Hacker News Reader

Comment by leeoniya | Hacker News Reader

50MW seems pretty significant. if Wikipedia is to be believed, "According to Google their global data center operation electrical power ranges between 500 and 681 megawatts"

View on HN

It's less significant if you consider that the Capacity Factor of a wind farm is around 20-40% (wikipedia).

That comes out to more than 3 dollars per watt.

You say that as though it's a lot. And it sort of is a lot for an initial investment. But the cost of keeping a wind farm running is a lot lower than the cost of keeping fuel consuming power plants running. This wikipedia page has a bunch of estimates of overall cost per Megawatt Hour that show wind being a slightly cheaper source of energy than conventional coal plants, and considerably cheaper than more advanced coal plants.

http://en.wikipedia.org/wiki/Cost_of_electricity_by_source

Wind farms, as with solar farms and nuclear energy, have large up-front fixed costs and negligible marginal production costs. If the wind-farm produced _only_ 1 hours worth of energy, then $3.00 per watt would be spot on, as I understand it.

Likely the wind-farm will operate a number of years. Let's suppose it produces at 10% capacity for 20 years. That would be equivalent to 100% capacity for 2 years, which would be 50 MW/h * 24 hour/days * 365 days/year * 2 years = 876,000 MW/h

So now you have $50 Mill/ 876,000 = $57.08 per MW/h, or $5.7088*10^-5 cents per watt/hour.

Now, suppose the capacity is larger or the time lasts longer. The cents continue to fall per watt/hour

Ach, just realized they invested $75 Million, not $50 Million. Scale up by 50% :D

MW is the wrong unit to use for that comparision. TWh/year is more appropriate.

Out of curiosity, why do people use Watt-hours instead of Joules for this purpose?

Which is more practical?

A: "Hi, can you please supply us with 50MW for the next 12 hours?"

B: "Hi, can you please supply us with 2.16Tj at a rate of 50MW?"

It would be much nicer to get 2.16Tj in a bucket.

Tradition. Same reason we use centralized banks and sovereign currency instead of bit-coin and other independent currencies, or miles per gallon instead of gallons per mile.

Or for that matter, kilometres per litre.

Joules (or watts) is a measure of capacity. Electricity generating capacity, especially with intermittent sources like wind is almost meaningless. Who cares about capacity if the actual generation is zero.

Watt-hours, Megawatt-hours, and terawatt-hours are used because that's the descriptive term of energy actually produced.

I think MW is a reasonable unit. It's the power output of the wind farm.

No. MW-hours is the reasonable unit of power output. MW is a description of the capacity, but it says nothing of actual generation.

I understand the number is an ideal, maximum number, but I'm talking about the units. Are you saying power should be reported in megawatt-hours per year? That might be convenient, but it's the same as megawatt-years per year, which reduces to megawatts which is what I said.

MW tells us the scale of the plant. TWh/year tells us the production of the plant.

I've got a good gut sense for what a 3 MW wind turbine looks like, so when I read "50 MW Plant" - I can already envision about how large, and about how many turbines it will have (approx 15-30).

TWh/year doesn't tell me as much, because the plant might be located in a very windy place. Or maybe not in a windy place at all - the data is still useful, but I prefer MW when people are talking about a plant's "physicality".

Reply on news.ycombinator.com