Peltier Candle Powered JT

[unzapped]

I have had a weekend of experimentation... Mostly failures with a little success at the end.

I am trying to develop a replicateable petier(seebeck) candle powered JT circuit. I can get it to work alright however I would really like to get my hands on a really low input transistor ( currently using 3904). My lights do not come on until .7 volts now. This takes a few minutes.

Also I would like to power a small motor fan for the heat sink. From this same circuit.

Work flow goes like this...

light candle
adjust height under peltier
light comes on
fan kicks in
light gets brighter
no input except the peltier/candle

I have absolutely no idea how this circuit will look except that I am using the jeanna light circuit to power the lights.

Maybe even integrate and ultra cap as long as I can recharge it from the peltier.

ANY help is much appreciated.

Thanks
UZ

[pese]

I have had a weekend of experimentation... Mostly failures with a little success at the end.

I am trying to develop a replicateable petier(seebeck) candle powered JT circuit. I can get it to work alright however I would really like to get my hands on a really low input transistor ( currently using 3904). My lights do not come on until .7 volts now. This takes a few minutes.

Also I would like to power a small motor fan for the heat sink. From this same circuit.

Work flow goes like this...

light candle
adjust height under peltier
light comes on
fan kicks in
light gets brighter
no input except the peltier/candle

I have absolutely no idea how this circuit will look except that I am using the jeanna light circuit to power the lights.

Maybe even integrate and ultra cap as long as I can recharge it from the peltier.

ANY help is much appreciated.

Thanks
UZ

use for low Vce(at) an germanium transistor
OR an Silicon Type from FERRANTI
It is only 1 silicon transistor (in wold) that work
on voltages below 0,7volts properly.
perhaps i find out the type numeber if need..
prferred (not so hard to find) Germanium pnp.
Pese

[resonanceman]

I have had a weekend of experimentation... Mostly failures with a little success at the end.

I am trying to develop a replicateable petier(seebeck) candle powered JT circuit. I can get it to work alright however I would really like to get my hands on a really low input transistor ( currently using 3904). My lights do not come on until .7 volts now. This takes a few minutes.

Also I would like to power a small motor fan for the heat sink. From this same circuit.

Work flow goes like this...

light candle
adjust height under peltier
light comes on
fan kicks in
light gets brighter
no input except the peltier/candle

I have absolutely no idea how this circuit will look except that I am using the jeanna light circuit to power the lights.

Maybe even integrate and ultra cap as long as I can recharge it from the peltier.

ANY help is much appreciated.

Thanks
UZ

Unzapped

I can't help you with the electronics ......many others here have  alot more experience  than I have.

I do  have a few suggestions though

A peltier device is not tthat efficient
I have heard that they are only 5 or 6 % efficient.

It might  be easier to get it working well  if you used  a small sterling engine....80% efficiency is not  uncommon.
Of course  it would need a generator ..... I think  even with the losses  of the generator  you would come out  ahead with the sterling.

I also think that if you are only using one candle .......a fan is overkill
a heatsink made for a CPU should  handle the heat  easly with no fan

If you wanted to use more heat........ I think I would try adding a cooling tower  kind of like a nuclear reactor .....I think I would  try a clear plastic tube about 12 to 18 inches high
......slightly narrower  at the top

gary

[Pirate88179]

I agree about the germanium transistor.  I also agree totally with Gary that the peltier is not the way to go in this application.  You are burning up the wick and the wax to produce so little output.  Why not just use dead batteries, which are free, and be done with it?  If you can't do that, build an EER and run the JT for free.

The only way this would make sense to me is if you already have the heat, for free.  Like hot water heaters that you are already paying for. (hook it to the pipe that is always hot anyway)  There are just too many more efficient ways to run a JT that I would do long before this option.

Now, if you get your candles for free, then this would change the equation a little bit but still, as Gary said, then the Sterling would be the way to go.

Try it yourself and see.  None of us here know everything.  Peltier junctions are a neat thing to experiment with anyway so enjoy yourself, and let us know how it goes.

Bill

[pese]

you cane make with cascadin arry also ANY needed voltage.
GRUNDIG have produced for Arabia-export. an petrol-
lampe that was to use with black/whith TV Sets that consumed 220V at abiut 200 watts.
only peltiers-ekements made from some hundred point contacted wires, surrounding  the center petrol-flame.
GP
So zts not need to use any inerters for higher voltages.