Capturing Tiny Amounts Of Electricity At A Time?

[Jay Lakoda]

A small thermocouple would produce tiny amounts of electricity as temperature unevenly fluctuates. But if I tried to store that energy using a diode and capacitor, I would guess that the tiny reverse current flow through the diode would be enough to not allow for such a self charging system to work. Is there a "perfect" diode? Not necessarily a diode, but just something that would act like a perfect diode? Or is there another practical way to charge a low energy system that does not involve a chemical battery, using just natural fluctuations in air temperature, sound, electromagnetic noise, etc? I'm looking for something very small that could fit in a microchip.

 

[berkeman]

A small thermocouple would produce tiny amounts of electricity as temperature unevenly fluctuates. But if I tried to store that energy using a diode and capacitor, I would guess that the tiny reverse current flow through the diode would be enough to not allow for such a self charging system to work. Is there a "perfect" diode? Not necessarily a diode, but just something that would act like a perfect diode? Or is there another practical way to charge a low energy system that does not involve a chemical battery, using just natural fluctuations in air temperature, sound, electromagnetic noise, etc? I'm looking for something very small that could fit in a microchip.

The wikipedia article on Energy Harvesting has some good information: http://en.wikipedia.org/wiki/Energy_harvesting

Although on a quick skim of the article, I didn't see if they addressed how to deal with the small voltages generated by harvesting. If you can stack many small harvesting sources in series, you can use a low Vf Schottky diode (around 0.2V for low currents). But I'd be interested in hearing of other options...

 

[Baluncore]

Is there a "perfect" diode? Not necessarily a diode, but just something that would act like a perfect diode?

Perfect rectifiers are used for the freewheel diode in switching power supplies. They are called synchronous converters when the PN diode is replaced with a MOSFET.

To make a perfect rectifier, use a MOSFET and a comparator. When the polarity of the power source is correct, the comparator drives the gate of the FET to turn it on. You can make a bridge rectifier using that technique, it requires four mosfets and one voltage comparator.

 

[mfb]

Perfect rectifiers are used for the freewheel diode in switching power supplies. They are called synchronous converters when the PN diode is replaced with a MOSFET.

To make a perfect rectifier, use a MOSFET and a comparator. When the polarity of the power source is correct, the comparator drives the gate of the FET to turn it on. You can make a bridge rectifier using that technique, it requires four mosfets and one voltage comparator.

That might be possible, but it defeats the idea of energy harvesting if you need external power to drive your system - or does that system work with very small voltages as the only power input?

 

[sophiecentaur]

Before you get too involved with the idea of energy harvesting it is as well to do some calculations (same as you would if you were planning to buy a car and you had to decide if you had enough income - that's not being patronising, btw - just encouraging realism).
When you only need very small amounts of Power or Energy and it's needed somewhere very remote, then it may be essential to make use of very modest energy resources ('harvesting'). If you want to heat a house or run a vehicle then there is seldom anything like enough harvestable energy available -especially where there are no sources of wasted energy from man made systems.
How much energy / power do you actually need, for a start; what do you envisage using it for?
I have a theory that there is much more useful ingenuity that can be spent on Saving the energy in the first place. You can save hundreds of kWhrs by insulating and turning things off when not needed. This tends to compare with harvesting in the order of Watt hours in many cases.
Storing energy from a thermocouple implies that there is significant energy to be available in the first place. Thermocouples are incredibly small sources of energy - more useful as thermometers and in gas flame safety devices.
Can you answer the question 'How much Energy per day / week will you need and what free source do you propose to use?'

 

[Baluncore]

That might be possible, but it defeats the idea of energy harvesting if you need external power to drive your system - or does that system work with very small voltages as the only power input?

CMOS comparators and low capacitance MOSFETs consume very little power. Energy use is proportional to the number of switching transitions. These days it takes about 3 volts to operate the comparator and MOSFET gate, but that voltage can come from all sources of power available.

Voltage is not power until it is multiplied by current. Compute your energy budget to find the power available.

 

[Svein]

A small thermocouple would produce tiny amounts of electricity as temperature unevenly fluctuates. But if I tried to store that energy using a diode and capacitor, I would guess that the tiny reverse current flow through the diode would be enough to not allow for such a self charging system to work. Is there a "perfect" diode? Not necessarily a diode, but just something that would act like a perfect diode? Or is there another practical way to charge a low energy system that does not involve a chemical battery, using just natural fluctuations in air temperature, sound, electromagnetic noise, etc? I'm looking for something very small that could fit in a microchip.

What came into my mind was a CCD - and then I remembered digital cameras where you really have to catch and process minute amounts of electricity. Could you use that technology?

 

[Jay Lakoda]

Here's what I'm trying to do...

I don't yet have a reason or goal in mind. I'm trying to just learn more about building electronic systems. And I prefer to learn in an unconventional way.

So right now I'm trying to develop a pointless, but fun and educational system.

I want to litter a square mile of desert somewhere with tiny microchips the size of an rfid tag. Each chip would be roughly 100 feet from the next nearby chip. And when I send one chip an 8 bit rf code, it will repeat the transmition with just enough energy so that the next chip 100 feet away can receive the code, and re-transmit the code. It would be a repeater system. Every chip in the area will quickly end up repeating the same code. Then the system would take a day or so to recharge it's own power supply, so that each chip can again do the same thing tomorrow.

 

[Svein]

Here's what I'm trying to do...

I don't yet have a reason or goal in mind. I'm trying to just learn more about building electronic systems. And I prefer to learn in an unconventional way.

So right now I'm trying to develop a pointless, but fun and educational system.

I want to litter a square mile of desert somewhere with tiny microchips the size of an rfid tag. Each chip would be roughly 100 feet from the next nearby chip. And when I send one chip an 8 bit rf code, it will repeat the transmition with just enough energy so that the next chip 100 feet away can receive the code, and re-transmit the code. It would be a repeater system. Every chip in the area will quickly end up repeating the same code. Then the system would take a day or so to recharge it's own power supply, so that each chip can again do the same thing tomorrow.

Oh, so you want to use these small amounts of energy as power - I did not get that.What's wrong with a solar cell?

 

[Jay Lakoda]

Oh, so you want to use these small amounts of energy as power - I did not get that.What's wrong with a solar cell?

A solar cell is kinda like plan B. I would like the system to operate in any location, such as the bottom of the ocean, or in Antarctica where a solar cell would get almost no energy. Unless there are infrared solar cells. Do those exist?

 

[mfb]

In principle, you could build one, but then you have to cool it below the temperature of your IR emitter to work - not practical.
I guess specific applications will need specific designs. Temperature differences occur nearly everywhere, but on the size of a chip they are really tiny.

 

[Jay Lakoda]

Lol. Ok. So a quiet dark corner in the closet is not the best place to find energy. I'll have think about some other design. Thanks for the responses and the entertaining discussion. Science is awesome.