector losses that are probably something like 20% minimum as a practical matter.
- the reflected light is coming in at a more oblique angle to the panels which reduces POA irradiance to less than that coming directly from the sky and incident on the array.
- because the angle of the reflected light increases the reflectivity of the panel's glazing (even with antireflective coating), the panel glazing will reflect more of the already angle reduced irradiance.
As a practical matter, and as a SWAG, the gain in array output may be something like, maybe 10 to perhaps 20% max.

By the way, vertical surfaces to the north of an array will require some pretty beefy supports to withstand the wind. Think more money.
Also, using reflective film may not be a good idea, primarily because it is not completely planar like a rigid surface reflector. That can create real hot spots on whatever surface a reflection hits. Believe it or not, that may present a fire hazard.
Somewhat comically but seriously, the architects of the former Playboy Club and Casino in Atlantic city, NJ used reflectors of plexiglass for the facade of their joint. That were not completely planar. The reflections started the boardwalk on fire.

Also, reflected irradiance will increase the temperature of the panel to some degree. That will decrease a panel's efficiency to some degree, decreasing the output as a % of total input.

Then, there is the idea that any stationary reflector field as some impracticalities such as increased maintenance and perhaps limits on array accessibility.

In the past, although not always, reflector mirrors were more commonly seen on solar thermal applications where irradiance variation over the entire panel field is not as critical as for PV applications where the increased panel temps were actually an advantage.

Welcome to the neighborhood.

[bcroe]

Keep experimenting, better than just throwing up a generic design. It
took me 4 years to figure out what my array ought to look like, and after
7 years there were still some weak points.

Not sure if you are building a net metering setup, or an off grid? Do
go ground mount, I have a list of 52 advantages over roof mount.

I must agree with J. P. M. about mirrors. Other factors can be snow,
and the percentage of days clouds will prevent direct sunlight.

We do see some tracking arrays here, mostly east to west. An advantage
of these is the ability to set for dumping snow in a storm, since commercial
arrays are too big for clearing by hand. I do keep my array cleared, and
there are ways to make this much easier as shown on the sticky. But the
E-W trackers become less effective at higher lattitudes. Bruce Roe

[GodfreysGhost]

Thank you both for your replies, J.P.M. - yours was particularly detailed about things I had never thought of. I am going to be ground mounting my panels, in landscape format, only one panel high, in two rows, and was thinking of having mirrors on the ground in front of them. I did think about the risks of a flexible mirror, it could very easily magnify the sun and make one part of a panel incredibly hot, and I also thought about the weather in the UK in the Winter months, when cloud cover is going to be the norm, and therefore a mirror wouldn't make any difference, I presume.
Then I thought about the cost of the mirrors (I will have 22 panels of 2172mm x 1303mm, so it would take a lot of mirrors to go with them), and it would be easier just to buy more panels.
Here are the results of the tests I've done with the Canadian Solar 595w panel since I started a few days ago:
Day 1: 29th June 2022: Broken clouds. Angled perpendicular to the sun at midday, my Elejoy 800w multimeter showed me 580w, (the panel is rated at 595w, so I was very pleased), and then it showed me 680w!
One hour later, with thicker broken clouds covering the sun, I got 200w angled perpendicular, and 150w when angled at 70-80 degrees from the horizontal (maybe 20 degrees higher than it was when angled at the perpedicular.)
Day 2: 30th June 2022: 10am, clear sky, with the panel at 70-80 degrees from the horizontal, but facing South, so the sun was at a very steep angle from the perpendicular, only just round the side of the panel, I got an incredible 200w - I wasn't expecting anything like that amount. When I turned the panel to face the sun directly to the perpendicular, still at 10am, I got 580w. This huge difference got me thinking about the possibilities of tracking the panels, but if you're going to track the sun from East to West, as well as changing the azimuth, you'll need a huge amount of space between panels, unless you mount many on a tracking platform, and I am using my back garden, and don't want to ruin the view for the neighbours (or myself!)
Day 3: 1st July 2022: 11am - panel flat on ground, on a West facing slope (downwards slope towards the West), overcast weather with no sun visibiel, starting to rain - 70w.
One hour later, less overcast, I got 250w - 350w, constantly changing - very impressive, seeing as the sun was still not visible, but the cloud cover was thinner.
7pm - full sun with no clouds, but the panel was at least half shaded by four feet high plants to the West, I still got 95w! I was very impressed with that, considering the angle of the sun was very low in the sky, and the very large amount of shading.
Day 3: 2nd July 2022: 10am - panel flat on ground, very overcast, just starting to rain lightly - 60w. Still impressive considering the weather conditions, and with 22 panels I would be getting over a kw coming in - if they could do that all day, that's still a load of useful kw/hrs coming in.

I'm not going to be doing any more testing because I've seen what I need to, the Elejoy 800w multimeter is a joy to use and so much easier than trying to use a conventional multimeter, though it did cost me about £100, I think it's worth it, as I will be able to use it in the future to test panels if any faults occur, etc.

I am going to be using off grid inverters (or single inverter, still haven't worked out which one to get yet), but will be taking electricity from the grid when needed, but not exporting TO the grid.)
bcroe - do you have your list of 52 advantages of ground mount? I'd love to see them.

I am just working on some diagrams of how I am intending to mount the panels - I will be using metal ground screws and recycled composite plastic beams, that look like wood - both of these should last for decades without any maintenance needed. I am going to make the frame so that the panels' angle is easily adjustable, just by lifting up the front of one panel at a time, moving it backwards or forwards a few inches along the horizontal beam, and setting it down behind a short length of wood that will be screwd in, across the beam. It will probably have four or five different angles.

[J.P.M.]

For anything I may have provided you are most welcome.

First I'd suggest you peruse the first 2 chapters of "Solar engineering of Thermal Processes" by Duffie & Beckman, which I simply refer to as "D & B". It'll provide a lot of background and make your journey easier. It's a free on line PDF download. Lots of very good background on the solar resource and its availability and some ideas on how to gather it that you may find useful or at least interesting food for thought. A read will probably or at least keep you away from a blind alley/dead end or two. It'll also answer a lot of your questions.

Even though the book was written for solar thermal, there is also a chapter later in the book on PV that's probably the best, most concise and to the point primer on PV I've seen.

Reflectors, be they specular (planar mirrors) or diffuse (as in painted white surfaces for example) will provide some irradiance augmentation under cloudy skies, but since irradiance under various cloud regimes is usually ~ 10 % or so of that under clear skies so don't expect much. There is one thing in D & B. Because of the nature of diffuse sunlight, reflectors of any type reflect a smaller percentage of their incident radiation than they do for beam radiation. However, they do reflect about, maybe half the diffuse vs. maybe 80+% or so of the beam radiation. But note, that does not mean an 80% increase in array output. As a practical matter, expect maybe 10 % increase in P.O.A. (Plane of Array) irradiance and chalk of the wild estimates and peddle hype to simple B.S. and a maybe 5 - 10 % increase in annual output per m^2 of array surface.

Your testing sounds (reads) familiar to me. But be careful. Curiosity can be addictive. I build my first thermal collector and mounted it on a hand truck in such a way that any azimuth or tilt was possible and so (manual) tracking was relatively simple. I also had a handheld pyranometer about the size of a pack of cigarettes that was a great help. I learned a lot and also learned I was largely ignorant of a lot of stuff.

So, I bought an earlier ed. of the D & B referenced above after I'd pushed the rig all over the yard for a few months and the book greatly increased the efficiency of my discovery process. That generated more question. So, I had a career change. I returned to school for more formal education. Back in my cart pushing days, I was a sales rep. of industrial process equipment. Now I'm a retired engineer. In a basic way, solar energy changed my life. So, beware. Just sayin'.

I reinforce your ground mount choice. Life will be easier even if a bit more expensive. a couple of possible concerns: You'll lose some use of the real estate, and depending on where you live and municipal codes, etc, you may need to put a fence or make other modifications to/around such equipment.

I'd respectfully suggest you shy away from wood or plastic for supports. The material will not last, is not as dimensionally stable as metal and will need more maintenance. It is not a long term method. A lot of folks use tubing or pipe of the sort that chain link fence framing is made of. There are a lot of accompanying fittings that also make life easier if that route is taken. Also, if you have not done so already, see what your local building codes require, not only for structural but electrical and such things as site prep./considerations as well. If you plan on tilt adjustments, make them easy to use, but equally or more importantly make sure they are as strong or stronger than the fixed connections (welds/bolting/etc.) they will replace (for wind and other external loading considerations). Think safety and long term serviceability before ease of operation.

As for the number of tilt adjustments, fixed is OK. many folks get by with 3 with four annual adjustments. When I engineered equipment for a living I preferred to build in as many options as possible/practical, and if I was designing an array support structure that had adjustable tilt, I'd probably go with something that offered continuously variable tilt. But knowing what I think I might know, and as a practical matter, I'd probably not use it more than 4 times/yr. with 3 angle settings.

Read Chap. 2 & 3 of D & B for more ideas.

Regards,

J.P.M.

BTW, from your portrait, you are the spitting image of a cat I once lived with.
Do you puke hairballs and destroy furniture ?

[bcroe]

Your Elejoy Panel MPPT tool is certainly educational. I wonder how
it dissipates that power, maybe they just take a short sample every
so often? My permanent aluminum and concrete mounts, with ability
to tilt, cost more than the panels. So yes, add more panels instead
of mirrors.

The problem with off grid, is a lot of power will be available when you
can not use it, it will be lost. With net metering all energy is saved.

Clouds are a big performance killer here, so adding more panels
does help. But then they could be too much for your inverter plant
under good sun. I managed to keep the same inverter plant by
connecting some east facing and some west facing panels to the
same inverter, they can never be maxed at the same time. Instead,
the length of the serious generating day is lengthened, here to
about 10 sun hours. This is a less than optimum use of panel
generating capability, but under these clouds it makes far better
use of the inverter plant.

52 reasons, not all will apply to any particular ap. Bruce Roe

52 Reasons for GROUND MOUNTING Solar Panels
1. Significant Other won't complain about looks. 1 July 22
2. No roof leaks
3. No reroofing problems
4. No 3' setback rule
5. No rapid shutdown requirements
6. Flexibility in location, to avoid shadows and save trees.
7. Easier panel repairs
8. Elevation angle change or tracking an option
9. Choice of initial angle
10. Less risk of harm in construction by not lifting components to the roof
11. Serviceable safely in all seasons
12. Fire hazard from panels distanced
13. No roof orientation direction issue
14. Layout not complicated by things like vents, chimneys, etc.
15. Easier to landscape mount
16. NOT feeling like I have a giant tick on the roof
17. No hiring a structural engineer to analyze roof loading
18. No walking on panels to inspect
19. Far easier to wash off dirt
20. Far safer to wash off dirt
21. Far easier to check for hot (potentially failing) MC4s
22. Ease of spot checking performance & tracking down any issue
23. Much easier to work on at night.
24. Less likely to be disturbed by high wind.
25. No problems trying to perfectly align with an imperfect roof.
26. Generally cooler operation than roof mount.
27. Avoid need for a building permit in some areas.
28. Can avoid glare problems by placing far out of sight.
29. Will not encourage squirrels and birds to build nests on your roof.
30. Trying to set up a solar pathfinder on a roof kind of s**ks.
31. No contract arguments about roof penetration points
32. No problem with snow blocks sliding off panels and damaging lower
panels, roofs, other equipment, or personal, no snow brakes.
33. No losing most winter production because snow slides are blocked.
34. No uneven heating up of panels due to restricted ventilation.
35. There is no limitation using older, less efficient panels to build up a system.
36. No concern with aesthetics with mixed equipment
37. Simpler to get back in service after a damaging violent storm.
38. Ease of using thru bolts to retain panels, instead of slip prone clamps.
39. Snow removal is at least viable.
40. Array can be designed to greatly aid in snow removal.
41. No roofers not willing to take panels off roof, due to liability
42. Much easier to directly ground against lightning
43. Removal of shade on a remote array will not impact the house.
44. Much less size limitation
45. Nobody wants to "experiment" on their own house.
46. No wind noise or possible vibration and deterioration issues.
47. No daily expansion/contraction noises.
48. No roof concerns if array is to be moved to another address.
49. Clumsy installers cannot break your roof tile.
50. No increase in house insurance cost.
51. No wires on roof
52. No panel ice clumps sliding off and damaging gutters.

[GodfreysGhost]

I will definitely try to find that book online tonight, J.P.M. The recycled composite plastic slats look very strong, but I am going to buy a few lengths of it first, to make a test frame for the panel I've already got. I'll see how strong it feels and only if it feels good enough will I do the rest of the panels when they arrive. I have only seen ground mounts made out of metal or wood, never recycled composite plastic 'fake wood', so I think they will be okay. But I am definitely open to using metal throughout instead, whatever will last the longest.
bcroe - the Elejoy MMPT multimeter only measures the power for a split second - there are three voltage regulators (or something like that, I'm not sure what they are called) inside, you can see them if you open the door on the back of the multimeter - there is a video on Youtube explaining how it works. If the watts it measures are small, (say 100w) then it counts down from something like 5 seconds to 0, then it reads the watts again, if they are 600w, then it takes thirty seconds to countdown before the next reading, presumably to allow the regulators to cool down sufficiently. I have stopped taking any further readings because as J.P.M. says, it's addictive!
I have to say that the whole idea of solar panels seems like magic to me - to be able to just buy a piece of glass with some silicon behind it, lay it on the ground, and get a decent amount of electricity out of it, for decade upon decade, without having to do anything with the panel except keep it clean, is incredible.
bcroe, thanks for the 52 reasons, I agree with all of them! There's no way anybody would be putting these 595w panels up on a roof, they are huge. With the ground mounts I can check the MC4 connectors with an infra red camera, I can keep the panels clean in any weather, especially when it's snowing, and I have a huge garden, so it just means less grass for me to mow!

[GodfreysGhost]

I have received the recycled composite slats I bought off Ebay, they are far too flexible (they are 3m long and 1" by 2") to use for solar panel mounts, I will cut them into shorter lengths and use them for a compost box. I have decided to take J.P.M.'s advice, and use metal - I am going to use C and U galvanised purlins, I want something that never needs maintaining and that will outlast me. I am going to be copying the design used by the Sinclair Sky Rack 2.0, as shown here: https://practicalpreppers.com/produc...unts/?swcfpc=1

This video, by Engineer775, shows the installation of this system: https://www.youtube.com/watch?v=DGp6IHUJ44I

I am not going to bother with the adjustable handle, I will just use a brace of flat metal to set the angle (and also tighten the bolt between the vertical and diagonal C purlins). I will be cutting all the purlins to length myself, but once that's done and I have drilled the necessary holes, building the ground mount will be a breeze, because it's just tightening bolts on the most basic structure possible. I am now trying to decide between using ground screws and using holes filled with concrete: I am only going to have one panel high per row, with eight panels in two rows, and six panels in the final row, making 22 panels in total. The panels are going to be mounted in landscape format, the width of the panel is 1303mm, length 2172mm. I will be mounting the panels at 76 degrees from the horizontal, so the top side of the panels will be about 1.4m above the ground, to keep the bottom side at least 15cm above the grass - therefore the vertical post (C purlin) will only need to stick about 0.65m out of the ground. I am drawn towards ground screws because they are much quicker to install, the total weight per ground screw is going to be much less than what it can really hold, as the panels are 31kg each, and I will probably have two panels per upright support. If you look at the video above, my system will be half the height and the upright post will be half the height too, so the leverage on it will be comparatively low. Therefore I am hoping that ground screws will work okay - there won't be much wind load because of the lack of height, and the five feet high hedge that is to the left of the panels, about six feet away.

I've just found another video from Engineer775 that shows something much more like what I am planning: https://youtu.be/ffmuaTsFn5c?t=267

[J.P.M.]

I didn't see/hear any reference to any design calculations in that video by Engineer 775 or any methodology used to calc any loadings.

I also saw several things I'd question and not do if it was my design, like, for one example, adding some diagonal bracing to the structure to avoid damage to the panels by not relying on them to provide structure resistance to torsional flexure from external loadings as will happen to the design as built.

But, it ain't my design.

Ghost: Do you have any municipal or gov. permitting or design review/requirements to conform to ?

[SunEagle]

GodfreysGhost. Unless you provide some evidence that you will have structural engineered drawings for you installation I will shut down this thread and give you a very long vacation from the forum.

[J.P.M.]

GodfreysGhost. Unless you provide some evidence that you will have structural engineered drawings for you installation I will shut down this thread and give you a very long vacation from the forum.

SunEagle:

Wrong poster/thread maybe ?

See thread started by eric@psmnv.

Thank you,
J.P.M.

[SunEagle]

SunEagle:

Wrong poster/thread maybe ?

See thread started by eric@psmnv.

Thank you,
J.P.M.

You are correct. I posted on the wrong thread.

[bcroe]

Pre engineered parts can work. I question if the variable tilt would be
working after a few years in the weather. Steel bolts here rust up pretty
fast, removal means breaking them. The pivot points on just a bolt is
not a great idea. The parts may get more attached than you want. Will
those screws keep working for long? The tilt range is limited before
you must unbolt the screw and remount it. That might be more
important if you go for a vertical, snow resistant mount.

Another plan here is 6061 aluminum held together with 18-8 stainless
hardware. Not cheap, but it has done well and still can be serviced
without damage after 9 years. 6 tilt bearings on tilting array are regular
machine shaft greased ball bearings. Multiple hole braces must be
unbolted to change, but for strength they are at a much larger radius
from the pivot. With a balanced array little force is needed for the tilt
change. Bruce Roe

[GodfreysGhost]

SunEagle, I was very worried there for a minute, I had no idea what I had done wrong!
I have been thinking long and hard about whether I need to be able to adjust the angle of the panels, and I think that for the moment, I will build a static ground mount. I have made a couple of diagrams that show what I am thinking of: I am going to use a ground screw to take the main weight of the panel, as the angle will be quite steep, to take advantage of the Winter sun more than the Summer sun, because I have 13Kw of panels and won't need the full output in the Summer. I will cut off one of the upright sides of the ground screw (the ones I am using come with a 'U' shaped top, which you can put a beam of wood, etc. into, and then put screws through the top part of the ground screw, into the wood, etc.) and behind this I will have a galvanised steel tube, which I will hammer into the ground, and about 2.4m-2.6m of it will be above ground. Then I will have a horizontal tube attached to the top of this vertical tube, onto which the top of the solar panel will lean. I can then lift up the solar panel's base so that it rests on two ground screws side by side, the same distance apart as the width of one solar panel, with two panels' edges sitting on each ground screw, and then I will use U-bolts on the horizontal bar to hold a flat metal plate on the top of the frames of the solar panels, and that's it. I will use diagonal poles every five or fix panels, from the top of the vertical pole down to the bottom of the next one, to stabilise the structure more.
I have tried to upload my diagram, which is in .PNG format, but the forum software keeps insisting that I provide a source URL for it, even though I'm using the 'Upload' tab. I browse to my image, then I press 'Upload it to the server', then when I press Okay it says "Image source URL is missing." Do any of you know what I'm doing wrong? I've had no problem uploading images on other forums.

[SunEagle]

SunEagle, I was very worried there for a minute, I had no idea what I had done wrong!
I have been thinking long and hard about whether I need to be able to adjust the angle of the panels, and I think that for the moment, I will build a static ground mount. I have made a couple of diagrams that show what I am thinking of: I am going to use a ground screw to take the main weight of the panel, as the angle will be quite steep, to take advantage of the Winter sun more than the Summer sun, because I have 13Kw of panels and won't need the full output in the Summer. I will cut off one of the upright sides of the ground screw (the ones I am using come with a 'U' shaped top, which you can put a beam of wood, etc. into, and then put screws through the top part of the ground screw, into the wood, etc.) and behind this I will have a galvanised steel tube, which I will hammer into the ground, and about 2.4m-2.6m of it will be above ground. Then I will have a horizontal tube attached to the top of this vertical tube, onto which the top of the solar panel will lean. I can then lift up the solar panel's base so that it rests on two ground screws side by side, the same distance apart as the width of one solar panel, with two panels' edges sitting on each ground screw, and then I will use U-bolts on the horizontal bar to hold a flat metal plate on the top of the frames of the solar panels, and that's it. I will use diagonal poles every five or fix panels, from the top of the vertical pole down to the bottom of the next one, to stabilise the structure more.
I have tried to upload my diagram, which is in .PNG format, but the forum software keeps insisting that I provide a source URL for it, even though I'm using the 'Upload' tab. I browse to my image, then I press 'Upload it to the server', then when I press Okay it says "Image source URL is missing." Do any of you know what I'm doing wrong? I've had no problem uploading images on other forums.

I believe the forum software limits the size of the picture that can be attached. I have had to resize a picture in the past and I am afraid you will also need to find a way to get the picture small enough to be accepted.