Hackaday Prize 2022: Solar Harvesting Is Better With Big Capacitors

The sun is a great source of energy, delivering in the realm of 1000 watts per square meter on a nice clear day. [Jasper Sikken] has developed many projects that take advantage of this power over the years, and has just completed his latest solar harvesting module for powering microcontroller projects.

The concept is simple. A small solar panel is used to charge up a lithium ion capacitor (LIC), which can then be used to power other projects. We first saw this project last year, when it was one of the winners of Hackaday’s 2021 Earth Day contest. Back then, it was only capable of dishing out 80 mA at 2.2V.

However, the latest version ups the ante considerably, delivering up to 400 mA at 3.3V. This opens up new possibilities, allowing the module to power projects using technologies like Bluetooth, WiFi and LTE that require more current to operate. It relies on a giant 250 F capacitor to store energy, and a AEM10941 solar energy harvesting chip to get the most energy possible out of a panel using Maximum Power Point Tracking (MPPT).

It’s a useful thing to have for projects that you’d like to run off the sun, and you can score one off Tindie if you don’t want to build your own. We’ve seen [Jasper] pull off other neat solar-powered projects before, too. Video after the break.

26 thoughts on “Hackaday Prize 2022: Solar Harvesting Is Better With Big Capacitors

  1. Interesting gadget, but I’m confused about the sweet spot use case.

    The LIC here seems to hold a little over 1000J, depending on max voltage. My reading of the datasheet suggests max voltage drops as temperature rises. Thats comparable to a 100maH 1S LiPo battery.

    The LiPo may be rated for ~1000 full discharge cycles, but a typical small its also ~10x bigger capacity wise. So if the discharge is the LIC full capacity then the LiPo would get ~10k cycles. Here the LiPo will likely die of old age well before it wears out. Assuming that the battery capacity at least runs the gadget through the night for either style.

    The LIC doesn’t seem to like heat, not sure its lifetime is that wonderful in a sometimes hot environment.

    1. I mean it’s great to investigate a nascent tech so you have plug’n’play knowledge when some breakthrough makes it a couple of orders of magnitude better, but at the moment, it’s not even demonstrating a compelling use case here vs 1970s NiCad capacities.

    2. ^ This.

      LiC offer no practical advantage to Li-Ion in this situation. Li-Ion stores at least 10x the energy (so for a given consumption, the number of cycles is the same as LiC). LiC is temperature limited to the same range as Li-Ion. It’s more expensive. Requires a specific electronic to charge them (since, unlike a capacitor, you can’t discharge them completely, you have to cut the discharge at 2.5V to avoid damaging them).

      The advantage of such capacitor is the exceptional charge/discharge current (compared to Li-Ion) and in the presented situation, it does not make any sense to use them since there’s no requirement for high current.

      At least other supercap could be used instead, but then the circuit is very expensive in that case (since for a capacitor, charging and discharging can be done with a asynchronous buck/boost converter).

      1. I agree. They are used in cell phone to draw up to 50A during HF pulsed transmission.

        But they won’t help solar panel performance ratio which is poor (and most important limitation of such solar powered device.

    3. placing the capacitor inside a thermally shielded enclosure might help with this issue but then again I see no real use of using a capacitor in place of a li-ion battery.

  2. These tiny panels are not very efficient when looking at manufacturing cost vs total lifetime energy. Probably better to put some full size panels on your roof, and charge your batteries with a regular charger.

  3. Wow, you can actually buy that e-peas AEM10941 Photovoltaic Energy Harvesting Manager IC. Mouser: 2,610 Can Ship Immediately, qty.-1 @ $7.38 each (ouch).[1]

    Check out the e-peas web site.[2] They have all sorts of energy-harvesting product examples like the ” Moovement Livestock Tracker” and the “Cartier Solar Watch” which took four years to develop. Quoting Mouser: “e-peas, based in Mont-Saint-Guibert, Belgium, is a fabless semiconductor company focused on low-power, high-performance circuits intended to make devices energy autonomous.”[3]

    * References:

    1. e-peas AEM10941 Photovoltaic Energy Harvesting Manager IC


    2. e-peas Mont-Saint-Guibert, Belgium


    3. Featured Products by e-peas @ Mouser


      1. I’m going to start looking into this today if I have time, I hope dyi at home pumped hydro is worth implementing… Going to be difficult to tell without actually doing it with real turbine and pump… After that I’m scared of how big the tank might need to be to store enough juice for overnight let alone a daytwothree

        1. You could do some maths first, what’s the weight capacity of your roof or other high place and how much energy does that let you store in kwh? At a guess if mine holds 1000kg and is 10m high then I could store… 27Wh. Or I could buy a battery pack for tens of pounds or dollars that stores more. I suspect hydro only makes sense if you go big, lake or stream big.

          1. so for the equivalent of a single Tesla powerwall you ‘d need to pump the 1000kg of water to 5000m altitude, or pump 500T over 10m and have two 500m3 tanks..doesn’t look very practical. also if you want to get similar 22kW peak power, you need to move 720m3/h over 10m
            just the pump and turbine would cost more than a battery
            a 2019 study pump as turbine applied to micro energy storage concludes 42% round trip efficiency and $0.6-1.3/kWh averaged over its lifetime

        2. It is certainly a valid approach, that can be astonishingly efficient, and hold full charge basically however long you like, in some cases even gain free energy thanks to rainfall – but at a domestic scale I’m not sure it can ever make sense – if you can build a giant water tower on your property perhaps but otherwise the volumes (or height) needed to meet even modest domestic power needs for any meaningful time make it problematic at that scale.

          As CampGareth says in his case the storage level is a pretty low, though I suspect they are under estimating the carrying capacity available in his building with a little engineering work – but that really depends on how your house is built and thus how much load spreading and reinforcing you can do.

          There are advantages to the tech, but also big disadvantages and from a maintenance cost to power throughput at such small scale I don’t think water makes any sense, Compressed Air perhaps, probably lower energy density (as density of pumped hydro depends on how damn high the storage container is, unless you count all the empty space between it and the turbine/pump tank) but its equally simple, lighter and you don’t have to deal with water and the problems it can pose so quickly corrosion and erosion wise – still won’t easily provide enough power at a domestic scales (and you have to make a choice between efficiency and peak energy output – even a tiny CAES can output really high burst of power but the efficiency will be awful) but IMO it scales down much better than pumped hydro in practical terms making it a more useful ‘backup battery’.

          However at such small scales I don’t think there is any getting away from the chemical battery as being the best choice – if you really want simple and reliable perhaps stick with Lead Acid, to get more energy density there are many Lithium chemistries to choose from (I’d pick on of the longer lifespan lower energy ones in a domestic setting, but that is definitely at least somewhat situational).

  4. in our head from our experiences with disc capacitors we believe capacitors have no messy chemistry, no mechanical parts to leak, break, catch on fire, etc. if you had a disc capacitor that has 1E12 times more capacity, it’d be something!!

    but the thing is, if you scale it up, even if you scaled up that same extremely simple technology, you’d find that in fact it does all of those things and at scale they start to become relevant.

    i remember how astonished i was to learn that LEDs *do* make heat and if you run them hard enough to make a lot of light then they make a lot of heat. the idealized solid state component in your head often falls apart when it meets real world scaling effects.

  5. Let me ask a stupid question…
    Is it possible to build, let’s say 3-4 capacitor, homade, 3m tall and 1m in diameter :) put them in a ground for cooling purposes maybe… Use them to collect all that energy from the sun and use batteries and inverters to make AC power…. Since capacitor that large, if it’s possible at all, will discharge insane amount of DC at once, would batteries will be able to catch all that energy or will simply burn everything with inverters and the house all together? :)

    1. That “insane amount” is called a short circuit. A stupid but otherwise capable EE could design something that allowed signicant damage. but then they wouldn’t get permission to market it. The cures are well known: fuses and circuitbreakers.
      As a backup, you can specify a second level of protection: that the copper hoses connecting the battery/capacitor be small enough to act as fuses -exploding spectacularly beforw the house catches fire.

    2. Why move it from a capacitor to a battery and then inverter for AC – that just adds losses in the transfer of energy from the capacitor to the battery, and such giant capacitor (if you could build them) should hold enough energy its not worth worrying about a little self discharge over longer periods, at least IMO so just go straight into the inverter.

      As to if an inverter can handle it, same as all electronics if its built too it do so it will handle as much as the caps could discharge, but you don’t HAVE to let the caps empty as quickly as they are able (really can’t see why you want to either), so you can keep cables to a more normal scale – say only as thick as your wrists, and use off the shelf industrial inverter – all safely under control.

      I do however doubt you can build a capacitor of that scale that actually stores as much energy as you would expect – something like a leydan jar can be scaled up, probably to almost any scale and work just fine (at least as well as they ever do, being interesting but really very primitive) but making a ‘normal’ multi layer type capacitor at such a giant scale would I think prove very challenging, especially as a DIY project.

  6. Except these are lithium ion capacitors, pretty much a hybrid of lipo and capacitor, I think they do still have a chemical element to them and a limited number of charge discharge cycles because of the lithium ion part of it.

  7. I think the operating spec for the LiC is -25 ~ 85℃. The spec sheet says -25 ~ 85℃ (-40 ~ 85℃ @ in Li/SOCL2 battery system). I believe to -40~85C spec requires a combo of the LiC and a separate Li/SOCL2 battery.

  8. Energy harvesting is an interesting subject. I wouldn’t go for a solar panel and cap as large as shown in this project, but just a small energy source (heat, solar, RF) and use it to power a circuit for a few seconds every few hours. When your micro is using only micro amps this can easily be achieved.

    In the video it was claimed 10 to 20 years life, but are there ways to prove this, except for expensive thermal cycling and duration tests? What are the DIY options that give reasonable assurance?

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.