http://enews.lbl.gov/Science-Articles/Archive/MSD-fuel-cells.html

Looks like Lawrence Berkeley Labs has solved one of the knottiest
problem regarding solid-oxide fuel cells.  Instead of using a
structure of zirconia, the researchers deposited a stack on top
of a piece of sintered stainless steel powder.  This has the
advantage of being very cheap compared to zirconia ceramic.

At $200/kw and 50% efficiency, a furnace burning 50,000 BTU/hr
of fuel would cost about $1500 and be able to produce about
7.3 KW of electricity.  The typical home uses about 1 KW on
average.  The home would also have the use of 25,000 BTU/hr of
waste heat for heating water or air.

If we were looking for ways to slash our fuel consumption, this
is a golden opportunity.

37 responses total.

I expect my new house to use as much as 8000 btu/hr of heat during the very
coldest days, which comes out to the amount generated by producing 1.1KW
electricity, which is probably a lot more than the house would use but could
be used for hot water.  Sounds promising.  From your short summary I do not
understand what fuel is used or what it costs per KWH - can you give more
specific details?

Re #1:  A watt is about 3.41 BTU/hr, so your heating requirements
actually run to almost 2.5 KW.  Perhaps you misread my meaning, or
intend to meet part of that figure from other sources?  (My figures
were based on an efficiency of 50%, so half of the input energy
would appear as electricity and half as waste heat.)

Total cost is the sum of fixed cost and variable cost.  Variable costs
depend on how much the fuel costs and how much wear and tear things put
on the system.  At 50% efficiency and fifty cents a therm for natural
gas, the fuel cost would be 3.4 cents per KWH, plus you'd get about
3400 BTU of heat; that's enough to heat about 6 gallons of water from
50 F to 120 F.  "A kilowatt-hour from your morning shower"?

Hopping on my hobby-horse for a second, if your house uses 6,800 BTU
per hour, you'd co-generate 2 KW.  If this charged an electric car
which used 250 watt-hours per mile, the electricity could run the car
192 miles a day.  If you drove a more typical 50 miles a day, you'd
have an excess of ~35 KWH.  Used in an EER-15 heat pump, this 35 KWH/day
would give another 22,000 BTU/hour.  Total production:  50 miles
travel, plus ~28,000 BTU/hour of space heat from ~14,000 BTU/hr of gas.

(This looks like "something for nothing", but all the "excess" energy
comes from low-grade heat outdoors; the trick is making maximum use
of high-grade energy to push heat around instead of diluting it
directly to low-grade heat and just accepting the entropy increase.)

The solid-oxide fuel cell that I recall could burn just about any alcohol
or hydrocarbon.  As it works by moving oxygen (O--) ions through the
electrolyte, anything which can consume oxygen at the anode will create
the oxygen gradient required to make it go.  I expect that you could use
this device with natural gas, ethanol, methanol, gasoline and perhaps
biodiesel.  There may be some restriction on fuel sulfur content.

A typical gasoline engine is roughly 30% efficient at its peak.  If this
thing hits 50%, it could boost the Honda Civic Hybrid from 50 MPG up
to eighty-plus.  That's just the beginning.  If this can be translated
to a solid, reliable product, it is going to be huge.

The calculated heat requirement for the house at 0 degrees F outside is about
2.2KW, or about 6000 BTU if I recall correctly.  We can heat on time of day
rate for about 2.3 cents/KWh.  My electric bill at two addresses (without heat
or hot water) totals as much as $20/month at 8.x cents/kwh, or about 8 KWh/day
(to run a large freezer, two refrigerators, electric stove, several computers
since I work at home).  The 6000 BTU is the maximum figure for heat and might
average half that over the winter.  3000x24 = 122,000 BTU/day for space
heating and I think you calculated 6 gal of 120 degree water (more than I am
likely to use in a day - I shower at about 95 degrees) = 3400 BTU, call it
125,000 BTU/day heat.  At 50% efficiency the same amount of electricity, 3400
BTU/h per watt, divide 125,000 by 3400 = 36 KWh/day, over three times the
electricity I could use in order to generate heat on a cold day, but
reasonable for not so cold days.  I calculated at one point that in January
the heat bill would be about $40/month at 5-6 cents/KWh = about 25 KWh/day
of heat, which is still 3 times the electricity I would be using.

This means having to sell back electricity to the grid even during cold
weather when running the system enough to produce all the heat and hot water.

In November or March it might balance out, but during April-October I would
still have to buy electricity from the grid (or invest in some expensive solar
equipment).  

Am I off by any powers of ten somewhere?  

The power company does not particularly want more electricity in the winter,
which is when such systems would generate excess, and most people use more
heat at night when it is colder out, which is just when there is least demand
for electricity.  People with air conditioners use more electricity during
hot weather when nobody will be cogenerating electicity along with heat.
You would really have to combine solar generation with this sort of system
to balance the loads out.

I think that Detroit Edison (or whatever it is called now) is paying such low
prices for electricity that you sell to the grid that you would get less for
your electricity than it cost to generate it.  I could heat cheaper just by
buying off-peak electricity at 2.3 cents/kwh.  But perhaps in 5 years the
economics of this will have improved.  It might help to throw in an electric
bike except those are not practical during heating season.  My heating season
coincides with icy sidewalks.

Can the fuel cell be run backwards?

Re #3:

>This means having to sell back electricity to the grid even during cold
>weather when running the system enough to produce all the heat and hot water.

Only if you insist on generating more electricity than you need.  If you
need 8 KWH/day (333 watts average) of electricity (which is converted to
heat) plus another 6000 BTU/hr, the fuel cell would produce 1136 BTU/hr
waste heat from the electric requirement leaving 4864 BTU/hr.  If you
revved up the fuel cell by another 264 watts (total electric output 597
watts, extra waste heat 901 BTU/hr) and fed that to a heat pump with an EER
of 15 (264 watts in, 3960 BTU/hr out) you'd have another 4860 BTU/hr of
heat.  Your energy input would be 4072 BTU/hr for this output; not bad, eh?

Using a heat pump as a dump load gives you a lot of flexibility in the
heating season.  Heat pumps can heat DHW too.

>In November or March it might balance out, but during April-October I would
>still have to buy electricity from the grid (or invest in some expensive solar
>equipment).  

Or generate juice and throw the heat away, just like the plants which
make your juice now.  It's not a panacea, it's just a way to take a huge
whack out of annual energy needs without a bit of lifestyle change.

>People with air conditioners use more electricity during
>hot weather when nobody will be cogenerating electicity along with heat.

I did say it wasn't a panacea.  However, this problem would not be
exacerbated by such a system; the use of highly efficient ground-coupled
heat pumps for air conditioning might even reduce it somewhat.

>I think that Detroit Edison (or whatever it is called now) is paying such low
>prices for electricity that you sell to the grid that you would get less for
>your electricity than it cost to generate it.

Answer:  net metering.  I'm sure the electric company would be happy if
you produced a slight excess during peak hours and consumed a bit in the
off-peak; it would make their lives a lot easier and reduce their costs.
Or you could generate a substantial excess during the day and sell it
for 5 cents a KWH, and the electric company turns around and wheels it
over the lines to all the people charging their electric cars at the
office where they sell it for 10 cents a KWH.  Everybody's happy (except
the oil producers).

Re #4:  For most realistic purposes, no.

Heat pumps are not efficient at below-freezing temperatures so an air-to-air
heat pump would not make sense in this scenario, since it would only be
practical during the times of year when you don't need more heat than you
would get by generating enough electricity for your non-heating needs.
Ground-water heat pumps are rather expensive and tend to require that you have
a pond or a deep well, and they are also high maintenance, as we know from
talking to a couple of people who have them.

During hot weather I would not want to generate heat near the house in order
to produce electricity.  I suppose someone could rig up a system in which the
waste heat ran an air conditioning system.  In the summer solar power would
make a lot more sense than cogeneration.  

I don't see this as being really practical except during the spring and fall
when heating needs matched electrical needs more closely.

Does this fuel cell produce AC current (in the $1500 model)?

It might help to use hot-water heating so that excess heat at any moment could
be stored for later use.  And incorporate solar water heating in this somehow
as well.  

I think Detroit Edison is paying people about 3 cents/KWh.  I will try to find
out.  

                     S T A T E   O F   M I C H I G A N
               BEFORE THE MICHIGAN PUBLIC SERVICE COMMISSION
....   
   At the February 14, 2001 meeting of the Michigan Public Service
   Commission in Lansing, Michigan.
....
   ORDER APPROVING TARIFF RIDER

   On February 9, 2001, The Detroit Edison Company (Detroit Edison) filed
   an application for approval of its Standard Contract Rider DG relating
   to distributed generation. The rider establishes terms and conditions
   for customers operating less than 100 kilowatts of on-site distributed
   generation capacity to sell energy to Detroit Edison at the utility's
   top incremental cost of power. The rider also specifies
   interconnection standards and metering requirements and makes
   provision for the parallel operation of a customer's equipment with
   Detroit Edison's system. Because the application does not request an
   increase in any rate and will not result in an increase in the cost of
   service to customers, Detroit Edison requests that the Commission
.......

What is the 'top incremental cost of power'?

Re #7:

>Ground-water heat pumps are rather expensive and tend to require that
>you have a pond or a deep well, and they are also high maintenance, as
>we know from talking to a couple of people who have them.

There are designs for ground-side loops which merely run a length of
tubing under the yard, or vertically down a small-diameter hole.  It
is not necessary to pump water or even have a water supply for heat.

>During hot weather I would not want to generate heat near the house
>in order to produce electricity.

People using air conditioners are doing just that, so I doubt they would
have the same objections.  However, the economics are different if you
cannot take advantage of the waste heat.  (At worst, you are back to the
current model of buying juice generated somewhere else.)

>I suppose someone could rig up a system in which the waste heat ran an
>air conditioning system.  In the summer solar power would make a lot
>more sense than cogeneration.  

If you do not have a use for the waste heat, there is no "co-" in the
generation.  I expect that the capital cost of a heat-driven A/C system
would make it unpopular, despite how much I find it attractive from the
standpoint of cleanliness and elegance.

Something that was mentioned is that the fuel cell operates at about
800 C (~1470 F).  This temperature is high enough to operate a gas
turbine, and turn perhaps 20% of the waste heat into electricity in
large units.  If the fuel cell grabs 60% off the top and the gas
turbine converts 20% of the remainder, that's 68%.  Then you can run
a steam cycle off the gas-turbine exhaust...  It should be possible
to get 75% net efficiency out of very large installations.

>I don't see this as being really practical except during the spring
>and fall when heating needs matched electrical needs more closely.

Did you fail to grasp the point of the heat pump?  You could get as
much heat as desired by generating electricity to pump heat; the
limit on usefulness is not set by cold weather, but by warm weather.
If you need less heat, you generate less juice and turn off the heat
pump.  When you don't need as much heat as is generated as the byproduct
of your electric demand, you've got issues.

>Does this fuel cell produce AC current...

Electrochemical cells produce DC.  (Duh!)  DC can be inverted to AC.
Cheap inverters seem to be running ten cents a watt or less these days.
As the price of mass-market electronics continues to drop, I expect that
we'll see 5 KW sine-wave inverters for $300 or less if demand creates a
market for volume production.

>... (in the $1500 model)?

This is not yet a product and does not have a price tag attached.
The only thing that's currently known is the raw-materials cost,
given as $37/KW by the article.  Multipliers between that and price
of the theoretical product can be estimated, but no more.

>It might help to use hot-water heating so that excess heat at any
>moment could be stored for later use.  And incorporate solar water
>heating in this somehow as well.  

One of many opportunities for efficiency and price arbitrage, if the
markets develop far enough for homeowners to be players.

>I think Detroit Edison is paying people about 3 cents/KWh.  I will
>try to find out.  

Let's see.  If my fuel cost is $.017/KWH and I can turn that into
electricity at 50% efficiency, I could sell the electricity for
$.03/KWH and enjoy heat at an incremental cost of $.004/KWH.  So
long as I need heat anyway, it sounds like a great deal.

Re #8:

>What is the 'top incremental cost of power'?

The price of the highest-cost power that's actually required to meet
demand, I suppose.  That's how much the incremental kilowatt costs.
As demand falls, the utility stops using (or buying from) the highest-cost
generators first and the incremental cost of power drops.

The contract rider makes a lot of sense for Detroit Edison, because the
more sources they have for power the less they'll have to use the
highest-cost suppliers, and the less their power costs the more money
they can make at the same rates.  Also, the closer the power producers
are to the consumers, the less the load on their lines and the longer
they can go without expanding.  The only thing that would concern me
is the metering and interconnection requirements, which might be
prohibitively expensive for really small (less than 10 KW) producers.
Environmentally it may well be a win regardless - I'd have to learn
more to be sure.

I have some recent energy figures that I'll try to incorporate into
an analysis of fuel-cell possibilities sometime tomorrow.

I did some research today, and found that the average gas
consumption of a house with gas service is 83,000 cubic
feet per heating season.  68% of this goes for space heat
(presumably the rest is for cooking and clothes drying).

Based on my preliminary numbers, a fuel cell/heat pump/electric
car combination could eliminate motor-fuel consumption during
the heating season, produce a substantial amount of electricity
and fully replace the energy lost to heat with the heat pump.
The gross savings based on the DOE figures would be about $558
per year, and elimination of 245 gallons of fuel use based on
a fuel economy of 30 miles per gallon (probably pessimistic).

When I find figures for the total residential natural gas
consumption, I'll try to convert this to a net reduction in
national oil consumption.

I dug up a DOE figure claiming US residential usage of 4.940
quadrillion BTU of natural gas in 2001 (preliminary numbers).
If my figure of 245 gallons savings per 83 million BTU of gas
is accurate, use of the fuel cell/heat pump/electric car
scheme could save 14.6 billion gallons of gasoline per year.
This is without increasing natural gas consumption.

Cogeneration possibilities for commercial and industrial properties
would probably increase this figure by a substantial factor.

I have a different approach to saving fuel - reduce consumption.  My household
could not possibly save $558/year.  My estimate of total cost of electric heat
at 9 cents/KW would be under $200/year, and at time of day rate about
$50-75/year.  Another $50/year or so for hot water (electric).  No air
conditioning needed due to insulation and timed ventilation.  Even if a fuel
cell setup could reduce the cost of heat/electricity, it is unlikely to ever
pay for the cost of the equipment with this level of consumption.
Nobody seems to make small enough heating systems for a well insulated house.
Russ, what length tubing at what depth are people using for ground-heat pumps?
What is the smallest manufactured system designed to produce (BTUs)?  

We have not ruled out any type of heating system yet, but it would never pay
for itself to have natural gas coming to the house ($100/year to read the
meter).  The current idea is to splurge and put electric radiant heating in
some of the floors, plus a 1500 watt heater in the ductwork used for
ventilation and dehumidification, meaning you could switch over to any other
heat source and heat via hot air.  Anything with pumps tends to cost more to
start with and need maintenance.  

The $50 for hot water may be too high but it is based on 2.3 cents/kWH in the
winter and a high rate in the summer.  If it were worth the bother, this could
be coupled with a non-photovoltaic solar water heating system.  

Non-infloor radiant heat equipment is very cheap compared to heat pumps or
furnaces or boilers.  Radiators start at $25 for room sized and can be had
for free since many people are removing them.  We have a collection of used
and even unused free ones.  Kiwanis was throwing them out.  I spent $1000 on
exterior insulation instead of more than that on a furnace.  (Plus the added
cost of building thicker walls).  The fuel-cell approach makes more sense on
larger installations since the equipment probably does not cost ten times as
much for a building ten times the size.

Russ, can you find out what exactly DTE is paying homeowners with solar
systems for energy sold to the grid?  And what it costs to add to your solar
system the equipment needed to do this?

I think you would also need some way to store the electricity at times when
you only want heat, as well as a way to store the heat when you only want
electricity.  

Sindi, I don't think you quite realize that most people are living
in houses which can have their heating plant and water heater replaced
(and are designed for that), but can't be insulated to R-30 for love
nor anything but huge sums of money.  Nor are most people willing to
live with the discomforts that you are.

Most people expect their houses to be no colder than 65 in the winter
and no hotter than 78 in the summer, and they will not give up personal
vehicles absent an utter collapse of the economy.  I prefer to work with
people's desires, not against them.  Fortunately the gains which can be
achieved even within these constraints are enormous.

Jim has been adding insulation to his house, built in the late thirties, along
with a third layer of glass and lots of weatherstripping.  It finally, last
week, fell below 50 degrees with no heat added.  (Admittedly his housemate
bakes a lot and leaves lights on all over the place).  It is not all that
difficult to add insulation to the outside of a house and might not cost more
than you are proposing for a ground-water heat pump plus fuel-cell furnace.

Do you consider a tripling of the cost of gasoline (which would still make
it cheaper than in a lot of the world) an utter collapse of the economy?  If
trains again became cheaper than cars or airplanes I would expect them to
become the normal means of long distance transportation again (long distance
meaning the distance many people commute to work), like they are in most of
the world.  People will give up things that cost them money if there are other
cheaper ways to accomplish the same thing.

I am not saying that this fuel-cell cogeneration does not have its place, but
other factors should be taken into consideration.

Trains would not become much more useful until they build more stations
and there are better local transportation options at their termini. If
cars became much more expensive to operate, and whole transportation
infrastructure would have to be (re)created.

"People will give up things that cost them money if there are other
cheaper ways to accomplish the same thing" 


I think that this is one of your most out-of-touch-with-reality
assumptions.

One of the bases of 'affluenza' is that people are NOT acting that way.
Most of the excesses of consumerism are based on people behaving in the
opposite way.  

think "conspicuous consumption". 

A lot of train stations that already exist are not currently in use (such as
Dexter and Chelsea).  

They would  need new bathrooms (and other updates and upkeep). Also, a
LOT of people have filled in the spaces between the old passenger
stations. More would have to be added. Even then, many people would
not be able to walk from their homes to a station in a reasonable
time (or in bad weather) as conveniently as they can get in their car.

In big, densely populated, urban areas, trains are still widely used, and
stations are frequent enough, although even then one must have a car to
get to the train stations. I am reminded of this by staying with relatives
in Rye NY, and travelling into the city on the train.

I suspect it would be a lot cheaper to rebuild the old train routes and
stations than to keep building new highways.  There is already a Toledo to
Ann Arbor train route with no passenger service since 1950.   Public
transportation within cities to take people to train stations is also possible
and is the norm in most of the world.  (Chicago seems to be building its new
train stations on the theory that everyone travelling by train will arrive
there in a car.  We had to walk along a 4-lane highway in the mud to get to
the station a mile away because there were no sidewalks leading to any of the
businesses, all of which were behind huge parking lots).  In Boston the subway
system goes to the Amtrak stations and the airport.  It is not difficult to
arrange bus routes to include train stations.

Well....I was a passenger on the Ann Arbor to Toledo (and return) train
sometime in the 80's. It had to run at ca. 30+ mph because the trackbed
isn't in very good shape (and maybe the engine and cars too). Also, we
had to stop every now  and then for the photographers to get off and
take pictures of the train coming, or going, or stopping in a station
(which were pretty rare). I think it would take some doing to update
the tracks, stations and rolling stock, to be practical. 
But it was fun.

My grandparents insulated the first floor of their house to modern
standards.  It was time consuming and very expensive -- it basically
involved gutting the whole first floor and building new, thicker
interior walls, since the studs weren't thick enough to hold more
insulation.  Definately not an easy process.

I do not understand what interior walls have to do with exterior insulation.
To add exterior insulation, you can leave the original walls where they are
and build an additional non-load-bearing wall on the outside of them, or just
nail on sheets of styrofoam under the new siding.  Jim just added insulation
between his existing studs (they had none) which helped a lot.  His estimated
heat bill is three times mine for a house about the same size (which should
go down if he ever insulates the basement walls and floor, or at least the
basement ceiling).  He is in the process of switching to off-peak electric
heat at 2.3 cents/KWh.  (DTE has not gotten back to us for over a year about
this, but they did put in a new pole at the bottom of his yard so he could
safely upgrade from 60 amp service and he put in the trench and conduit).
If you have enough insulation, you can heat up the house before 10 am and it
should stay warm enough until 7 pm when the peak ends.  Or you can, for about
$300, buy an electric storage heater that puts the heat into a ceramic storage
medium overnight and then releases it as needed, but this takes some good
predicting of how much you will need (which the more expensive models try to
do for you based on outdoor temperatures).  These are common in Germany and
Scandinavia.  Or you can store the heat in the form of hot water.

Sorry about the long paragraph.

Since Jim does not mind a few degrees of temperature swing it will just be
baseboard heaters, which we already have for free.

Re the cost of refurbishing rail lines.  A friend was assessed $10,000 for
his share of paving an already existing city street.  Assuming his property
is about 100 feet wider, or less, and that the property across from him paid
for half of that particular 100 feet, paving appears to cost about one million
dollars per mile for already graded, already in use, city streets (two lane).
I suspect highway construction costs are much much higher than the costs of
upgrading existing railway lines or even building new ones along with
stations.
And they don't need frequent repaving.  The Ann Arbor Chicago train was going
85 mph and passing highway traffic but we were told it could have gone much
faster if the tracks were improved.  Modern trains can do 150 mph or so.


Any heating system based on natural gas would never pay for itself in my small
well insulated house unless DTE stopped charging $100 per year to read the
meter.  The electric bill includes a minimum $6/month usage charge but at
least you don't pay for meter reading plus minimum usage if you use less. 
In Boston things are reversed - no meter reading charge for gas but there is
one for electricity.  Maybe now that it is one company they will make things
more consistent.

I note that Sindi admits that increasing the depth of the the
insulation of a house involves re-covering the outside at the
very least (an option not available with brick structures, to
give just one example).  This is expensive and involved,
whereas replacement of the heating plant can be done in a day
or two.

And Sindi, if you must drift on topics such as rail transport,
please make your own item for it.

Grex crashed last time I tried to reply.  Russ, you really cannot control
drift in an item.  I was responding to Rane and I don't recall how the train
thing came up.

We put 2" styrofoam and then drywall on my bathroom basement wall.  No reason
why you cannot glue styrofoam on the outside of a brick wall and cover that
with some siding material such as cement board (screwed on through the foam,
or with foam between furring strips).  This would put a thermal mass inside
the insulation, making it easier to heat during offpeak hours and store heat,
or to keep a house cool on hot days after ventilating at night.  You would
then just add some trim around the windows.  

The cogeneration would make sense on a larger scale, maybe for neighborhoods.
In Belgrade heat is pumped to apartment buildings, possibly from a local
generating plant.  

What do you estimate to be the cost of a fuel cell or other cogeneration plant
plus a ground-heat pump even without the ductwork and fans needed to move the
heated air around?  Include control equipment.

Re #24:  If you think that someone is going to cover their expensive,
relatively classy, maintenance-free brick exterior with foam and
cement board, you need to have your head examined.  That would destroy
a large fraction of the value of the house right there.

If fuel cells make sense on the scale of an individual laptop computer,
they'll make even more sense on the scale of a furnace or water heater.
A lot of expense comes on a per-unit basis and doesn't vary much with
the size of the unit.

As for cost, I seem to recall that the production cost of an item falls
by 20% for every doubling of the number of units made.  If we assume
that materials cost $37/KW and production costs add another $200/KW
for the first 10,000 units, by the time you've built 10,000,000 units
your production costs are down to $27/KW, total $64/KW.  Ten million
units is a small fraction of the number of gas furnaces and water
heaters out there.

Most houses either have ventilation ducts or hot-water systems
for heat.  A pre-existing system is not an additional cost when
it comes time to upgrade or replace the heating system.  As for
the cost of heat pumps, I haven't priced them lately.

Last, water heaters could be a real gold mine for co-generation.  The
one downstairs serves 12 dwelling units and is rated at 199,990 BTU/hr
(call it an even 60 KW thermal).  Recovery is rated at 181.5 gallons/hr.
If you replaced the gas burner with a 20 KW-electric gas-fired fuel
cell and a 40 KW-thermal heat pump with a CoP of 3.0, you'd be burning
gas at a rate of only 40 KW while putting 60 KW of heat into the water
and generating a surplus of 6.7 KW.  As demand fell to 40 KW, you could
either throttle back the heat pump and increase the electric surplus
to 13.3 KW, turn off the fuel cell and consume 13.3 KW from another
source, or anything between; at 20 KW heat demand, you could put the
entire 20 KW electric on the grid.

300 gallons/day would be 1.7 hours of operation; call it 102 KWH thermal.
If I can sell 102 KWH/day at 3 cents/KWH, that's $3/day or $1095/year; at
$64/KW, my fuel cell costs $1280 and pays for itself in 14 months.  If I get
clever and buy off-peak electricity at 2.3 cents/KWH for the heat pump at
night (preparing for morning showers), then heat the tank up during the
afternoon using fuel cell waste heat and sell surplus juice for 10
cents/KWH, my gain increases to $4.71 per day or over $1700/year.  My gas
cost in this last scenario would be 87 cents a day or $316/year.  I could
pay for a $5000 water heater pretty handily this way, and sell "green"
feeling to my tenants.  Those things would sell like hotcakes.

Is the expense of fuel cells mainly an issue of production cost, or
material cost?  I thought they needed some fairly exotic materials.  If
that's the case, the cost may not drop much with volume.

Oh, one other question -- who is going to pay you $0.10/kwh for your
surplus electricity?  In a lot of areas electricity doesn't even sell
for that much, and the power company is going to buy it back for a lot
less than they sell it to you for.

Russ was talking about selling surplus electricity for 3 cents/KWh - you must
have misread him somewhere.

I agree with Russ that his setup might make sense for apartment buildings or
other heavy users, but possibly not for small houses that don't use much
energy to start with.

Stucco is also considered a classy building material.  You can instead of
putting on cement board (which you decorate with thin strips to look Tudor)
spray on a three-coat system like what we applied by trowel, which is usually
put over styrofoam already.  It does not resist dents well but it looks nice
for a few years, anyway.  Probably longer if you don't build next to a truck
loading area.  The NEW center was built that way.  

Re #26:  In the case of room-temperature fuel cells, the catalysts
can be very expensive (platinum).  The solid-oxide fuel cell runs at
temperatures high enough to crack molecules apart with no assistance,
so it appears to get by with nothing much more exotic than nickel.
The article (read the link) implies that the zirconia is the costly stuff.

Re #27:  At least one California utility will, if you have a time-of-day
meter and generate the power during peak hours.  This was reported with
some apparent glee in a recent issue of _Home Power_; you are credited
for power (net metered) at the current time-of-day rate, so you can
generate power during the afternoon and sell it dear, then charge your
electric car cheaply in the dead of night.  (Actually, I think they'll
pay up to $.25/KWH...)

Re #28:  If you use almost nothing you'd probably be best off buying
from one of your neighbors who has an excess to sell, and in a world
of abundant natural gas (so far) and cheap efficient fuel cells (coming
soon), that is probably what a lot of people would have.  (You really
wouldn't pay your neighbor, you'd just pay the electric company and
let them sort things out.)

Re #29: Interesting.  With the push for building more power plants in
California, though, I don't expect prices like that to last.

I found a few graphs of costs of ground-water heat pumps in horizontal closed
loops.  You need at least 10' spacing between the loops, which should be 3-6'
down.  At 6' ground temperature is always at least 45 degrees (in the US,
anyway).  20' spacing is best.  They did not say how many feet you needed for
what size installation but it is in multiples of 500' loops.  You can get
three runs of 45' each into my back yard or about 150' total.  THe costs of
for horizontal loops but you can also drip 150-450' deep holes and use those
instead, at greater installation costs (but possibly warmer ground water?).

Ground loop costs - about $2000 for 3-5 ton units.  How many tons does it take
to heat a small house?  They calculated for AC only.  We don't need AC. 
Probably there is no unit small enough for our needs.

Heat pump only - 2 ton $1600, 6 ton $3600, other tons intermediate.

Interior installation without ductwork (assuming you have that already);
3 ton - $4500, 4 ton $5500, 5 ton $6000
Ductwork an additional $1500-2000 installed.
These figures do include the heat pump itself, I think.

Total installed cost of a 3-ton unit $4400-9000.  Not sure what this figure
refers to.

Cost of the smallest unit (2 ton) with interior installation and underground
piping is thus $2000 plus $4500 - $6500.

These units can last up to 20 years, meaning you have to replace the pump part
of them after a while.  Let's say another $2000 within my lifetime for
replacement if labor is cheap, or $8500 minimum for an oversized unit.

There is some maintenance involved -cleaning filters and coils.

We can put in radiant infloor heating for under $5000.  This saves at least
$3500 on equipment costs.  Spread that out over 35 years (assuming I live to
age 87) and it saves you $100/year not to install groundwater heating.

Calculated cost of radiant electrical heating under $100/year.  Calculated
cost with geothermal heat pump if it costs half as much, $50 year.  If costs
of electricity double, the system could pay for itself within my lifetime.

Oops, forgot the $1500 for ductwork if we don't do our own.  Add $50/year to
additional costs of the heat pump.  

Conclusion:  for small well insulated houses the most energy efficient system
may never pay for itself as the installation costs will be not much less than
for a larger poorly insulated house.  

The people we know with groundwater heat were using a river or a pond, which
saves on installation costs but the water may be colder. 

It'd be sort of silly to make a heat pump that *didn't* include A/C,
since all the necessary parts are already there.

We have an air conditioner of the type designed with a waist, so you can close
the window onto the middle of if and have the condenser on the outside.  We
are thinking of building a frame around it and putting it into the doorway
between the house and the solar porch, condenser side inwards, and using it
as an air to air heat pump when the porch is below 60 degrees.  Will an air
conditioner run under these conditions?  We might need to turn it off
frequentlly to melt any ice that forms.  Another option is a dehumidifier run
backwards but they are harder to separate parts on.  They are now making Sears
models that stop when they ice up and wait until the ice melts.  We could then
extract more heat from the porches on semisunny days or late afternoons. 

Re #31:  A ton is 12,000 BTU/hr.  (Literally, it is the rate of heat
transfer required to melt one ton of ice per day.  Melting ice requires
144 BTU per pound, so a ton is 288,000 BTU/day or 12,000 BTU/hr.)

A 50,000 BTU/hr furnace would be about 4 tons.  A 2 ton unit should be
adequate for a small, well-insulated structure.

Let's see, just offhand, consider a 2000 ft^2 2-story house.  It has
1000 ft^2 of attic floor insulated to R-30, 2280 square feet of wall
insulated to R-15, and 120 square feet of windows insulated to R-3.
Total U-value:  1000/30 + 2280 / 15 + 120 / 3 = 225 BTU/hr/F.  A
12,000 BTU/hr heater could maintain this house at 53 degrees F above
ambient; 24,000 BTU/hr could maintain it at 107 degrees above ambient.
That's 70 F in -37 F conditions.  I think that would do for Ann Arbor.
Someone using straw-bale construction for R-34 walls, similar treatment
for the ceiling and better windows might be able to get by with 1 ton.
Lessee:  1000 / 40 + 2280 / 34 + 120 / 4 = 122 BTU/hr/F.  Marginal in
the coldest weather, but nothing will freeze.

Oh, for what it's worth:  cost of deep-well heat tranfer loops would
all but certainly come down in price the more common they became.
Everything gets easier and cheaper with experience, and the materials
aren't exactly rare or terribly expensive (copper tubing).

Note:  R-value is in the units of square feet-fahrenheit hours per BTU.
It is the inverse of U-value (heat conductivity), which is in the
units of BTU per square foot per degree fahrenheit per hour.  Divide
square feet by R-value, you get BTU per hour per degree F.

I doubt very much that an air conditioner will function well with
a cold-side temperature too much colder than room temperature; the
compressor is designed for a certain vapor density and superheat, and
if things are too cold it just won't have much to move.  The worst
possibility is that it could get slugs of liquid, or accumulate so
much liquid in the evaporator that the compressor gets oil-starved.

Your biggest headache may be controls.  Trying to pump the *hot*
side up to a given temperature isn't part of the design.

If you don't care about losing the A/C to an experiment, go ahead
and try it; it cannot be less efficient than a resistance heater.

We have no other use for an air conditioner, and we have several air
conditioners and keep finding more working ones, so all we have to lose is
our time.  
We would just set the thing to run continuously whenever it was above maybe
40 on the porch, during the cold season.  I doubt it would overheat the house
in Dec through March.  So we would not be pumping the hot side to a given
temperature.  We would need a way to have it turn off whenever it ices up,
which might be most of the time.

Thanks Rane.  Your calculations are consistent with ours.  I don't consider
2000 square feet to be a small house, but that seems to be what is being built
nowadays.  A really cheap and easy way to cut heat consumption in half would
be to build smaller houses again.

Our house would need 6000 BTU to bring it from 0 to 60 degrees assuming no
heat coming from the sun into the house and nobody cooking or using lights.
So that is 1/2 ton.  I bet nobody makes a groundwater heat pump under 2 tons,
which is the smallest I found listed.  

I think it is the well drilling which is expensive as you need to go up to
450 feet deep.  Our house is actually in a good location for groundwater heat
as it is between Spring and Fountain St. and we hit water five feet down while
digging the foundation on one side, and again while digging under the sidewalk
to put in city water pipe.  The systems work a lot more efficiently if the
pipe is actually underwater.  

Another way to heat the house is to use as much electricity as the typical
American household - leave all the lights on, bake instead of pressure cook,
run a large color TV, etc.  Might not need any added heat then (of course it
would be rather hot in the summer to live like this).  

THese groundwater systems seem to be more popular in the south.  THe example
I ran across was for air conditioning in Louisiana.

You don't need straw bales to get R-34.  We have more than that from 9"
fiberglass in a staggered stud wall.  R-21 plus R-13 (or even R-15 for 3.5")
plus a few more R for a shiny vapor barrier and wood sheathing.  
The house is staying at 40 or above with the heat from one large freezer, one
small refrigerator that hardly runs, and one space heater keeping a 50 sq foot
room at 50 degrees.  On sunny days it gets to 46.  So we would not need to
raise the indoor temperature much to make it livable.  With good insulation,
lower air temperature feel comfortable because you don't have cold walls and
floor and ceilings.

I forgot to mention:  "total installed cost" is the number that you
write on the check that you give to the installer just before s/he
leaves after installing your new whatever.  It includes hardware and
labor.

Response not possible - You must register and login before posting.

- Backtalk version 1.3.30 - Copyright 1996-2006, Jan Wolter and Steve Weiss