And mirrors generally don't like being outside in the rain (even though this is the middle east).
Might be easier to mount the solar PV's on an array that rotates and tilts through the day so that they are always at the optimum position.
"Easier" than using mirrors??? Have you ever tried making up a jig that will do this? I have and I can tell you it's not as simple as you might think! The solar panels are quite sensitive to alignment and only give of their best when placed *directly* in front of the light source. This, in the case of the sun, requires constant re-alignment in 3 axies plus a 'map' for the track of the sun for each day of the year (okay, you could get away with a 'weekly re-map' but it's sub-optimal). This will require quite a complicated assembage of electro-mechanical parts plus a micro-controller of some sort to do the brainy bit. Best of British luck in your endeavour!
From past comments, I suspect jon is based in NW england?
Plugging 0.23 kWp into the PVGIS calculator, the panels would generate a total of 22 kWh from start November to end of January so only enough to recharge the batteries less than once a week. If the continuous load is greater than 14 Watts, you're screwed, I suspect it might just power the hub alone, forget the computer/monitor.
It'll be about four times better in summer, about 92 kWh from start April to end June, so perhaps it could support a continuous 60 Watt load in summer?
It's called a solar tracker, they exist, in numbers, and can be moderately expensive (around 500). People take, like, a bunch of panels, make a super-panel of them, then mount the assembly on the solar tracker. The reason for doing that, is to amortize the tracker over as many panels as possible.
If you look on Aliexpress, they try and sell you just the transport, without any control electronics. bad monkeys.
As for the sun tracking, you can do that with a four quadrant sensor. Someone in my experimental physics was messing around with something like that.
What you don't want, is you don't want a sensor in this application, to be continuously honking on the two motors (that wastes all the power you're generating with the panels). Like, you can wake the thing every ten minutes, give a 100 millisecond squirt on each motor, and put the controller back to sleep and call it "good enough".
You can also do it with stepper motors, maths, and open loop. And only approximately trace an arc across the sky to track the sun. This has the advantage of not being confused by clouds. The controller needs to know the date, so it can pick a good track for the adjustments (every ten minutes).
Using a tracker, adds an extra hour in the morning and an extra hour at night, to collection. And is mainly to make you feel better :-)
But really, nobody wants to lay a fortune in panels on their house roof, and the house roof is all wrong for solar, and the panels aren't even remotely close to correct. So even if you made the mounts in the above picture, and you didn't run a tracker, you could still pick a suitable average position for the time of year and do a static pointing.
We have solar powered traffic radars here, where the panel at some part of the year, is not even remotely close to pointing in the right direction. And that's the kind of crap you want to avoid. Is the "dumb pointing error" thing.
Paul
There are a few here, and they have cost a hell of a lot more than that.
Bullshit. The array on the local ones is a massive great thing, at least 6x6 panels mounted on a f****ng great metal column. That is a massive great sail and it isnt even possible that it will always be end on to the wind in a major gale. That's why they are much more expensive.
Or flying around in a gale next to your house either.
Makes more sense to calculate where the sun is and drive it that way.
That's the best way because the position of the sun is completely predictable.
There's probably more to it than that or this wouldn't have been done:
I think your logic is sadly at fault, Whereas a quality engineer would do the sums first and then confirm the results with experiment, the inferior engineer tales lots of measurements and baffles brains with big words and bullshit.
I reckon that somebody has had a very successful sales drive in my area. Currently half a dozen properties within a hundred yards of me are scaffolded for solar PV installs.
Notably, several of them have the ridge aligned N-S.
I guess the power situation has really changed the payback calculations.
Chris
It did surprise me how little being quite a bit off-south affects the performance of solar.
It's quite a good paper, but I didn't follow why they didn't compare the realised performance with a series of theoretical models, and then drill down on significant variables. Maybe the next paper . . .
Your comment led me to read it. I must beg to differ. You have already noted the omission of the obvious comparison with calculated output. But the paper fails even to mention the issue of microclimate - whether Halifax is more likely to get cloud/rain at different times of day. All it really shows is that you get a bit more over a year from panels fang South.
Lots of people have done sums which show that's totally expected for Halifax's latitude where more is generated in the summer when the sun rises in the NE and sets NW.
play with
it'll let you see the expected annual (and varied monthly) output for a given panel capacity, in any location, with whatever alignment
I'm not sure what the three available insolation databases take into account?
I thought it did (3rd para of s.1) - indeed, that was one of the points of the paper in the first place?
Well, the paper I linked to used Huddersfield as the case study, so I'm not sure what you mean.
Useful to plan an allotment too :-)
There's a brief explanation here:
I meant that the paper doesn't answer the question of microclimate in Halifax.
Sorry, that was not a comment on the paper, it was in response to your surprise about the variation with azimuth there.
Got it, thanks.
Could that not be done by installing a photo-electric cell at the bottom of a tube, then moving tube and panel to seek maximum light level, within certain bounds.
Four quadrant sensor.
Some of these schemes, they get into a "seek loop", where the steering moves the panels back and forth, from 79 degrees to 81 degrees, when the correct value is 80 degrees. That might happen if you have stepper motors where the discrete step size is too large.
Designing *a* scheme, isn't the problem. It needs to be the *best* scheme :-)
An example of open-loop transport, is the clockwork mechanisms used on telescope mounts. You can see here, that a number of small companies think this is a good business to be in. The prices of these items seem unattractive for any sort of DIY experimenter application. With a good mount on your telescope, you can open the shutter on your film camera for 30 seconds, and the picture of the stars is not full of streaks. Of course if Musks fleet crosses your field of view, those will leave streaks. There is no guarantee of a "good time" these days when running an observatory.
Any tracker that actually works can beat no tracking. If you're homebrewing, a tracker that hunts can often be fixed by just reducing the motor power. If running ot from small differential solar panels, a power zener would do, preferably a rubber zener.
Not if it uses more power than it gains.
Bullshit.
Makes more sense to point it where the sun is calculated to be since the location of the sun is completely predictable.
That makes even less sense.
To follow myself up, another couple of scaffolds have just appeared on adjacent nearby properties. A stroll around the village reveals a surprising number of installations in progress.
I do hope that these customers have been given realistic financial projections, insofar as anybody has a clue what is happening these days. ;-)
My guess is that it is a knee-jerk response to rising energy costs, possibly without fully understanding that it is not a short-term solution.
Chris
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.