On August 2, I published an open wager on National Review Online. I offered to bet up to ten people $10,000 each that I could take my 2007 Chevy Cobalt, which is not a flex-fuel car, and, running it on 100 percent methanol, get at least 24 miles per gallon on the highway. Since methanol averages less than half the price of gasoline — and can readily be made from coal, natural gas, or any kind of biomass without exception — this would demonstrate superior transportation economy from a non-petroleum fuel that is producible from plentiful American resources.
Unfortunately, no one took the bet. That fact alone says a lot. Of the 7 billion people on this planet, there are about a million or so who know a great deal about cars. Clearly, not one of them was sufficiently doubtful that it could be done to put his money on the line. Although it left me short a nice chunk of easy cash, the refusal of anyone to accept my challenge should have settled the matter. But some people, while refusing to take the bet, still demanded that I conduct the test anyway. I did, and here are the results.
First, I ran the car on 100 percent methanol. This required replacing the fuel-pump seal made of Viton, which is not methanol compatible, with one made of Buna-N, which is. The new part cost 41 cents, retail. In order to take proper advantage of methanol’s very high octane rating (about 109), I advanced the timing appropriately. This dramatically improved the motor efficiency and allowed the ordinarily sedate sedan to perform with a significantly more sporty spirit. As measured on the dyno, horsepower increased 10 percent. With these modifications complete, I took my Cobalt out for a road test. The result: 24.6 miles per gallon.
When I first made the bet, many commentators thought that I would aim for high-efficiency performance with high-octane fuel by increasing the compression ratio of the engine (which is how race-car drivers using methanol have done it for the past half-century). However, with modern cars using electronic fuel injection, this is unnecessary. Instead, the necessary changes to the engine can be made simply by adjusting the Engine Control Unit software. Thus, except for switching the fuel-pump seal as noted above, no physical changes to the car were required.
Other critics commented that while I might be able to achieve good fuel economy, the idea was impractical because the emissions would not be acceptable. In response, I had the car tested for emissions with 100 percent methanol (M100), 60 percent methanol (M60), and ordinary gasoline (i.e., E10, which contains about 10 percent ethanol), and for comparison, did mileage tests for these alternatives as well. The results of all these tests are shown in the table below.
It can be seen that, far from failing to meet emissions standards, the Cobalt running on methanol was extremely clean, beating both the strict Colorado emissions standards and the national EPA averages by an order of magnitude. The complete elimination of carbon-monoxide emissions when using M60 is particularly remarkable — so much so that I initially thought it was an experimental error caused by faulty equipment at the emissions test station. I tested it again at a different station and got the same result.
Returning to the subject of fuel economy, this can be evaluated by dividing the miles per gallon by the pre-tax spot price of the fuels in question in order to obtain the pre-tax miles per dollar shown in the table above. It can be seen that when methanol is used, fuel-economy improvements of 40 percent can be achieved. (The spot price shown in the table is the New York Harbor spot price of gasoline and the non-discounted Methanex spot price, both averaged over the past year.)