A pulse-charge energy harvester

Hi Kjetil,

The difficulty with this is that a capacitor by nature has a voltage that varies with the amount of charge that it stores. A specific capacitor may have 100 volts at full charge. Then when you go to remove the charge every little bit of energy that you withdraw lowers the voltage. Yes you might be able to use a voltage regulator to control the voltage for a short period until the capacitor's voltage gets below the requirement for the regulator. If you wanted it to be a pulse charger you could used a PWM or square wave circuit to generate pulses. The bottom line is that due to the usable voltage range of the capacitor, the inefficiency of the regulator, and the energy cost to power the other circuitry you will only get perhaps 10% of the energy of the capacitor to actually add any charge to the battery. A more efficient use of the energy available would be to charge your battery more directly from the same source that you intended to use to charge the capacitor. You have not given any information on the rest of your system or why you want to design the way you have explained so it is difficult to comment further.

John

You have summed up the problems, but that is no solution.

10% would be just fine if the result was a charger with a stable voltage and amp in every pulse.

The problems connected doing a  'charge your battery more directly from the same source' is the reason I want a charges like the one that I described. If the energy has to be used directly ther will just not be enough to drive a PWM or square wave circuit to generate pulses most of the time,.

If you can charge a capacitor with your energy source you should be able to charge a battery. The energy that is store in the electro chemical battery is much more usable as a power source than a capacitor's electric field. You could charge some secondary batteries and then reconfigure them to power your pulse charger that charges the primary system. If I am still not answering your question you will have to provide more information so I am not swinging in the dark.

John

It's diffult to charge batteries when the energy comes from something like a 50kg droping 3 meter to the floor. This short burst off energy can be stored fast in a capasitor, not so fast in a battery.

I can get 5 super capacitor 2.7V 100F for 10$. This type will fully charges loose 0.4V in an hour, but that no problem since they ar not primary storage.   .

Hi Kjetil,

I too didn't understand the need to charge multiple devices in a chain, but maybe you have your reasons. I think it would be difficult and quite complicated.

An alternative approach is to just use your supercaps as the primary storage - i.e. not charge a battery, but just directly power your projects from the supercap. You'd need some circuit to step-up and regulate the voltage to suit your circuit's needs.

You need to incorporate intelligence to only switch on bits of circuitry when you have sufficient power to complete the expected task.

For that approach, check out some posts by tomaja and janisalnis and others - they are using some TI devices that have this capability although their topology may be different than what you need, it would be a good start.

How about a capacitor in parallel with a chargeable battery. The capacitor would capture the energy and then you could bleed it to a battery through a resistor allowing the battery time to do the chemical process. Your 50 kg fall of 3 feet gives us about 1500 Joules. In freefall the mass takes about 0.78 seconds to get to the floor. This means you will have to be able to have a mechanism that can convert about 1900 joules per second of kinetic energy to electrical energy. This is a limit as practically your conversion device will probably slow the fall of the mass and a share of the potential energy will be lost to friction and conversion inefficiencies. What kind of an energy capture and conversion device do you envision?

John

That parallel setup sounds good , but there is now slowing down included. Energy will be generate using several coils and strong magnet that get pushed true the coil. This should double up as a set of shock-absorbers. Springs will lift the plate back into start position.

If you take the energy out and convert it to electricity the mass will settle onto the springs like a feather and slowly compress them to equilibrium. You will then have to put in at least 1500 Joules of energy to get it back to the start position.

John

Sorry that is't wasn't clear, but the the weight that drops will move off the plate. Then the springs can lift the plate.