💬 Effective Solar Supercap Boost Charger for Small Solar Panel
-
-
@gohan said in 💬 Simple Solar Supercap Charger:
what does it actually do in detail?
It starts with a basic boost converter circuit, the same as before:
https://www.openhardware.io/view/279/Adjustable-Boost-Converter-with-Pass-ThroughIt then adds a voltage detector which controls whether or not the boost converter is enabled. When the 4.7uF capacitor is charged up to 0.83v, the voltage detector enables the boost converter, current flows from the capacitor in a spike, and the voltage is boosted. When the voltage detector detects that the voltage has fallen below 0.80v, it disables the boost converter so that the capacitor can charge up again. The cycle then repeats indefinitely. The circuit doesn't care how fast the capacitor charges, which is why mini and micro solar panels that deliver voltage but very little current can still use it. As long as they can eventually charge a 4.7uF capacitor to 0.83v, the circuit works.
If you were to instead to connect the mini solar panel directly to the boost converter, the current would be insufficient, and it would just spin it wheels, producing no boosted voltage. Especially at the beginning, it seems that most boost converters need a kick to get them going. i.e. the in-rush current can be much greater than the operating current.
-
So basically it keeps pulsing 2.7v on the output or do you have also a voltage limiter that will not overcharge the supercap?
@gohan said in 💬 Simple Solar Supercap Charger:
So basically it keeps pulsing 2.7v on the output or do you have also a voltage limiter that will not overcharge the supercap?
Yes, it's pulsing 2.7v. Whenever the 4.7uF capacitor charges up to 0.83v, it delivers a pulse. If you hook up a blue LED to it instead of a supercap, you'll see it flash with each pulse. Did you want to charge the supercap to less than 2.7v? As is, the supercap should top out at around 2.7v. The exact number will depend on the resistor tolerances for R1 and R2. If you want less than that as an additional safety margin (or more than that, if you're charging a supercap with a higher voltage rating), you can change the values of the two resistors R1 and R2 (see the MCP1640 datasheet for details). After all, it is an adjustable boost converter, which means you can set the output voltage by selecting the appropriate resistor values for R1 and R2. That's their purpose. That's what they're there for.
The exception to the above would be if the input voltage is actually greater than 2.7v (or whatever voltage you've programmed with R1 and R2). In that case, the output voltage would be about equal to the input voltage, but still pulsed. As before, the rate of pulsing will be governed by how fast or slow the 4.7uF capacitor charges to 0.83v. So, if that's a possibility that you want to prevent, you'll need additional circuitry to address that. Or, you could handle it in software by, for instance, possibly turning on the Rx of your radio and/or other loads to burn off extra power before the supercap reaches your target voltage limit. In my case, that would be easy to do, since my node periodically wakes to monitor and report the voltage on the supercap anyway.
-
Hello @NeverDie, I would like to give it a try but I have quite limited supply options. I see some MCPT1640T on AliExpress but no S-1009 and not any other voltage detectors, did you happen to see another reference during your searches that could replace it, even with less interesting specs ?
-
Hello @NeverDie, I would like to give it a try but I have quite limited supply options. I see some MCPT1640T on AliExpress but no S-1009 and not any other voltage detectors, did you happen to see another reference during your searches that could replace it, even with less interesting specs ?
@Nca78 said in 💬 Simple Solar Supercap Charger:
Hello @NeverDie, I would like to give it a try but I have quite limited supply options. I see some MCPT1640T on AliExpress but no S-1009 and not any other voltage detectors, did you happen to see another reference during your searches that could replace it, even with less interesting specs ?
Digikey carries a bunch of different voltage detectors, which is how I found that one. So, you might try looking for part numbers on Digkey, and then checking to see whether your supply sources carry any of them.
The other, perhaps more interesting, alternative would be to build your own voltage detector out of more basic parts. I haven't researched how to do that, but I would certainly like to detect voltages at less than 0.8v, and 0.8v was the lowest that any of the voltage detector chips on Digikey could detect.
Also, it's possible that a mosfet or transistor or something just naturally switches at a low voltage. That's probably the least expensive option, and something that you might find on Aliexpress. However, I wasn't sure how to search for that easily, so I haven't explored that. It doesn't have to switch at a precise voltage, or even a repeatable voltage: just a voltage that's something near what you want. In that respect, the voltage detector is overkill, because it's more precise than it needs to be.
In addition, I suspect that a switch/detector with a wider hysteresis would be more efficient, but I haven't explored that either. However, in that regard, you might get a similar efficiency benefit from simply using a higher capacitance value.
Hope that helps. Please do let me know what you come up with!
-
Hello @NeverDie, I would like to give it a try but I have quite limited supply options. I see some MCPT1640T on AliExpress but no S-1009 and not any other voltage detectors, did you happen to see another reference during your searches that could replace it, even with less interesting specs ?
Not sure, but perhaps some kind of Schmitt Trigger used with a part that naturally switches at low voltage could do the job.
Here's a good overview on Schmitt Triggers: https://www.youtube.com/watch?v=Ht48vv0rQYk
-
So basically it keeps pulsing 2.7v on the output or do you have also a voltage limiter that will not overcharge the supercap?
-
good to know, the problem is when using a bigger solar panel; protection circuits usually have a transistor that gets activated when voltage gets to 2.7v and start discharging cap with 4-10 Ohms resistor and that would be my goal. I am waiting for Adreas Spiess to review the supercaps protection boards on his channel and I'll see from that
-
good to know, the problem is when using a bigger solar panel; protection circuits usually have a transistor that gets activated when voltage gets to 2.7v and start discharging cap with 4-10 Ohms resistor and that would be my goal. I am waiting for Adreas Spiess to review the supercaps protection boards on his channel and I'll see from that
@gohan
If you're comfortable having the MCU manage the charge limit, you could simply use a PFET between the solar panel and the supercap. Have the PFET gate connected to ground through a pulldown resistor, so that the PFET initializes to "open" and thus default to charging mode. When you want to shut-off the supercap charging, drive the PFET gate pin high using one of your ATmega328P digital pins. I've done this previously, because I was attempting to automatically measure the open circuit voltage on the solar panel. I don't recall now whether it worked fine just like that, or whether I had to drive the PFET with an NFET, but however I did it, it seemed to work fine as far as disconnecting the solar panel and preventing it from charging anymore. -
@Nca78
Interestingly, in some scenarios it's possible to just do a kickstart and then switch over to non-pulsed mode. So far I've only done it manually, so I haven't yet worked out circuitry to do it or circuitry to decide when it's appropriate and when it isn't. Based on preliminary work, though, it looks as though the same basic technique may be applicable to buck converters also. In a way, kick starting buck converters should be easier to solve, because the starting voltages are (obviously) higher, so finding components which work at those voltages will be easy. -
Where did you get S-1009N081-I4T1U voltage detector please? Cannot find it on Farnell or Arrow.
-
@NeverDie It is not available on Digikey either currently. Any idea what I could use as a replacement please?
@alexsh1
It turns out Digikey does have it, but their own search engine couldn't find it. Very odd. Here's a link:
https://www.digikey.com/product-detail/en/sii-semiconductor-corporation/S-1009N08I-I4T1U/1662-1182-1-ND/6601322Actually, any of these would work in theory:
https://www.digikey.com/products/en/integrated-circuits-ics/pmic-supervisors/691?k=S-1009N&k=&pkeyword=S-1009N&pv846=33&mnonly=0&newproducts=0&ColumnSort=0&page=1&quantity=0&ptm=0&fid=0&pageSize=500
except that not all of the would fit the land pattern.Notably, though, I see they are showing a "new" product entry which wasn't there before:
https://www.digikey.com/product-detail/en/sii-semiconductor-corporation/S-1009N08I-M5T1U/1662-2290-1-ND/7228582
which is a good thing, because it is larger and would be easier to solder.Are you able to get that one? I could re-do the PCB and customize it for that one instead.
-
@alexsh1
It turns out Digikey does have it, but their own search engine couldn't find it. Very odd. Here's a link:
https://www.digikey.com/product-detail/en/sii-semiconductor-corporation/S-1009N08I-I4T1U/1662-1182-1-ND/6601322Actually, any of these would work in theory:
https://www.digikey.com/products/en/integrated-circuits-ics/pmic-supervisors/691?k=S-1009N&k=&pkeyword=S-1009N&pv846=33&mnonly=0&newproducts=0&ColumnSort=0&page=1&quantity=0&ptm=0&fid=0&pageSize=500
except that not all of the would fit the land pattern.Notably, though, I see they are showing a "new" product entry which wasn't there before:
https://www.digikey.com/product-detail/en/sii-semiconductor-corporation/S-1009N08I-M5T1U/1662-2290-1-ND/7228582
which is a good thing, because it is larger and would be easier to solder.Are you able to get that one? I could re-do the PCB and customize it for that one instead.
-
This remains the best boost charger for small solar panels that I know of. All the other low voltage boost chargers made by others that I've tried fail, probably because when a small solar panel has low voltage, it typically also has very low current. Solar panels don't produce much current indoors, and if it's outdoors you probably don't need a boost charger anyway. This is the only boost charger for small solar panels that I know of which works indoors under low light conditions.
It's conceivable that a solar charger based around either the ADP5090 chip or the SPV1050 chip might work as well or better, but I haven't tried either one. If anyone else reading this has tried them, please post and let us know how well they perform on small solar panels.
Enocean previously sold a solar charge module, the ECT-310, but I can't find it in stock anywhere:
https://www.enocean.com/en/products/enocean_modules/ect-310-perpetuum/The AEM10941 claims to work under indoor light on small panels. It won the Hackaday prize, so maybe it's one of the few that actually work under those conditions:
https://www.tindie.com/products/jaspersikken/solar-harvesting-into-li-ion-battery/ -
This remains the best boost charger for small solar panels that I know of. All the other low voltage boost chargers made by others that I've tried fail, probably because when a small solar panel has low voltage, it typically also has very low current. Solar panels don't produce much current indoors, and if it's outdoors you probably don't need a boost charger anyway. This is the only boost charger for small solar panels that I know of which works indoors under low light conditions.
It's conceivable that a solar charger based around either the ADP5090 chip or the SPV1050 chip might work as well or better, but I haven't tried either one. If anyone else reading this has tried them, please post and let us know how well they perform on small solar panels.
Enocean previously sold a solar charge module, the ECT-310, but I can't find it in stock anywhere:
https://www.enocean.com/en/products/enocean_modules/ect-310-perpetuum/The AEM10941 claims to work under indoor light on small panels. It won the Hackaday prize, so maybe it's one of the few that actually work under those conditions:
https://www.tindie.com/products/jaspersikken/solar-harvesting-into-li-ion-battery/@neverdie they work super well. I've been testing them for a few months now, very reliable.
One of the first supercap setups I made, "SolarRed". It's been running non-stop for 8ish months sitting indoors on a windowsill.
6 Month History SolarRed
Testing platform for different solar panel / super cap combos
