Methods for the Study of Energy: the Case Study

I do not count myself as a top down thinker. Top down thinkers spend a great deal of time thinking about their methods before the start working on problems. During my father's scientific career his approach to an assignment was to always do a literature review first. Once he completed the literature review he had identified what was known about the subject, how new knowledge could be acquired, and obstacles, if any to acquiring that knowledge. His approach can best be illustrated by the assignment he received in the mid-1950's to report of the compatibility of plutonium with liquid fluoride salts. People in the small community of interest that focuses on the Liquid Fluoride Thorium Reactor recognize the importance of the assignment my father was given, to the rest of my readers I will only say that my father's answer to the question may have important implications for the future of the world's energy.

When my father did his literature review he discovered a significant obstacle to his research. His primary research tool, the glovebox, was defective. During the 1950's gloveboxes were used by AEC facilities both to conduct plutonium research, and to machine plutonium for nuclear weapons. But there had accidents including fires with plutonium gloveboxes at AEC facilities, My father did not like the idea of working with unsafe tools, so hew set out to perfect the glovebox. In short he found solutions to the problem of designing and building safe gloveboxes. His glovebox techniques quite literally were text book. In fact he wrote the glovebox chapter in a manual on physical chemistry techniques.

Once my father solved the glovebox problem he proceeded to answer the plutonium question. That is a top down approach.

Where my method diverges from that of my father is that when I start looking at a question I google it, and then see what I come up with. Once i get an answer in hand i start analyzing it. Then on the basis of my analysis, I formulate a question, which I Google again. I then do another analysis. Then I look for comparable cases and start the process.

I often do case studies. I have a number of ongoing case studies which I conduct on the future potential for wind generated electricity. One of my most useful case studies is based on the simple question, "can wind provide the power that will run my Dallas, Texas air conditioner during the summer?" My answer has been, in a single word, no! And if you live in Texas and a power plant cannot provide electricity to run air conditioners during the summer, it just can't cut the mustard. Well in Texas wind can't cut the mustard during peak hours of summer electrical demand. I did a case study to find if this was a local problem in Texas. It is not, indeed it turns out that there are similar problems with summer wind in California, the Southeast, New England, the Great Plains, New England and Canada.

It has been argued by Sanford University researchers that by linking many carefully selected wind generating sites the wind can be made reliable enough to be considered base power. The Stanford study found that by linking windmills at 17 Southern Great Plains locations, 21% of their rated power was reliable enough to qualify as base power 79% of the time. Unfortunately, this approach does not solve the Summer wind problem. There were several problems with the Stanford study. It did not address the Summer electricity issue. The study briefly noted a rapid drop drop off of wind availability after the 79% threshold, but did not say when. However, enough data is available about the wind performances of the 17 locations to get an idea, and clearly there is going to be a problem. If we looked at the idea of linking the 17 locations as a means of providing summer peak electricity the whole project would be a non-starter. Summer wind generated electricity in texas is not simply unreliable, it is largely unavailable during periods of summer peek electrical demand.

Renewables advocates have an answer to the summer wind problem, build solar generating facilities to handle peek electrical demand. There are some simple but obvious problems. First electrical demand remains high during summer evenings in Texas. Temperatures may remain above 100 F at 10 PM, and wind speed in the Southern Great Planes does not return to annual average as soon as the sun goes down on hot summer days. So it looks like we are going to have an evening shortage of peek electricity. There would also be another problem with the solar peek approach - its cost. My own review of the cost of solar thermal generating facilities suggest that the current cost is at least $4 billion pre nameplate GW output. But we are not talking about facilities that will be built today, inflation makes cost a moving target. During the next decade when such facilities are likely to be built, inflation is likely to drive their costs to $8 billion or even higher per name plate GW. Now this is an interesting figure. because it is the current cost of nuclear power plants, being bandied about as too expensive by nuclear critics is also in the $8 billion or above range.

Now, lets look what the southwest base wind system will cost. Renewables advocate Dr. Ben Sovacool recently put the figure of $1700 per nameplate KW in play in discussions with me. That figure is probably low. I have reason to believe that the cost of a fully installed windmill in November 2008 is closer to $2500 per name plate KW, but the lower figure will serve to illustrate my point. If we assume that our project to replace Texas fossil fuel generating plants with renewables by 2030, as the Google plan would require, how much is it going to cost? Lets assume that we decide to go with a all renewables system, with wind base power. Assume that the same rate of inflation for electrical generating facilities that we have seen during the last 5 years. That would bring our wind facilities capital costs to $3400 per nameplate KW. But remember that only 21% of nameplate capacity can be counted as base load electricity. In order to figure the cost of building base load electricity we have to divide the cost of a KW of of wind generating capacity by 21%. That gives a figure of something over $16,000 per KW. But hay, that is not the end of our cost, since our base load electricity cannot be relied on during summer days, we are going to need back up solar facilities. We have already counted that costs as $8000 per KW during the next decade. That gives us a cost of $24,000 per KW of semi-reliable wind and solar generated electricity. Semi-reliable because we know that there will be after dark hours of high electrical demand when our wind system will not be able to supply electrical demand. So far we have a system that is not 24 hours a day reliable. How much will it cost to give us some assurance that we can keep those Texas air conditioners running 24 hours a day? We could use sodium-sulfur batteries @ $350 per KWh capacity. 4 hours of battery back up brings out price to $25,400 for each 24 hour a day KW provided to Texas by a renewable system. Needless to say renewables advocates have not and will not perform this exercise, but the it does illustrate the value of case studies for exposing future energy costs.

My texas case study did not come from a top down approach. Rather it came from asking a simple question, how will the electric company get the power to run my air conditioner on hot summer days and nights. Looking at that question over time, has produced some answers, but raised more questions. There are other cases that could be studied.