I have an NB-IOT Transmitter which sleeps most of the time, but one a day wakes up and transmits a file over FTP to my web server. My device runs on 3.7V and needs at least 1A for very short bursts during tranmissions.

I wish to use the following power supply:

  1. 5.5V, 200mA Solar Panel
  2. 2.7V 100F - 500F Super Capacitor, which will be charged via the above solar panel
  3. 2.7V to 3.7V DC-DC Up Converter, which will be connected to the Super Capacitor
  4. A non-Rechargable 3.7V Battery

I wish my IoT device to use the power from the Super Capacitor, unless its depleted, in which case it will use power from the non-rechargable battery.

The building blocks seems to be

  1. Current Limiting Super Cap Charger, between solar panel and the super capacitor
  2. DC-DC convertor to take the 2.7V from the Super Cap to the required 3.7V
  3. Some kind of switch to decide if power comes from 6) or from the non-rechargable battery.

The device is a remote monitoring sensor.

I have searched everywhere to find a circuit that comes close to what I want to do, but can't find it, so am hoping to get some pointers here. How can I do what I want to?


There are some problematic elements to your idea, but I think that it could be made to work.

Solar cells have a property called the Maximum Power Point, very simply this means that just connecting them to a load will typically have extremely poor results. You need to use something called Maximum Power Point Tracking to get good efficiency.

Supercapacitors typically have very poor leakage characteristics. This literally how much charge continually leaks out of them via internal processes, visualise a bucket with a hole in the bottom. You will need to choose both the topology and the supercapacitor very carefully to ensure you don't waste your captured energy in leakage and efficiency losses.

Supercapcitor charging ICs are readily available (e.g. LTC3625). The MPPT technique is shown in this App Note:

MPPT Technique for current-programmable charging ICs https://www.analog.com/en/design-center/reference-designs/circuit-collections/ltc3625-lt1784-solar-powered-super-capacitor-charger-with-mppt.html#cc-overview

For your power source selection into the load I would look at a 'dual ideal diode' configuration. This will allow you to 'OR' your two supplies together without allowing one to drive the other. This is usually the highest-efficiency way to accomplish this type of operation. (e.g. LTC4415). You would have the 'primary' supply regulating at a couple of hundred millivolts higher than the 'backup' supply (but still within the acceptable range for the load), this ensures the primary feeds the load whenever possible.

You will need to look very carefully at your power-budget. It will take a very long time to charge your supercapacitor, your circuit will be taking current to operate its various components, and the supercapacitor will be leaking charge. You will need to balance the charge time and total energy per cycle that the panel can support with the energy being constantly lost, and also the large burst of energy that will be consumed at transmit time for your IOT device.

  • \$\begingroup\$ Thank you for the response, i will follow up researching the LTC3625 and LTC4415. Any schematics i can study with the above chips ? \$\endgroup\$ – phantomrose1999 Jun 7 at 14:03
  • \$\begingroup\$ Thank you for the response, I looked up LTC3625 and it uses two caps to give a much higher output voltage than i need. I wish to have use a single 2.7V super caps, and thus need a 2.7V charger to suit. From the Super Cap to my device will be a DC-DC buck converter, taking in the 2.7v and upping it to 3.7. \$\endgroup\$ – phantomrose1999 Jun 7 at 14:12
  • 1
    \$\begingroup\$ I looked up the LTC4415, and its EXACTLY what i want to switch the supply rails. Much appreciated. 2.7 DC to 3.7DC Buck converters are avail, but i still need a 2.7V charger chip / circuit for the super cap please. \$\endgroup\$ – phantomrose1999 Jun 7 at 14:18
  • \$\begingroup\$ I installed LTSpice and downloaded the simulation for the LTC3625. With no load its showing a current drain of 5-6A, This cant be correct ? Please see screen shot. \$\endgroup\$ – phantomrose1999 Jun 10 at 7:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.