[FoRK] Solar power

Eugen Leitl eugen at leitl.org
Mon Feb 9 08:34:55 PST 2004

On Mon, Feb 09, 2004 at 03:38:34PM +0000, Russell Turpin wrote:
> Eugen Leitl:
> >Why can't I use a small generator for overnight/
> >spike use, and avoid batteries altogether?
> (1) The general principle behind PVs or some
> other form of passive generation is to bank the
> spare amps from peak times, for use at other
> times. Note that the amps don't just turn on

I'm familiar with the classical approach. I'm interested
in whether prices have come down sufficiently so I can address some issues
with brute-force (overdimensioning the PV array).

> at dawn. The amount supplied varies depending
> on the height of the sun, its angle to the

I'm familiar with that.

> panels, and how much it peaks from behind the
> clouds. On a sunny day, you get peak amps only

Amorphous does better for diffuse radiation, is also cheaper W/EUR -- it
ages less well, though, and it probably doesn't integrate into the facade as
well as trimmed down monocrystalline. Right now it seems some 1 kW is about 6
kEUR (peak), though I presume I can do a lot better if buying directly from
the manufacturer. If integrated into facade, it would protect the insulation,
and should last 25-30 years with proper cleaning. 

> for a couple hours either side of solar noon.

I'd be a fool if I planned to run the thing by peak amps alone.
Unfortunately, you can't buy water electrolysis systems/hydrogen pressure
tank for home use yet, so I can't store the big oversupply (other than e.g.
running the refrigerator, electric cooking, heating, pumps, etc). Neither 
fuel cells, nor fuel cell cars (a fuel cell
car would do about optimally for overnight power spikes).
> (2) Generally, engines work inefficiently and wear
> out faster if they are constantly turned on and
> off. Most generator engines are designed to run
> at a constant speed and relatively constant load.
> Their use model is precisely that they are either
> (a) matched to a proportionate load, e.g., some
> piece of machinary they power, or (b) they
> dump their amps into a battery bank, and then
> come on again when it is drawn down.

This is good info, thanks.
> (3) You're going to need at least a small battery
> to turn on the generator when it is needed,
> unless you're planning to hand-crank the
> generator every time. So really, you're talking
> about eliminating the "house" bank, not all
> batteries in the system.

I'm not planning to eliminate all batteries altogether, just big banks for
high-power use. LED and fluorescents draw very little power, and of course
I'm not planning to touch the generator apart from refuelling and servicing.
Even low-end models are push-button nowadays.
> (4) The generator doesn't come on instanter.
> The time it takes for some sensor to detect
> that your PV amps have dropped, to signal
> the ignition, to throw the solenoid, to turn
> over the starter, to catch the engine, to
> start generating amps is plenty for your PC
> to reboot from loss of juice. Of course, you

The house lighting and the 12 V/small inverter-powered systems will be
battery-buffered. It doesn't require a lot to buffer a few min of continuous

> can solve this with some kind of UPS. But now
> you're opting for expensive, small capacity
> batteries over cheaper, big capacity batteries.

An UPS is a good thing to have in case you get spikes in your power system.
Can you at all run a vanilla PC off a nonsinus power? 
> (5) What are you going to do for light when
> your generator needs some work? It will.

LED and fluorescents should be good for several days on battery charge alone,
if no other sources are draining the battery. 

> And if sailing experience is anything to go by,
> it usually chooses the most inconvenient
> time to stop working, say a cold, cloudy
> morning before you've had your coffee.

Both the heating and cooking isn't affected 
by the generator. Light wouldn't be a problem in above scheme as well.
> Bottom line: lots of people have lots of
> experience powering homes, work posts,
> and boats with a generator and/or PVs, not

In most cases, these are systems providing few 
10-100 W, not kW.

> because they have chosen to live "off the
> grid," but because the grid doesn't go
> where they are. Universally, they use some
> kind of deep-cycle battery bank to store
> amps and buffer the need for generated
> amps. What you're proposing runs against
> broad engineering and practical experience.

I agree I'm pushing the envelope here. Question is, is this far enough to be

-- Eugen* Leitl <a href="http://leitl.org">leitl</a>
ICBM: 48.07078, 11.61144            http://www.leitl.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE
http://moleculardevices.org         http://nanomachines.net
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
Url : http://lair.xent.com/pipermail/fork/attachments/20040209/4c1c82c5/attachment.pgp

More information about the FoRK mailing list