[Gasification] cost per kW, cost per kWh - was: Re: Ash to land? - separated ash fractions
dmc at danielchisholm.com
Sat May 12 10:08:05 CDT 2007
Sorry, long email follows.... once I get started spinning numbers, it
seems difficult to stop...!
Hopefully we can develop some sort of "budget" for what a viable biomass
power system can or should cost. There's nothing wrong working on
$20,000/kW lab gear, so long as we know that we must (for example)
somehow produce $1000/kW gear for commercial use.
On Fri, 2007-11-05 at 15:09 -0700, Tom Miles wrote:
> Thanks Oscar and Daniel,
> Applying Oscar's cost structure:
> Diesel genset $1152/kW
250kW-1000kW Cat diesel gensets cost $200-$300/kW.
(I see that AJH wrote, "Except even a prime time diesel genset in the
Megawatt class shouldn't cost more than GBP200/kW." So we seem to be in
the same ballpark)
With fuel today in a $20/MMBTU ballpark (diesel fuel at $2.75/gal, 70
cents per litre) and 33% thermal efficiency, your fuel costs are about
20 cents per kWh. (see near the end of this email for the scenario
under $170/bbl crude).
If you can get natural gas for $10/MMBTU and keep your plant efficiency
at 33%, then your fuel costs are 10 cents per kWh. Heavy fuel oil
(Bunker C, #6) is also in the ~$10/MMBTU ballpark (OK, perhaps closer to
Which is why a gasifier system at $3600/kW seemed shockingly expensive
to me. It seems like it is simply too high a price to pay, for the
ability to use a $3/MMBTU fuel.
> It sounds like a target system cost should be $1800/kWe or less. Prototypes
> have been costing $5,000-$40,000/kW.
Let me take a crack at coming up with a budget for a biomass fueled
electricity generating gasifier system. Let's assume three possible
- selling wholesale electricity at 5c/kWh (i.e. you're competing with a
conventional power company)
- selling or displacing "retail" electricity at 10c/kWh
- selling premium-priced "green" electricity at 15c/kWh
Assume biomass at $3/MMBTU delivered (that's $30/tonne 45%mc)
If your plant has a net 20% efficiency of biomass->electricity, then
your fuel cost is $15/MMBTU-electricity, which is 5 cents per kWh. Ugh!
So right off the bat, you're not going to be able to produce wholesale
electricity. You're going to need a plant that is 30% to 40% efficient
before it even makes sense to run any numbers. Even at that, it would
probably be easier to make a go of it if you had a cogen application
that could earn you *something* from some of the waste heat.
So let's look at the 10c/kWh and 15c/kWh markets. You have 5c/kWh and
10c/kWh respectively to spend on capital, operations and maintenance.
Since this is an exercise in capital pricing, let's ignore O&M for now.
Perhaps we can justify this by saying that in a 250-1000kW plant you can
probably achieve greater than 20% efficiency, so let your actual
efficiency there pay for your O&M, and perhaps some profitability too.
Yes, profitability would be nice.
Assuming very durable industrial equipment with an 85% availablity rate
(let's say it has a 10 year lifetime, so you can amortize the cost over
75,000 hours), and assuming that you are blessed with relatively cheap
capital at 8%, every thousand dollars of capital equipment will cost you
2.4 cents per kWh (let's call it 2.5c/kWh for convenience).
So, for selling power at 10c/kWh, your plant must absolutely not cost
you more than $2000/kW (5c/kWh for fuel, O&M and profit; 5c/kWh for
interest and amortization).
And for 15c/kWh, your plant must absolutely not cost you more than
$4000/kW (5c/kWh for fuel, O&M and profit; 10c/kWh for interest and
And if you want to even bother running calcs for selling competitively
priced 5c/kWh power, then your plant better not cost more than $1000/kW.
This puts your capital cost at an uncomfortable 2.5c/kWh, which is half
your selling price. And if you were able to achieve a very impressive
40% plant thermal efficiency, thats a fuel cost of 2.5c/kWh. Which adds
up to your net selling price. If you're going to survive, you better
have a market for your waste heat....
How's this for a depressing calculation: what is the cost of a simple
oil change for a 250kW plant? Let's say you change your oil every 250
hours, and you need 20 litres (5 USgal) at $2/litre. Throw in a $10
filter, and pay your mechanic $20 for the labour. That's a $70 oil
change (of course I'm being generous here; how much does a 4L/1-gal oil
change for your car cost?). 250kW*250hrs = 62500kWh are produced in
this period. This is more than 0.1 cents per kWh, just in lube oil
Even worse is operator/attendant labour. What if you have a 4x250kW
plant, with one full time operator? Let's say your loaded labour rate
is $15/hr (and good luck with that, eh? ;-). You are paying him 1500
cents per hour, and his plant is generating 1000kW. That is 1.5 cents
per kWh of output (!).
Certainly from one vantage point, 1MW is a "big" system. But from the
point of view of running a business that pays for itself, 1MW of power
is only worth $50, $100 or $150 per hour (at 5, 10 and 15c/kWh). You
can kinda-sorta-just-barely run a business at this revenue rate, but you
better stay sharp...
Don't even think about a what a 100kW plant ($5-$15/hr of electricity)
looks like, argh...!
I would SWAG that to produce "Grid-ish" (i.e. 5-15c/kWh) power, anything
under 1MW pretty much has to be unmanned and fully automated. Which is
an understandably big challenge, and therefore perhaps not the first
approach to try.
OK then, what about premium-value power? For example, remote power or
emergency backup power is certainly worth paying $1/kWh, and perhaps
even ten to a hundred times that depending on the circumstances (ask me
just how much, when it is in the middle of the winter here! ;-). But at
these rates, you've got to keep in mind that you could always use a
diesel genset that costs $200/kW capital and 20c/kWh fuel.
Even if the price of oil goes "sky high", you'll still be able to afford
to buy diesel fuel to produce $1/kWh power:
* $57/bbl crude (roughly where we are today) means:
-- $10/MMBTU heavy fuel oil
-- $20/MMBTU diesel ($2.75/USgal, 70cents/litre)
-- 20c/kWh-electric, via a 33% efficient genset
* $170/bbl crude (is this "sky-high" enough?) means
-- $60/MMBTU diesel fuel - $8.25/USgal, $2.10/litre
-- 60c/kWh of diesel genset electricity (not so bad if you're in the
market for $1/kWh power)
The problem of course is that a biomass power system fits poorly here,
where the goal is to produce expensive power for only some of the time.
A biomass system is, just like a solar, wind, hydro, tidal, OTEC, waste
heat ORC, nuclear or coal plant, a "cheap-fuel, expensive-plant"
solution. They only end up paying for themselves if you can put them
into a baseload-like operation. You must use that cheap fuel to pay for
that expensive plant.
For non-baseload sales, i.e. where you can get paid 25c, 50c or even $1
per kWh, they will never be competitive with "cheap-plant,
expensive-fuel" solutions (gas or liquid fueled, spark engines, diesel
engines or gas turbines).
I really don't want this email to be a negative diatribe. Are there any
lessons to be had? Here's what I can think of, perhaps others will add
- can't go small. A 100kW plant produces only $5-$15/hr of electricity,
no matter how you slice it that can't pay for much of anything.
- overly high fuel cost plants won't ever work
- overly high capital cost plants won't ever work either
- you must play capital cost off against fuel cost:
-- 2.5c/kWh per $1000/kW is (baseload, 85% capacity factor)
-- 2.5c/kWh per $3/MMBTU fuel cost at 40% plant efficiency
-- 5.0c/kWh per $3/MMBTU fuel cost at 20% plant efficiency
What then can/should we do if we are to make money (or do good, if you
prefer) with biomass gasification? Here are my thoughts on the facts we
- biomass is a relatively cheap fuel
- it's relatively clean
- it's relatively costly to handle/process
- but we need a reasonable capital cost
- we need to find a product to produce that is:
-- valuable enough to pay for our cheap fuel
-- valuable enough to pay for our reasonable capital cost
-- well fitted to our fuel's cleanliness
-- well fitted to our fuel's difficulty of handling
That deserves an email of its own, eh? I do have an idea for one
approach that I'd like to throw out for consideration, but in the
meanwhile I hope that discussion of this email proves worthwhile, then
we can get on to that...
Fredericton, NB Canada
More information about the Gasification