Regulated Supply with Low Voltage, High Current Solar Cells

Hey there,

My project scope involves harvesting the energy from qty2 low-voltage high-current solar cells (I have taken two of these single cells from a 265W full-size panel). I am locked into these solar cells and cannot change them. These cells have the following specifications:

Voc = 630mV
Isc = 9A

Vmpp = 510mV
Impp = 8.42A

mpp = maximum power point

So, the above is my potential input to the power supply, and obviously this input would be varying dependent on irradiance. I plan on connecting the 2 cells in series, giving 1.2V @ 8.42A maximum potential input.

The output of the supply would go directly to a BQ24195 power management and battery charger IC that accepts an input range of 4-12V. Current can trickle in, but considering these solar cells are nearly 4W, having at least 2A available would be fine.

A couple ideas are wrangling in my head:

• Joule thief with regulated output
• Custom energy harvesting IC tailored to small solar panels (the problem with these ICs are that they have a maximum allowable input current that my solar cells cannot be directly connected to).
• DC-DC converter IC (I've looked at a couple, but there haven't found anything that suits this application well)

Obviously I would like to make this as simple as possible with basic off-the-shelf components. My biggest fear is the high-current (8.4A) in the solar panels will damage components and reduce the life of the project. I could limit this current with a resistor, but then the resistor would fail.

Are there any other circuit types that might work with this specific application?

Thanks for reading!

Your approach should be to have two DC-DC converters operating in parallel. One will be a low power converter which only needs the 0.6V input voltage to start up. The other will be a higher power converter which is bootstrapped by the output voltage. So on startup the low power converter will trickle charge the output until it's high enough to start the second converter, which will maintain the output while the first converter is disabled. For the low power converter, you could use a simple energy harvesting circuit like a joule thief. Linear tech also makes some converters with very low starting voltages which would meet your requirements, like the LTC3105 or LTC3108.

According to the spec sheet it needs a minimum input voltage of 3.9v to work.
And you have available only zero to 1.04v (MPPT) input.

This will first need a very small low power boost converter to generate enough voltage to run the control system of a more powerful main boost converter.

The problem will be that the main boost converter will need to draw very high intermittent peak currents to work, and you cannot draw any higher current than the solar cells can produce. So a very large low ESR input capacitor will be needed between the solar cells and main boost converter. All this will need to be located right at the solar cell end of the system.
Its just all too difficult, too expensive, inefficient, and impractical.

My advice would be to junk the solar cells you have, and start again with a larger number of much smaller cells in series that can operate at a much higher and more usable voltage.

Thanks for your reply, folks. It sounds like you are both saying the same thing (dual DC-DC converters). Though yes, the LTC3105 and LTC3108 do require higher input voltages to starts, so I cannot use them.

Unfortunately the cells I am using are special because they are specifically made to measure irradiance at a fine-level. I am actually switching them on at intervals to then perform the energy harvesting. The irradiance measurement takes precedence over the energy harvest.


I forgot to tell you that the BQ24195 power management and battery charger IC is connected to a Li-Ion battery. The primary reason for the energy harvesting through these unique solar cells is to recharge the battery through the BQ24195. Would it not be possible to design a joule-thief circuit with regulated output and connect this output to the power management IC. The BQ24195 has the following input parameters:

Vin = 3.9V to 17V (though safely, I'd say no more than 12V)
Iin = 3A max (there is no minimum)


From the circuits I've seen, this is possible with a regulated joule-thief. Though I have never made a joule thief before but I fear the high current could damage components if not selected carefully.

The difficulty is, that on a cloudy day there may be only 150mA of current available from the solar cells.
If your boost converter needs to draw very short one amp pulses, to deliver say 80mA of current at a much higher voltage, its simply not going to work.

You may need something like an ultra capacitor that can charge at 150mA (or whatever is available) and deliver very short surges of relatively high current into whatever increases the voltage.

These ultra caps are physically large and expensive, but its the only way I can see this ever working.

You would then allow the ultracap to charge up to the MPPT voltage, and pulse your voltage booster on and off at a rate that just holds the ultracap voltage at the most efficient 1.02 volts.

In twilight conditions it might only be able to pulse once every ten minutes, in bright sunlight it will probably be going pretty fast.

It would be most efficient to pulse straight into the battery, let the pulser set its own rate to hold the MPPT voltage constant.

Then use your high tech battery management chip only to turn the pulser on when charging is required, and off when the battery is fully charged.

If your PV cells are like raw crystalline types I've worked with, then you can cut them into pieces, and hook them up in series. It gives you increased voltage at reduced current. The trick is to break them only where you want them to. It's a further trick to attach wires for an electrical connection. The liquid silver (tube or epoxy) is ideal, I found.

Unfortunately the cells I am using are special because they are specifically made to measure irradiance at a fine-level. I am actually switching them on at intervals to then perform the energy harvesting.

Click to expand...

o/k fine use these special cells for your monitoring.

But use something completely different for your main power source.
Doing it any other way will be complex and inefficient.

fishmastaflex said:

Though yes, the LTC3105 and LTC3108 do require higher input voltages to starts, so I cannot use them.

Click to expand...

The LTC3105 requires 0.25V and the LTC3108 requires 0.02V, so I'm not sure what you're talking about.

A large, low ESR cap at the input might indeed be necessary, depending on the cells' load line. A small ultracap should be fine.

If an ultracap is used, the system can be designed to start and run at 1.0v and need never see voltages much lower than that.

The required eight amps average charging current is a pretty big big ask.
Peak currents will be several times the average, there will be no nice simple one chip solution to that !