Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I wish someone with the skill to do this would use their time to push the state of the art in small generator design. Here is a sampling of parameters, all of which I think could be improved:

Cheapest: Harbor freight's "Storm Cat" $129, 2-stroke, 700W, 91 dB, 63cc, 38.2 lbs, 18.3 W / lb, 1818 Wh / Gal.

Best power / weight, most efficient: Honda "EU2000i" $1000, 4-stroke, 1600W, 53 dB, 98cc, 46 lbs, 35 W / lb, 6737 Wh / Gal.

Quietest: Honda "EU3000is" $2000, 4-stroke, 2800W, 49 dB, 196cc, 134 lbs, 20 W / lb, 5939 Wg / Gal.

Lightest: Honda "EU1000i" $800, 4-strike, 900W, 53 dB, 50cc, 29 lbs, 31 W / lb, 5700 Wh / Gal.

(there are ~33440 Watt-hours of energy available in a gallon of gasoline).

The most recent innovation seems to be the inverter/generator. It means that the engine does not have to run at any specific speed to get 60 Hz, so it can be run slow to make it quiet. Perhaps an extension to this could be the generator equivalent of a "hybrid"- supply power from a battery to cover surges, but use the engine to supply the average usage.

It will be interesting if any recent work on engine design ends up in small generators. For example, there is the OPOC engine from the Bill Gates backed startup http://ecomotors.com

You would think there would be more competition from micro-turbines, but I don't think they scale down so well. There is this 2.6 kW "Dynajet" micro-turbine: https://www.youtube.com/watch?v=wa8kS8EZWmQ

Edit: also.. what ever happened with small fuel cells? http://www.forbes.com/sites/michaelkanellos/2013/01/31/why-a...



The other day I saw a story on a new engine design which claims to be 125% more efficient: https://www.youtube.com/watch?v=ChLFFT08eHg

Here (http://www.nytimes.com/2000/09/19/science/putting-a-darwinia...) is an old story about a group who used genetic algorithm to design a more efficient diesel engine.

>using a Silicon Graphics Origin 2000 supercomputer with 32 processors it took more than two weeks of continuous operation to find an optimal set of parameters.

>Dr. Senecal's test engine consumed 15 percent less fuel than a standard engine while producing one-third as much nitric oxide and half the soot.

Unfortunately I have no idea if anything ever came of this or not. I can't find anything. It'd be incredibly sad if this actually worked and then was never implemented.

This was done just with a dumb genetic algorithm on 2000 era computers, a population size of 5, run for 50 generations, and only tweaking a handful of parameters. I believe there is more potential in stuff like this.


It sure has not been from lack of trying new designs: http://www.douglas-self.com/MUSEUM/POWER/unusualICeng/unusua...

(Also: Big News! Douglas Self has been updating the Museum of Retro Tech again)


this guy builds models,they work on compressed air, they are not actual combustion engines.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: