"Pulse" gel-cell charge strategy?

Farad level caps are probably your best bet. I found they have got quite a bit cheaper/more available, and with tens of milli ohm impedance.

-jg

Makes sense. What's the rise time of your pulses? If it is fast you might want to add a big hulking inductor to treat the caps nice and reduce IR losses and/or wire size. I'd sure like to know what your power source is.

It's a sort of internal combustion engine powered by a slow seepage of gas. The gas, while mainly methane and oxygen with a little hydrogen, is severely impure and not suitable for feeding to a fuel cell. A gas bubble forms slowly; when it's big enough (minutes), my controller fires an igniter that burns the gas and fires a rare earth magnet down (or rather, up) a coil that provides my electrical supply voltage. The magnet has a groove around its lower rim that provides turbulent obturation during the expansion phase. The controller has other responsibilities too, but that's basically it for the power system.

What kind of rare earth magnet doesn't lose its magnetic properties over time with mechanical and heat shocks?

Great idea though.

An open end at the top of the coil would produce nice smoke rings

w..

That would be a "not my problem" problem :) I don't do thermodynamics and mechanical engineering; all I know is I get somewhere between

400-600W @ roughly 20VDC for a few tens of milliseconds every couple of minutes or so. Having said that, I'm not sure how vigorous the expansion phase is, nor how warm the magnet actually becomes.

I prefer this expressed more like, "about 10 Joules every few minutes." Do you have a figure on the Joules yielded and an estimate on the methane volume (I assume at atmospheric pressure?) Just out of curiousity...

Jon

"10J every few minutes" implies to my mind slow and steady delivery - which would be a different problem to be solved, I think.

Each stroke burns enough fuel to release maybe 350J of energy. It's quite variable though because the concentrations of the reagents fluctuate. There is a lot of argument in the pistons-and-PV-diagrams camp right now about the design of the hardware; I'm not sufficiently erudite to follow it. I've been assured that the electrical characteristics won't change much if at all.

Earlier this was 300A - did you mean 30A ?

The 10J mentioned is not large in the scheme of things, I get around 3.2F at 2.5V and 32mF at 25V (assume full dV) - at such short times, the series R of the 'SuperCaps' becomes an issue, so you might do either a combination, or just look at a FAT 25V Elco ?

ie a 100mF 25V cap, changing 11.25 Joules moves from 20V to 25V

-jg

No, I meant 300A, I was tinkering in the simulation spreadsheet and accidentally deleted a zero :) I blame the fact that I am currently about 150lb heavier than normal due to being stuffed with Thanksgiving food.

Wow - So that's 6000W (?!) of peak power, or 150W over 1 second, or ~1W over 150 seconds

How big is that magnet ? ;)

How much of this 6000W do you actually need to capture ?

How many milli-ohms is the 'generator loop' ?

-jg

It's roughly roughly 6kW for a few tens of ms; on average, somewhere around 25ms, once every couple of minutes (that couple of minutes figure is extremely loose, but it will never be LESS than a couple of minutes - it could stretch to four or five, though).

Big enough that you have to plan a route exactly when transporting it, you may not allow more than one in the same room, and you should not bring it within spitting distance of any CRT monitors. But it's more the "propelled at speed through coil" aspect that makes it electrically frisky.

Very little. My circuit VERY RARELY dissipates about 25W; however most of the time it's cruising in the region of 200mW. The more I could harvest, the better though - all sorts of interesting things could be done with the excess. The other thing is that the fuel source is very erratic. Sometimes the gas mix that's harvested won't contain enough available energy to actually drive the cycle, in which case it gets "burped" and collection resumes.

It's about 52mohm; and, I might add, a scrap metal thief's dream come true. Which is a bit of a problem, but again not mine :)

Very cool project. Thanks for sharing more about it. Pictures would be greatly appreciated if ever possible.

I get the impression of a free piston engine and when enough methane and air accumulate then bang. The piston races up through a cylinder wrapped in wire. You capture the transient. The piston returns maybe even with gravity and the electronics lies in wait for the necessary gas to accumulate before sending the piston again for a ride of a life time.

Mr. Goldberg would have been proud.

Nice project. With global warming I can just see great opportunities for such a device in the Canadian arctic. The permafrost is melting the amount of constable gases being released is huge.

Regards,

Walter.. __ Walter Banks Byte Craft Limited

That's it.

That is exactly why the "piston" uses turbulent obturation. The compression/exhaust stroke works by slowly venting the exhaust gases around the edges of the magnet and simultaneously condensing the steam that formed during the expansion phase. If the piston was close-fit to the cylinder, it would be necessary to have all manner of valves and such. Yes, we lose some energy to leakage, but we get it paid back in reliability. Mortars often use the same principle - you drop the shell down the barrel, and it moves relatively slowly so there's plenty of gas leakage around the edges. Once it hits bottom, the propellant fires and the huge pressure differential creates rapid flow, lots of turbulence around the obturating ring and your shell gets pushed off towards the enemy...

Sounds like a neat idea for dairy farms and landfills. Lots of methane generated in manure digesters and in the land fill. Our local landfill runs a gnerator with some of their gas---but the gas requires a lot cleaning lest H2S corrode the generator.

I wonder if it might be simpler to have the coil be open circuit on the upward movement of the magnet, then capture the engergy on the downward fall of the magnet. You could arrange the coil impedance and gas venting rate so that the down stroke might be 10x slower than the upward.

Another off-the wall idea: Use the main piston as a pump to push the air above the piston into a reservoir with a simple flap valve at the entry. You can then bleed the pressurized air through a turbine to capture energy.

This sounds like enough fun that I may have to go out and buy some PVC pipe and indulge my inner Rube Goldberg.

Mark Borgerson

Yes, this project has similar problems to solve. Hence the desire for no unnecessary moving parts.

Don't forget to add 1100ft of AWG 0000 copper wire to your shopping list. You may wish to rent a truck... and an armed guard ;) Of course you could make a scaled-down version with magnet wire, but that simply wouldn't be fun.

I should point out that in my case the bottom (harvester) end of this gadget is underwater; gases are evolved by reactions at the bottom, and fill the combustion chamber. Atmospheric air is also bubbled in. The "breech" seal, as it were, is simply a column of water. This is also a safety valve - overpressure in the "barrel" is impossible.

There are a lot of mechanical engineering subtleties in the design, most of which I don't know (and likely would not understand unless they were explained to me patiently).

So this project has no dollars per watt metric, that it needs to meet ? ;)

Surely the power generation is _not_ the primary goal here ?

We have Bird Scarers here, that use similar 'sporadic explosion' basis

- called 'Bird cannon'

-jg

It all depends where you put your break-even point; a thousand-year payoff? :)

No, this is waste energy in fact.

They're propane though, aren't they?

I was considering my alternative plan to (in smaller scale) use the methane combustion to compress air---then vent the compressed air through a turbine. That lets me spread the power out over a much longer time and avoid the requirement for all that copper wire.

Mark Borgerson