Jim Thompson (jthompson@taii.com)

Week of 8 Feb 10

Click here, then click "Trans lator" in top navigation bar for other languages.

Some will consider this column to be very negative and indeed it may be--but the following needs to be said.

There is the old story about the fellow that quit the US Patent Office in the 1790's because he thought everything had been invented. Your author does not feel that way, even today. However, some of you may think he does unless you follow the logic in this column very closely.

If a teenager asked me today if it would be wise to spend their university years as a student studying how to make energy usage more efficient, I would have to very carefully consider their question and probably tell them, "No, you cannot have a full and complete career studying and perfecting energy efficiency improvements." Carefully notice I did not talk about a career studying energy extraction methods, but limited myself to energy efficiency improvements.

Why do I say this? When it comes to energy efficiency improvements there are hard physics-based limits that are easily defined. These limits are two and only two. The first one is the conversion of one form of energy into another: the limit is 100% perfect conversion in an ideal world (in effect a perpetual motion machine, a claim for which the US Patent Office will honor only with a working model). The other limit is the opposite of the first: this is the limit of insulation, which in a perfect world would allow zero energy exchange between two surfaces or two volumes.

We know that both of these limits, in even a world without financial constraints, are impossible. In a world where one must trade capital expenditures for operating expenditures, in other words, the real world, the constraints are much less than 100%.

This is much different than trying to set a speed record or even see the famous Moore's Law of computing power march forward. Those limits do (in the case of speed) and seem (in the case of Moore's Law) to have no hard boundaries. Granted, Einstein told us the absolute limit on speed is the speed of light*, however, that is so distant from anything we can do, for all practical matters we can think of speed as being limitless.

Yes, energy exchange and non-exchange (insulation) have very finite boundaries. They are similar to the basic studies of limits in high school mathematics. I remember our high school math teacher, Mr. Shaffer, positing the problem of the frog jumping towards a well. This frog had a unique jumping technique: with each jump he could jump half the distance from himself to the well. The question was, does he ever get to the well? In an ideal world, he does not, of course, but does approach it asymptotically.

Energy efficiency research is the same. Whether seeking 100% conversion or -100% conversion (perfect insulation) there are finite limits. Probably in an ideal world those limits are around 90%; in a world where one has to pay capital to achieve the best limits possible, they may be even lower, perhaps 80%. Who knows for certain where these are, but the point is we will reach limits in both cases.

So let's posit for a moment that best available technology says we are at +/-50% in our energy conversion expertise (the positive being conversion from one form to another, the negative representing insulation). We'll set the rules slightly different than the frog's jumping for argument's sake. Let us say we can improve by 5% the first year (2010) and each year after that we only improve by 1/2 of what we did the year before. Hence, in 2011, we will be at +/- 55%, in 2012 we will be at +/-57.5% and so forth. The argument goes asymptotic in 2031 at 60%. Don't like these numbers? Start with +/-70% now and an improvement of 5% the first year, hold all other conditions constant. The model goes asymptotic at +/-80% in 2031, probably the practical limit.

If we do use the frog model (improving by half the remainder each year) and start at +/-50%, we'll go asymptotic at 99% by 2016. I don't think we are that good.

The poorest form of energy conversion of which I am aware is solar cells (one might, in the strictest sense, call this extractive technology). Sharp, of Japan, apparently has the currently highest efficiency cell at 35.8%. Yet, if you set the improvements at only 2% for the first year from this base and assume you can march forward at a rate of improvement of 75% of that of the year before, the model still goes asymptotic at 43.8% by 2034.

Of course, all of this is just an interesting mathematical exercise, but it does give us a framework and a dilemma box for our current scientists and engineers studying energy conversion. For if they say they are making improvements faster than this, they are obviously going to run out of work sooner; slower, and we may ask, " What is delaying your progress?"

The point is, in 20 to (let's give it an extra ten years for uncertainty) 30 years, we are going to be at the end of practicality in energy conversion and insulation technology. We'll call all of this Thompson's Law, just for the fun of it.

So, again, what do I tell the high school senior thinking about studying energy conversion technology? You might want to think about having to prepare for another career to jump to about the time you are 50. This gig will be over.

For safety this week, it is appropriate to talk about hypothermia (a human body too cool) and hyperthermia (a human body too hot) for a moment. Both are medical emergencies. If your safety meetings have not covered these recently, I recommend you do so.

Be safe and we'll talk next week.


*Let's chase a rabbit for a minute: I am told if you build a pair of scissors with blades a couple of miles (3 km or so) long and close that pair of scissors at a rate of 1 rpm, when the closure gets to be about a mile from the point of rotation, that closure point will be traveling faster than the speed of light. However, that point has no mass and exists only theoretically, so it is possible. By the way, there is a certain technical manager in Wisconsin that may want to check this calculation and report back to us next week.