Lots of modern cars have trip computers that include things like “miles to empty”. I got mine to read zero once, on the freeway, after having driven many miles trying to maintain as constant a speed as I could (I wasn’t trying to test it, I just wanted to get to my exit, where I knew there was a gas station, rather than get off somewhere before that and hunt for one). I drove probably three or four miles with the “miles to empty” reading zero. Not surprising, really, that they’d fudge it down a bit, for people like me.
I can’t add to the fuel gauge question, but I wanted to point out that half is always half, but that doesn’t mean all halfs are the same.
Door to door from work to home is 20 miles (I’m making this up). The first 10 miles are open highway with a 60 mph speed limit that I cover in 20 minutes. The last 10 miles are city blocks with a stop sign on every corner that take me 30 minutes to cover. When am I halfway home?
The same idea applies to everything else. In the gas tank, half can be volume or height or something else.
Now you are telling us the spedometer doesn’t work or are you that guy with a line of traffic behind you?
OK, I’ll revise my earlier statements to:
Car makers are stoopid!
Heh. I can guarantee you a gas gauge isn’t a precision instrument.
I drive a 1994 Cutlass Ciera. The amount of gas in the tank changes with the speed of the car, whether you’re driving uphill or down, when you brake…I’ve seen it bounce from the half mark up to the full mark and down to the empty mark in one trip.
Why do you say that?
Do you often run out of gas? Do your friends often run out of gas? If the answer to these questions is no then I submit that the accuracy of the gauge is close enough.
Could a car maker make a more accurate gauge? Sure they could, but how much accuracy is good enough?
Compare your fuel gauge to building a house. The plans for the house call for the studs to set 16" on center. Do I have to get a giant micrometer to set those studs to within + - 0.001"? Of course not, the marks on a tape measure are plenty accurate enough. The same is true with your gas gauge. Is it necessary to measure the volume of the tank down to the nearest milliliter? No, plus or minus a half or a full gallon is plenty good enough.
Compounding the problem for the fuel gauge designer at the car maker is you insist on driving up and down hills, and around corners all of which make the apparent fuel level change.
It’s not hard to measure the volume of a known container. There are computers in cars today running everything else, why not have the height-gauging floater translated into gallons (or liters). Hell, you wouldn’t even need a processor, just adjust the placement of the marks on the dial.
It’s just another example of poor human-interface design. Ask Cecil why headlights don’t turn off with the rest of the car…
On many makes they do. Some people like that they don’t have to worry about accidentally leaving the lights on. Then again, others don’t like that if they want to turn the headlights on after the car is parked, they can’t simply do it with the light switch.
Having a fuel gauge that’s consistently accurate throughout its range isn’t something that can be easily achieved just by throwing sophisticated electronics at the problem. Even with a tank that has an uncomplicated symmetrical shape, there are issues of sloshing and being off level to deal with. There are variations in float density, float placement, and potentiometer resistance, both from the manufacturing process and from changes with aging and use. Throw in the fact that an ideal shape for this purpose is at odds with other considerations (exhaust routing, trunk room, etc.) and it gets even harder. These are physical obstacles. They can be overcome, but it would be so costly that the price wouldn’t be justified.
People get along just fine with the gauge systems now in use. While the indication for the first 3/4 or so can be fairly coarse, they often have a finer reading for the last 1/4, so drivers can pretty well figure how close they’re getting to empty.* How many would be willing to pay an extra $1000 for a really accurate gauge? No, I don’t know what it would cost. But I do know that it wouldn’t be cheap, and that there’s no real need for it, and that there’s no significant demand for it. It’s not a matter of poor design - it’s a matter of adequate and cost-effective design.
*This varies from car to car - some run you out of gas when the needle hits the edge of the “empty” mark, others may go a needle width past the mark. You can find out where “true empty” is for your car by running out of gas in it. But the main thing is, gauges tend to give easily discernible indication of the progress toward empty during that last 1/4.
On every car I’ve ever had, I’ve always topped off. This really means that I’m overfilling the tank. If you don’t top off, you’ll see that the needle stays on “F” a lot less time than otherwise.
As for headlights, that’s an old article by Cecil. There are good systems for automatic headlight control. Heck, I remember them working with full manual override control as far back at a mid 80’s loaded Dodge Lancer turbo.
I was always under the impression that most car makers left a little ‘safety margin’ in the fuel gauge calibration - i.e. set it so that you could still go a dozen or so miles after it hit empty, just to prevent damn fool customers stranding themselves by misunderestimating the distance to the next garage.
This seems a good thread to ask whether I have been wrong all these years…