|
|
re: honda generators vs the big boys
24 oct 1998
buzz cut aka dave wrote:
>...years ago i asked one of our power engineers how they sync'd the
>multi-megawatt generators we used to the grid. he told me that the
>generator didn't have a chance and would be forced into sync...
why not sync, if you were a generator? they don't have crystals like
watches. no rigid phase or frequency references. no digital governors,
yet. it seems to me that it would only take a small push to get them
to slowly sync to the grid, ie a large series resistance to start with.
>how do you plan to transport the waste heat to the house?
i'm not sure. i was planning to circulate house air through the generator
box when it wasn't running, but you bring up an interesting point. i don't
want my house to smell oily... yuck.
what i had in mind was to put this 90 pound $900 1500 w gasoline generator
in a small insulated room in the basement, surrounded by lots of smallish
sealed containers of water, (eg about 600 4 gallon recycled 9x9x13" tall
stackable plastic tubs that used to contain dried cherries) with a fan
blowing on it. and a small exhaust fan from the room to the outdoors that
runs when the generator is running to slightly depressurize the room,
as in a radon exhaust system.
the generator would only run a small fraction of the time, in my mostly-
solar heated house, when it's cloudy outside for more than a few days in
a row. i'd planned to have another fan to circulate house air through
the box, but maybe the box should have some fin-tube pipe near the top
(which would let me heat water for showers, too) or an old auto radiator
and fan inside like the 1984 dodge omni now under my living room floor,
or less insulation so more heat can leave the box by conduction. maybe
a concrete block enclosure in the basement to deaden the sound, with the
blocks stacked up sideways to expose the holes and increase their surface,
and some rocks inside each hole. that would cut the number of 4 gallon
tubs needed to about 200.
my house has stone walls, with insulation outside, and i'm filling the
basement with more thermal mass, lots of water containers and about 3 tons
of 20' long steel channel, so the house temperature changes slowly. the
genny box doesn't really need thermal mass in my house, just lots of
internal and external surface area to transfer heat by conduction, and
maybe some internal heat storage to allow a lower average continuous
heat transfer rate, some sort of heat accumulators as used on the outside
of 1930s refrigerators, and efficient sunfrosts, today.
>up until about the mid 60's vw used the engine cooling air to directly
>heat the interior. it worked great until the engine developed even a
>small oil leak. then the smell was quite strong...
not good.
>they changed to a exhaust heat exchanger scheme and still use the same
>method with the mexican air cooled bugs made today. either way,
>overall efficiency couldn't be too good in my opinion/guess...
a 55 gallon drum full of water with a few feet of 1/2" copper pipe with
holes in it near the bottom might make a good exhaust heat exchanger, but
i don't want to waste the rest of the engine heat. i'd like this thing to
be at least 90% efficient, adding the heat and electrical output power,
then dividing by the fuel's heating value, ie less than 10% of the fuel's
heating value leaves the house, vs about 2/3 in a power plant. intelligen's
5 kw residential diesel cogen system was 93% efficient, in this sense,
with a 150 f exhaust leaving the house in a 1.5" pvc pipe.
>your idea sounds like it should work and will protect the equipment even if
>it doesn't. let us know how it works!
i'm going to try this sometime with a rental unit... i'd like to find out
if my meter's racheted so it only goes one way, as well. if so, i may have
to give away the electrical power, or get a few batteries and a trace
inverter or bugger the meter until the pa puc eliminates all the regs
about parallel operation. hmmm. peco now pays 22 cents/kwh for pv power
delivered at peak times... maybe this system needs a nominal pv panel.
>>seems like the expensive part is the grid-tie interface. suppose we use
>>a couple of 100 light bulbs in series with a circuit breaker across them.
>>start up the honda, wait until the light bulbs are dark, then connect
>>the breaker...(?) anyone tried this?
one more thing. something to make the power delivered to the grid about
1.5 kw, eg a variac to step up the output voltage, and maybe another in
series...
nick
|
|
