A solar supercap powered Moteino (15Farad charged by BQ25504)

[WhiteHare]

For this exercise, I found that cutting back on the amount of deadwood debugging code that had accumulated over time helped a lot.  At the moment I'm able to do a full power-up from zero and then transmit at least a couple packets from just a single 220uf ceramic capacitor that was charged to 3 volts.  Presently I send two packets when it starts up: the first is really just a throw-away packet that pretty simply means "I'm alive", but, more importantly, allows the node to measure the cap's voltage under Tx load.  The second packet reports that voltage.  As currently configured, the voltage that gets reported is 2.31v.

The 220uf ceramic cap that I'm presently using is this one:  /www.digikey.com/product-detail/en/murata-electronics-north-america/GRM31CC80G227ME11L/490-13250-1-ND/5877458

Unfortunately, Murata doesn't make it easy to figure out what either the ESR or leakage of it is.  It's not in the datasheet, and I suspect neither is very good. 

However, it does at least provides a ballpark for what may be required.  I would certainly welcome suggestions for a better cap to try.  If nothing else, what type of 220uF capacitor is likely to have the lowest ESR and/or the lowest leakage?  Would it be a ceramic capacitor, or something else? 

[WhiteHare]

Good news.  I soldered two 47uF ceramic caps in parallel, and their combined 94uF (charged to 3v from 2AA cells) was enough to do the startup and send two packets.

This has me wondering now just how low I can go.     I'll try 47uF next.

[Edit: I just now tried a 47uF ceramic cap.  With my current un-optimized code, it's still enough to start-up and send one packet, but not two packets.   In practical terms, 47uF is probably fine, as the Moteino would be getting some extra current from the solar panel as well.  Or, it could wait for the cap to charge up enough before sending the second packet.  The tests I'm running here are more conservative and involve just the cap, charged only once, and with nothing else powering the Moteino.  That would then cover the case where the small cap is charged by an itty-bitty solar panel which can't provide any meaningful assist.]

[WhiteHare]

Hey my pleasure.  I'd like to see this project as small and cheap as possible since I plan on having a play at some point.  My hope is you can get the power requirements down low enough that a few F will get it through a few overcast days.  I found something else interesting that might be able to serve as the primary supercap.  http://www.cytech.com/products-ips Scroll down to the bottom and check out the MEC202.  The size is shockingly small and 2.2mAh used a few ms at a time should last a decently long time.  If that could work, then the whole project would shrink to not much larger than a mote.  Pair that with a very small solar panel and you're golden.

If you find a good source for reasonably priced tiny solar panels, please do let me know.  I'm drawing a blank on Aliexpress.  Digikey has some amusingly small ones, such as:
https://www.digikey.com/product-detail/en/ixys/KXOB22-01X8F/KXOB22-01X8F-ND/4840081
that would be fun to try.  With my existing configuration it should be able to powerup a Moteino and send packets--after charging the 94uF cap that is.

On the other hand, there are these, which are larger but still roughly Moteino size and the unit cost is much lower:
https://www.aliexpress.com/item/10Pcs-Lot-5V-30mA-53X30mm-Micro-Mini-Small-Power-Solar-Cells-Panel-For-DIY-Toy-3/32253274314.html?spm=2114.01010208.3.44.ohS6oq&ws_ab_test=searchweb0_0,searchweb201602_5_10152_10065_10151_10068_10130_10136_10137_10060_10138_10155_10062_10156_10154_10056_10055_10054_10059_10099_10103_10102_10096_10148_10147_10052_10053_10142_10107_10050_10051_10084_10083_10080_10082_10081_10178_10110_10111_10112_10113_10114_10181_10037_10032_10078_10079_5110011_10077_10073_10070_10123_10124,searchweb201603_9,ppcSwitch_4&btsid=db355a0d-73ee-4596-8073-becf4670eb25&algo_expid=8bca9805-03f3-4551-be53-5d8397a46035-8&algo_pvid=8bca9805-03f3-4551-be53-5d8397a46035

Or, perhaps these, which are maybe a bit smaller:
https://www.aliexpress.com/item/10pcs-pack-DIY-5-5V-23MA-Monocrystalline-silicon-solar-cells-solar-panels-38-30-3mm-solar/32644217312.html?spm=2114.01010208.3.200.mRsVqC&ws_ab_test=searchweb0_0,searchweb201602_5_10152_10065_10151_10068_10130_10136_10137_10060_10138_10155_10062_10156_10154_10056_10055_10054_10059_10099_10103_10102_10096_10148_10147_10052_10053_10142_10107_10050_10051_10084_10083_10080_10082_10081_10178_10110_10111_10112_10113_10114_10181_10037_10032_10078_10079_5110011_10077_10073_10070_10123_10124,searchweb201603_9,ppcSwitch_4&btsid=5fa03c0a-4084-4cf5-8442-03aec929454d&algo_expid=4a0ae799-9d57-4722-b7ff-cc772b1066b0-28&algo_pvid=4a0ae799-9d57-4722-b7ff-cc772b1066b0

[ChemE]

I like those little digikey ones.  I'd be curious if something that small could put a decent amount of charge in that little solid state battery that I posted.  If a 22mm x 7mm cell works, it certainly would not define the form factor.

[joelucid]

I like those little digikey ones.

Yeah - these are cool!

So one of these, a smd supercap and the voltage controller. We could make a little "power" board, round, solar in the middle. super cap on one side, voltage controller on the other - as CR2540 replacement. Just solder these over your motes, the radio underneath and you'd get a nice little tripple stack.

I wouldn't even bother with the 2 cap approach. Just pre-charge and go.

Everything else can be done in software. E.g. burning off excess voltage should that ever become a problem. Or drastically reducing ping frequency when voltage sags.

Joe

[WhiteHare]

In that case, it's possible you guys might also like this:
http://www.mouser.com/ProductDetail/IXYS-Integrated-Circuits/CPC1824N/?qs=sGAEpiMZZMs50KUSuyRkpmuv51Ai%252bapo7wdM%252bGmE0Ac%3d
It's an array of photo diodes meant to function as a solar cell, and it's even smaller.  It's from the same company, though it's now EOL.

[WhiteHare]

Yeah - these are cool!

So one of these, a smd supercap and the voltage controller. We could make a little "power" board, round, solar in the middle. super cap on one side, voltage controller on the other - as CR2540 replacement. Just solder these over your motes, the radio underneath and you'd get a nice little tripple stack.

I wouldn't even bother with the 2 cap approach. Just pre-charge and go.

Everything else can be done in software. E.g. burning off excess voltage should that ever become a problem. Or drastically reducing ping frequency when voltage sags.

Joe

I like these ideas!  You have a true knack for boiling things down to an elegant simplicity.   

[WhiteHare]

I'd be curious if something that small could put a decent amount of charge in that little solid state battery that I posted.  If a 22mm x 7mm cell works, it certainly would not define the form factor.

You mean the MEC202?  Is anyone selling it?  I certainly could be wrong, but from looking at the company's website, I gathered  the impression that it's probably vaporware. 

[WhiteHare]

Thanks for the links.  This project is pretty interesting to me.  I've found some SMD mountable supercaps and wonder if any of them could work as the small/boot capacitor.  I'm always looking to make these sorts of things smaller after all.  Something along these lines:

http://www.mouser.com/search/ProductDetail.aspx?R=0virtualkey0virtualkeyCPX3225A752D

After your post I procured this part and tried it out.  It works, but because of the high ESR there's not, by itself, much headroom left for transmissions.  After charging it to 2.6v (which is its rated maximum), then after powering up from it, my Moteino reports that the loaded voltage at the end of the first transmitted packet was 1.92v.  Just FYI.  That's for an RFM69W Moteino set to the lowest RFM69W transmit power setting.  Of course, it could have other uses where the high ESR would not get in the way so much, such as powering the Moteino during a long slumber.  Or maybe (?) you put a capacitor with a low ESR in parallel with it, so that it's charged and ready for transmission when the need arises.  But what parallel cap to use?  Or maybe you use it to charge up a low ESR cap just prior to its being needed so that you preserve most of the low leakage characteristics?

Maybe you also have ideas on how best to make use of it?  It certainly is small, which is always nice.   

[ChemE]

You mean the MEC202?  Is anyone selling it?  I certainly could be wrong, but from looking at the company's website, I gathered  the impression that it's probably vaporware.

Yeah, so far I've only found a single source: https://www.tindie.com/products/Bazinga/ultrathin-lithium-battery-mec202-22p/ and $23 shipped is not a very attractive price to me.  But 2.2mAh, 100,000 charge cycles, and a 10nA leakage current in a postage stamp sized package is pretty interesting to me.  I plan to contact the company and see if they have any distribution in the States.

[WhiteHare]

I like those little digikey ones.  I'd be curious if something that small could put a decent amount of charge in that little solid state battery that I posted.

That's a good question.  Anyone know if there's a way to predict what voltage ithe IXYS could charge it to?  It seems likely that it could charge it to at least 3.4v, because the DS lists that as a typical voltage for the IXYS.  Beyond that, though, the only other datapoint seems to be the open circuit voltage of 4.7v.  What, if anything, can one infer from the open circuit voltage?  I'm assuming that it represents an upper-bound on what it might charge the capacitor to.  Since the capacitor's rated voltage is 4.1v, is it fair to assume that, given enough time, the IXYS solar cell could eventually charge it to 4.1v?  Anyone know?

[Edit:  Assuming this chart is correct,

http://www.pveducation.org/pvcdrom/open-circuit-voltage

and assuming that the amount of self-discharge experienced by ChemE's flat capacitor is negligible, then I would think that the answer must be "yes."  ]

[WhiteHare]

So, I think a more complete answer to ChemE's question would be:  (1) if you were to hook up your voltmeter directly to the Solar Bit KXOB22-01X8F with nothing else connected to the Solar Bit, and (2) if you were to supply the Solar Bit with enough light that your voltmeter showed 4.70v (aka "the open circuit voltage"), then by definition the Solar Bit should, under those lighting conditions, be able to produce 3.8ma at 3.40 volts.

Sadly, I rarely ever see a datasheet which states how much LUX is required to get the rated open circuit voltage.  I guess you could calculate it from the stated 22% conversion efficiency given in the datasheet, but it would be a lot more convenient if they would just state it.

More importantly, what if the open circuit voltage is less than 4.70v?  What can you expect then? 

[WhiteHare]

On the other hand, if you already have the solar panel (or Solar Bit, or whatever) in your possession, then there's a conservative yet simple way to find out whether there's enough light to power your Moteino via the method of charging up a capacitor: plug a blue LED into the panel.  If the blue LED glows, then you've got enough light.  Although this can obviously produce a lot of false negatives (i.e. you may still have enough light even if the blue LED doesn't visibly  glow), I've done this test with a blue LED plugged into a mini 5.5v panel at my house, and during a sunny day, even with all the indoor lights off, it glows practically everywhere that has a window in it--even if the panel is pointed away from the window!

Give it a try.  I think you'll be pleasantly surprised.

[WhiteHare]

In case anyone is interested, I received the Solar Bit from digikey (above), and yes, from mere ambient indoor sunlight (i.e. not direct sunlight), it's adequate to charge a capacitor and then power a Moteino through the methods we have been discussing above.  I even attached a blue LED to it, and the blue LED glows.     QED. 

[ChemE]

Outstanding!  I love that a solar panel (bit) smaller than a Mote can power a Mote.  I agree with joelucid then, a little shield on top with this solar bit, one or more SMD supercaps, and perhaps a chip to run the panel at its maximum power point would make an amazing little add on for a Mote.  Slap the whole thing in a teeny clear case and you've got a supremely capable little package.