At first glance, adding solar power to your project might seem easy. Get a photovoltaic panel, point it towards the big ball of burning gas in the sky, and off you go. But in reality, there’s a bit more to it than that. Especially when you’re trying to do something on a small scale. Without a rooftop full of panels pumping out power, you’ve got to take what you can get.
If you’re looking to power small electronic devices such as sensors with a single solar panel, [Vadim Panov] has put together a very concise write-up and video on building a low-cost solar harvester. It combines a relatively small photovoltaic panel, a charging circuit, and a battery for energy storage into a easily mountable package. He’s provided all the details necessary to create your own version, all you have to do now is come up with the application for it.
As far as the electronics go, this project is about as straightforward as it gets. The three watt panel is connected up to a simplistic charging circuit, which in turn feeds into a single 18650 cell. You might be wondering why a charge controller is even necessary in such a simple set up. One problem is that the output voltage of the panel is higher than that of the battery. You also need a blocking diode that will prevent the battery from discharging into the cell during the night or in cloudy conditions.
While the electronics might seem elementary to some readers, we think the 3D printed case alone is worth taking a look at. Not only has [Vadim] come up with a design that perfectly encloses the fragile solar panel and associated electronics, but in the video after the break, he also explains how the entire thing can be made waterproof with an epoxy coating. As 3D prints can have a tendency to be porous, this technique is definitely something you should file away mentally if you’ve been thinking of deploying a printed enclosure outdoors.
Whether you’re looking to power environmental sensors for as near a century as is technically possible or a portable OpenWRT router for mobile anonymity, these small solar panels hold a lot of promise if you know how to work around their limitations.
When I saw the photo appear in my twitter feed I wrongly assumed this was a ‘smart ‘solar panel that contained actuators that tracked the sun. I know builds already exist for this, but all the same, slightly disappointed. The world needs these. I wince whenever I see fields of dumb solar panels, its not like they can’t work out where to get the power from.
Nice idea for protecting with epoxy though, and even nicer that yer’ man shows us how not to do it first. I like that.
Tracking collectors are great and can add to the efficiency ‘but’ also add a lot more mechanical complexity to a very simple mounting system. My rack mounted roof collectors have worked without fail for over 30 years. If I had designed and installed a tracking system, I’m pretty sure that I would have to do a lot more complicated maintenance.
This is the basic premise of being an engineer, do a cost-benefit analysis of fixed panels vs tracking systems.
It’s absolutely undeniable that a tracking panel will collect more energy than a fixed panel. The question is, “Is it enough more to offset the problems incurred?” (spoiler alert: it’s not even close, and getting worse as panel efficiency goes up and panel costs go down)
The detail I found most interesting in the last few years is that more panels are being installed at less than optimum angles. The cost-benefit has gotten so lopsided that extra panels mounted flat to whatever roof angle you have is better than the cost of a fixed mount that points to a more optimum angle.
More than that, even large field installs with adjustable mounts are often not optimized for “maximum energy over the course of the day” instead being optimized for “maximum energy at time of highest demand.” It’s a subtle nuance, but an important one.
I wondered if it would be possible to use a really really long length of material with a high thermal expansion coefficient to mechanically adjust panels based on temperature, you could add a thermal delay to expansion (and contraction) using a layer of insulation. But then you look at the actual numbers for current materials with a typical expansion ratio from 1 to 317 x 10^-6 per kelvin and think yea, there is no point at all even thinking about the idea. There is so much more bang for your buck in just adding a few extra panels.
Impractical, yes, interesting concept though.
There are other ways to get more significant motion from thermal expansion. Bimetallic strips flex rather than just a linear displacement. (Think rotary actuator rather than linear)
The other one that comes to mind is a “Wax motor”
If you are interested in exploring this further, search for ‘fluid solar tracker’. (Methanol seems to be a popular fluid for a number of reasons.)
Author of build here. My thoughts went along the same track as DKE.
When I was a teen, I’d probably prefer to build a solar tracker just for the hell of it.
Now, though, when I finally decided to design and build one, I had to weigh pros agains cons.
Pros: (potentially) more energy collected. I wrote “potentially” because you have to make the gain from optimal angle’s additional power offset the loss needed to power the tracking motors themselves – and if you only break even at that you’re getting as much power as a static harvester.
Cons: More complex design, harder to repeat.
Eventually I settled for the static harvester so as to come up with a useful elementary building block, that can be easily reproduced by anyone and incorporated into their own projects.
That is the case in suburban Australia.
And that apathy to obtain maximum energy thru tilt and orientation exist because solar panels (but not costly, complex installations) are subsidized by local and federal government.
So, who cares, put another solar panel even if facing south.
It is different if or when every cent comes from your pocket.
You’ll want each and every electron generated.
Ask the greenies off the grid.
The flip side is that a lot of residential solar with net-metering is installed for maximum energy capture, resulting in peak power in mid-day, rather in the later afternoon when demand is highest.
How about every two months you go around and adjust the elevation angles?
Tracking the sun daily is the hard part. Elevation angle seems to have less negative effect.
I’d love to say I was playing devil’s advocate in my original post, but I’d be lying.
So it seems that “not worth the bother” is the general view, and that is news for me. Thank you.
Yeah, I’ve studied these builds as used for camera pan/tilt movement with all their various points of failure. I’ve always suspected there was a simpler, low-tech way of achieving a sun-tracker out there somewhere.
Something which takes the principle of crickets in a box, who crave (or fear) the sunlight and whose weight shift acts as the counter-balance under the solar panel.
I see used silica packets like that often put into projects like this, but I hate to tell you that silica packet isn’t going to be doing anything. They have a finite amount of moisture they can absorb and most are ‘full’ by the time we get our shoes or tchotchke off the slow boat from China. Also, they don’t have the capacity to significantly reduce the moisture content of the air, they are usually only good for maintaining the humidity level of a SEAELED container. Amazon and eBay have plenty of nice rechargeable packets and pellets if you are serious about moisture control.
Best to always assume that water will always get in (unless potted) and add one or more holes at the lowest point for the water to escape.
And add heating element to keep it dry.
Definitely, tracking system is not worth it!
Shucks…. Well, good to know there are alternatives for that! (Rice?)
I’m assuming there no water gets inside the enclosure when the lid is on, but rather I’m trying absorb moisture (if any) that condenses out of internal air pocket due to changing atmospheric conditions.
Will look into replacing that bag now, thanks for the tip!
Turns out you can regenerate the used gel as simply as heating it up in an oven! https://en.wikipedia.org/wiki/Silica_gel#Regeneration
Will try that out for starters.
The solar panel will “adapt” to the battery voltage , it is just less efficient. Direct connect and be done with it, unless your application is particularly space sensitive.
That’s a good way to overcharge the lithium ion battery. It would work fine with a NiCd.
Sure, easily dealt with using a zener or similar
Please use at least a TL431 and a transistor, a zener is awfully inaccurate.
Or just use a Zener that is 0,2-0,3V less than max cell voltage. Cell will hold a bit less charge, but will last longer…
Zener diodes can easily be +/- 5% even for the better ones. The zener voltage is highly current dependent. The TL431 on the other hand is a more precision device within it load range, you can trim the output voltage by changing the value of the voltage divider.
It isn’t even that less efficient. I measured/simulated this for a 5V/500mA panel. The regulator has to be >90% efficient (and run at the MPPT!) to break even, assuming no other losses and perfect power point tracking. So I decided to not bother about it.
Bill is right, though – you need overcharge limitation. Fortunately, this can be bought/built for a few cents.
This is a nice project! You might consider using the TP4056 though. It does the same as the CN3065, except
– it survives up to 8V at Vin (6.5V for the CN3065)
– it consumes only 55µA when not charging (650µa)
– Aliexpress is flooded with TP4056 boards which have builtin protection against overcharge, overdischarge and overcurrent (IC: DW01-G) and which are ridiculously cheap.
I do have several of those laying around. Will definitely try TP4056 with a solar panel, although there’s a reason I didn’t even get to them when I was thinking of IC to collect solar energy into lithium battery. Looking at TP4056 datasheet, I haven’t found any mentions of solar charging applications – and they’re different from just charging a battery from a wall-wart because trying to pull too much current from a solar panel will collapse the voltage on its terminals, rendering it useless.
So I guess since I had no way to tell if that’s going to be the case simply by looking at the datasheet, I thought I’d just buy something explicitly marked as “solar charger”. Lazy me :)
I’ll try them out and see how well TP4056 can milk a solar panel, thanks for the heads-up!
“trying to pull too much current from a solar panel will collapse the voltage on its terminals, rendering it useless” – you are right, that would be a problem especially if you add a boost converter to the solar panel. The TP4056 and the CN3056 are just fancy current-limiting linear regulators, so worst case the solar panel output voltage will drop to the output cell voltage, not lower (since they can’t actively pull more current out of the panel).
For more info about solar panels and charging batteries, you can watch this video by “the guy with the swiss accent” where he compares several switching and linear regulators: https://www.youtube.com/watch?v=ttyKZnVzic4
Thanks you! The video hits the spot. Actually, it looks like smth I should’ve watched before building my harvester ;-)
That ball of gas isn’t actually burning, its fusing.
Great project.
I’d probably say that this is a Solar Harvester inside a 3D printed enclosure, rather than a 3D printed Solar Harvester. Usually when I see “3D printed something” it means that the box was 3D printed but all of the essential components of the something were not.
Thank you!
I’d love to be able to print electronics for harvester too. In a perfect world, there’s no mass manufacturing, instead you buy design files from tech companies to print whatever you need at home: white goods, furniture etc. Aand, if you’re a hacker, you also manufacture your own designs.