Help with Supercap charger idea

I am using an INA260 board to monitor the voltage of a Supercap (2.7V 50F) in a weather station to make sure it does not die. Normally the WS has it's own charging circuit for the Supercap but it is bypassed when it is on AC power, which it is...but I need to make sure if the power fails the cap is always ready to go and take over.

I will switch the circuit using a relay from the INA260 monitor to a charger when it reaches about 1.7V (which it will eventually since the INA260 depletes a little charge over time)..but I need a way to recharge this Supercap at a controlled rate (say, 500mAh until it reaches 2.7V), and am kinda at a loss as to what to use to do that -- I was thinking maybe a LM317T...

Does anyone have any suggestions what would work for that application?

Doesn't the charging circuit use AC being on to charge the cap? Otherwise, how would it get charged up?

I edited the post to reflect that - it bypasses the cap and does not charge when on AC power. It gets charged normally by a solar panel, but that is also bypassed when on AC power, for obvious reasons. I basically have to work around it here to keep it charged when on AC power since the INA260 depletes it from monitoring over time.

Do you have a way to disable the AC cutoff?

Unfortuanently, no. There are no schematics available to it either.

How long do you expect to lose power? Maybe just add a UPS to the system so it always has AC in. You can get one for $60.

Well, two things. One it's mounted on an exterior pole in a very small enclosure, and two it would be overkill. The 50F cap is meant to run the station for about 4 days by itself. So, I need to do it the way I originally intended.

Is this already part of the existing setup or part of your addition? If your addition why such a complex chip when all you have to do is measure the cap voltage?

Also consider the loss from the capacitor internal resistance. The internal resistance could be between 2 an 10 µA.

Is it possible to run a resistor and diode from the AC Source to the Super cap? You would need just enough to keep the super cap "topped off". Be sure the AC source is not so high as to overvoltage the super cap.

JohnRob,
It is part of the monitoring system I built. The reason I went with it is I noticed using a voltage divider board (type typically available for the Arduinos) that it was leaking current somehow and constantly backfeeding charge to the cap slowly. While that may seem good, it's not because it's constant and uncontrolled when I'm trying to just measure it, kinda like what you are describing in you last paragraph.

Your solution of of a resistor/diode would be doable if it can be switched on/off, perhaps with a relay. Do you have an example circuit?

What voltage does the mains supply provide? Is it stable?
What voltage are you running on the supercap?

I think I may have omitted some details or misspoke, even though the circuit is originally powered via AC, it is powered via a AC to DC converter, so for all intensive purposes the entire circuit is DC including the cap, and the AC input is actually DC. I apologize.

The simplest. 2.7V DC power supply + 2.7 Ohm , you charge capacitor through this resistor, max current charging 1A will drop as capacitor is charged, if charging is to slow reduce resistor value.
or
https://theorycircuit.com/constant-current-dc-power-supply-circuit/

No worries, I pretty much assumed you meant something like an AC adapter.

Assuming the SuperCap is operating at 3.3 Volts:

I would add the below Ckt:

R1 may have to change depending on the voltage from the AC/DC converter.

It’s 2.7v and yes the DC supply is 5v.
Any changes?

Yes change R2 to 8.66k (a std 1% value). I've biased all the values to be sure the supercap is not over charged. So if you find when mains powered the supercap has a lower voltage than you would like decrease R2 slightly, but never to below 5K.

JohnRob and 79galinakorczak,
Thank you so much for your help! I will put this together and let you know.

These are 1/4W resistors, correct?
Also, 1N751A are no longer available. Digikey recommends this for a replacement - https://www.digikey.com/en/products/detail/on-semiconductor/1N5231B/977572 - is this acceptable?

I was not following this discussion closely but I think this will work poorly: when the main power fails R3 will discharge the supercap quickly. 10 kOhm and 50 F gives time constant of only 50.000 s ~ 14 hours. Much less than the intended 4 days of operation.

1 Like

That would make sense if it remained in the circuit with the Supercap. The Arduino will control this circuit being switched in for charging when it is detected below 1.7V. Once it reaches 2.5V, the relay will cut out (it will switch back and forth to determine if it is charged enough, and with a power failure it will be out of circuit) so it is no in the circuit, the only things remaining will be the WS and the INA260, which will be dead anyways as there will be no power. It looks like it should work as intended.

3.3volts
wrong. you need a 5.5 volt super cap and you need the cornell dubiler super capacitor. other wise all the girls will make fun of you for buying some jamico junk. it even sounds prestigeous. it does not come in a 3.3 volt flavor. so you have to use a 3.3 voltage regulator. your drone isnt going to spontaneously combust on its own. you need the cornell dubiler super capacitor

I don't understand why do you have Arduino here? It has since other tasks?

I would simply use a voltage regulator with output set to 2.7 V to keep the supercap charged all the time (and make sure it do not discharge the cap in case of a power fail).

In your setup what if the power fails when the cap has 1.8 V?