Monday, July 27, 2009

Running on empty

On Saturday driving my 2004 Toyota Solara back from Denver I drove all day on one tank of gas. I covered 520.4 miles. I thought at the time that this was a personal record for distance driven on one tank but was surprised to find upon checking that in the first couple of years after buying my previous car (a 1986 Toyota Celica) in 1994 I had exceeded 500 miles on a tank on several occasions and in August 1995 had gone 523.4 miles.

The Celica got better mileage but had a smaller gas tank than the Solara so its range was similar. The Solara has an average mpg calculator which I reset each time I buy gas. It generally overestimates mpg by .5 or so (compared to dividing mileage by amount of gas required at the next fill up) but on Saturday was dead on estimating 34.0 mpg (as compared to 34.01 computed from the 15.300 gallons to fill up). The Celica required 13.472 gallons to fill (after the 523.4 mile run) for an estimated mpg of 38.86. Both cars have a reserve (if the specs in the owners manual for fuel tank capacity are to be believed) past the point where the gauge reads empty so the true maximum range is a bit higher.

Of course to be really precise you need to account for the fact that the Celica's odometer read about 2.4% high while the Solara's reads about 2.4% low which would mean the Solara holds the true miles record.

One problem for electric cars is that there is no prospect of their achieving a range of anywhere close to 500 miles anytime soon. And they take hours to recharge as opposed to a few minutes to refill a gas tank.


  1. So that begs several questions:
    1) Are you really keeping records dating back to 1995 about mileages and gas station visits? That is indeed admirable!
    2) How did you measure the 2.4% mileage bias down to such precision?
    3) perhaps a system should be devised that involves standardized, easily replacable batteries so that one could just swap batteries at a station onstead of waiting for hours to charge.
    4) electric cars should be equipped with a loudspeaker producing noises to immitate the good old combustion engines, and (optionally for some of us) artificial gasoline smell generators.

  2. I am in the habit of recording the odometer reading, amount purchased and price whenever I buy gas. I use a little notebook which I keep in the glove compartment. I started the current notebook when I purchased the Celica in 1994 and it isn't full yet. This practice is fairly pointless but it does occasionally come in handy.

    I divert myself on long drives by comparing the odometer mileage to the mileage from the mileposts on the interstate. If the odometer reads to tenths of a mile 100 miles is enough (in theory) to measure the bias to a tenth of a percent. In practice of course there are potential problems such as errors in the milepost locations. However I am confident that the Solara odometer is reading low and the Celica odometer read high.

    There have been proposals for swappable battery packs for electric cars. Possibly this is the way to go. However it seems unlikely to be as convenient as buying gas. And the limited range would still be a problem.

  3. That is a remarkable routine!

    The odometer inaccuracy may be caused by variabilities in tire/wheel diameter: wear, tire pressure, manufacturing variation, etc. the 2.4% error could just correspond to a 2.4% larger/smaller diameter, which on a 16" tire would correspond to about 3/8".

  4. It looks as though you won't be eligible for the "Cash for Clunkers" program any time soon.