I'm going to power my arduino using solar power and a battery. Solar panel practicality aside I'd like to house the unit in a half cylinder (like a a PVC pipe cut in half) and have a string of mini 1V monocrystal cells in series and parallel running along the entire surface of the PVC pipe. This would mean only a certain number of cells are receiving adequate sunlight throughout the day. Can anyone provide insight whether this is at all practical, if there is a simple isolating circuit that ignores series that are going to drain the overall output of the array or should I just settle for the flat solar panel.
So far all I can find are new age hi tech solutions that cost more than I'm willing to spend. Small monocrystal cells are cheap so the fact that I'm paying for half of them to not be in use for half the day isn't an issue if it means a slight increase in output.
Generally, power consumption is a big issue with solar power. I highly recommend that you spend some time studying Nick Gammon's tutorials on power reduction and a solar powered Arduino.
Great deals on small solar panels can be found here. I have a few and they perform as advertised.
acitta:
I'm going to power my arduino using solar power and a battery. Solar panel practicality aside I'd like to house the unit in a half cylinder (like a a PVC pipe cut in half) and have a string of mini 1V monocrystal cells in
Difficult to understand that explanation.
Solar cells are normally flat panel devices.
Using focussed sun can dramatically increase their output but the reflector has to actively track the sun.
i have lived on solar for the past 10 yrs or so.. best thing i ever did was tip some flat panels to receive
early light and others the receive evening sunlight..
no trackers to screw up, and works very very well..
i think a 6 in pvc would be a little too small to do much good, but the idea is great..! i would use a more gentle bend..
Thanks for the links jremington. I've read both of them many times. I've been playing around with using 500F supercapacitors instead of batteries, but I think I'll go with a couple of 18650 batteries instead.
Say I have 6 x 1V solar panels in series and 3 lines of 6V series in parallel. A 6 x 3 array. Instead of mounting them on a flat board I mount them on a curved surface. If I mount these on a PVC cylinder cut in half I'll have the middle line of the facing upwards, the two other will be mounted either side and facing off at a 45 degree angle. In the morning the sun will hit one side directly, partially hit the middle line and the back line will be in the shade. By mid day the middle line will be in full sunlight and the 2 side lines will be getting partial light. By the evening the first line of solar panels will be in the shade with direct sunlight hitting the other half of my semi circular array.
I'm just curious if anyone has done any tests to see whether I'd get more output as a result of having 3 times per day when I'm getting direct sunlight rather than just once. backwoodsjack seems to suggest it could work well, but I guess I wont really know till I try.
I'm not too concerned about the cost of 6V vs 6 1V panels. My only rational with 1V was to have a long narrow line of cells. With some digging I can probably find narrow 6V cells premade. Maybe I'm completely underestimating my power needs but the 18650 will run my unit for over a week. I just need to trickle feed it enough to charge in that time so I wont need an excessively large panel.
What's the logic behind the step down converter on the solar panels?
acitta:
I'm not too concerned about the cost of 6V vs 6 1V panels. My only rational with 1V was to have a long narrow line of cells. With some digging I can probably find narrow 6V cells premade. Maybe I'm completely underestimating my power needs but the 18650 will run my unit for over a week. I just need to trickle feed it enough to charge in that time so I wont need an excessively large panel.
What's the logic behind the step down converter on the solar panels?
Google on MPPT convertor.
A simple boost converter is not quite the same but can maximise the efficiency of the solar collection, not economic for very small panels though, can be better value by buying extra panels.
Long narrow line of cells.
This sounds as though there is a constraint on your design, Can you elaborate .?
A simple boost converter is not quite the same but can maximise the efficiency of the solar collection, not economic for very small panels though, can be better value by buying extra panels.
i didnt encapsulate the above quote from boardburner2.. i dont see how to do that.. anyway..
a boost converter is far less efficient than any buck converter...
simple logic of copper loss in the boost inductor..
anyway.. it sounds like acitta has a very tiny power requirement for his 18650 project.. however i dis-agree
that efficiency drops off at lower power levels(for MPPT).. quite the opposite.. MPPT done corectly-- like
boardburner2 referred to with localised per panel conversion is really the answer to overall system effeciency..
Boardburner2:
I think mounting them in a 6 in cylinder would give too much shading.
Put a reflective strip on either side of the U so that when sun is directly on the panels of one side the reflector on that side is angled to reflect onto the panels on the other side and v-v. The strips should wide enough to work for a few hours.
Flat white reflects 50%, you don't need a mirror to beat that. Compare to "buy more cells".
If you have a flat panel aligned E-W, the sun will appear to travel at one elevation all day. That could be set by hand about once a week without being much out of alignment.
backwoodsjack:
A simple boost converter is not quite the same but can maximise the efficiency of the solar collection, not economic for very small panels though, can be better value by buying extra panels.
anyway.. it sounds like acitta has a very tiny power requirement for his 18650 project.. however i dis-agree
that efficiency drops off at lower power levels(for MPPT).. quite the opposite.. MPPT done corectly-- like
boardburner2 referred to with localised per panel conversion is really the answer to overall system effeciency..
Its not efficiency in operation i am arguing against, its CAPITAL COST versus return.
For small panels its better value to buy more panels than throw electronics at it for long term return.
Where the trade off is these days i cannot say , a couple of years back i worked out that MPPT was only worthwhile for several hundred watts.
With modern higher efficiency panels and cheaper converters that has probably shifted somewhat.
Cost per watt for 'proper' MPPT is difficult to work out without substantial data though.
Yeah I have relatively small power requirements. My arduino will sleep a lot.
The reason for the cylinder isn't really a good one. Originally I saw my unit housed in a 3D printed object the shape of a bike helmet with a flat solar panel on top (for no reason other than I thought it would look cool). Then I thought i had all this spare real estate around the sides, why not cover the whole thing in solar panels if theyre cheap. This could actually help me pick up the morning and the evening sun. A cylinder just seemed like a simplified example of this. Having solar panels arranged around an object vs putting them on top of a flat box means I actually get more solar panels per sq meter of ground which makes my unit smaller (yey). While a third of the panels will be inactive twice a day I'm guessing that is the equivalent of not having direct sunlight on the flat solar panel for two thirds of the day so there isn't an obvious power loss.
It sounds like the idea isn't a terrible one so I'll keep working through it! Thanks
Hi,
If you are going to have PV cells in series/parallel array and only some of them is going to be producing power at any time, then you will need to install bypass and blocking diodes into the array.
This is to stop cells producing current from having that current sinked back into a part of the array not producing as much current.
