Does gasoline burn at different rates in my car's gastank?

April 7, 2007 9:35 AM

In three cars I've driven, I get about 300 miles per tank of gas. But with all three cars, I can get up to 200 miles by the time I hit 1/2 tank. But invariably, by the time I near the 300 mile mark, I'm on E. Am I nuts or do gas seem to burn faster in the lower 1/2 of my tank? Is this mental or by design?

I understand it could be mental, or even physical, as lots of cars can log 50 miles by the time they make it down to the F line (if topped off), but it seems weird.

Is the distance between topped off tank and 1/2 tank mark really 2/3 of the distance between 1/2 tank and the empty red zone?

posted by mathowie to travel & transportation (22 comments total)

It's the float and a function of how it works

This mechanism (the float) is one reason for the inaccuracy of fuel gauges. You may have noticed how your gauge tends to stay on full for quite a while after filling up. When your tank is full, the float is at its maximum raised position -- its upward movement is limited either by the rod it's connected to or by the top of the tank. This means that the float is submerged, and it won't start to sink until the fuel level drops to almost the bottom of the float. The needle on the gauge won't start to move until the float starts to sink.

from here
posted by fair_game at 9:43 AM on April 7, 2007

Do you drive until you run out of gas? That would really be the only way to find out how many miles you go in the last half tank of gas.

Also, is your gas gauge to your left when you are looking at it? If so, then when it appears to you that it is at half, it may really be at less than half because of the angle from which you are viewing it.
posted by flarbuse at 9:44 AM on April 7, 2007

Here is the How Stuff Works page that explains fuel gauges which are, in their words "notoriously inaccurate" and they explain why. The two main reasons: shape of fuel tanks and the way the little float thingamajig in the tank sends information to the needle on your gauge.
posted by jessamyn at 9:48 AM on April 7, 2007

Your fuel gauge doesn't measure the volume of fuel, it measures the fuel of your tank. You can imagine that the shape of the tank could cause the effect you notice.

Or, you know, what everyone else said.
posted by sexymofo at 9:49 AM on April 7, 2007

IME, gas gauges on cars are notoriously inaccurate. One explanation I've heard for this is that on many cars, the fuel guage is designed to pop an early empty to protect the fuel pump (which is cooled by gas in the tank). The fine print on some car manuals admit that empty warnings are signaled with 50-100 miles of range to make it to the next exit with services.
posted by KirkJobSluder at 9:51 AM on April 7, 2007

My car has a digital read out of miles before the tank is empty. Do they just take my average MPG and calculate it into how much the needle estimates? It seems to give very precise, if not accurate, readings. I noticed on a recent low-gas trip it read 36 miles, I coasted for some time and it ran up to 37 miles, indicating it was taking some very recent MPG estimates.
posted by geoff. at 10:00 AM on April 7, 2007

Do you drive until you run out of gas? That would really be the only way to find out how many miles you go in the last half tank of gas.

Not really -- just get the tank capacity out of the owner's manual and then subtract out what it took to fill it back up. Just sayin.
posted by rolypolyman at 10:01 AM on April 7, 2007

You're not crazy, I noticed this too.
posted by Brandon Blatcher at 10:02 AM on April 7, 2007

I get the same thing. I always assumed it's down to the shape of the tank, making the first "half" bigger. I've always wondered why they didn't have a cam or something to correct for that. I've never run out of gas intentionally, but I do know the tank's capacity and how close I get when I fill up, so it's not just a matter of refilling early.
posted by adamrice at 10:13 AM on April 7, 2007

1. You could build a very accurate meter, it would also cost quite a bit.

2. It doesn't need to be that accurate, as long as empty is somewhere near empty, but not quite, because everybody 'knows' that you can drive the car past empty, and I've known too many people who've done just that. Give them an accurate gauge, and many more people would be stuck without gas at the side of the road.

3. The 'mileage' calculators on cars just take a floating average of your mpg (you can get immediate mpg on any car with an ODB II connector on it btw, which is every car built after uhh... '94 or '95) and a tank level, which is the same cheap level that everybody uses. I've driven 'under' the miles left (not my fault, somebody else left the tank empty, and nearest gas station was 30 miles away!) and I didn't run out of gas, YMMV of course. :)
posted by defcom1 at 10:23 AM on April 7, 2007

The fuel gauge in my van changes according to whether I'm going up or downhill. Also, it's a 20 gallon tank. When I've filled it with the needle very nearly reading empty, I can't get more than about 15-16 gallons in it. I suspect vehicle manufacturers deliberately calibrate their already inaccurate fuel gauges to under-read near empty. Given their inaccuracy, this is, of course, the best way to engineer it.
posted by normy at 10:46 AM on April 7, 2007

In my little coupe, I get at least 40 miles after the low fuel light comes on. (I've never been brave enough to push it farther than that!)
posted by chrisamiller at 10:57 AM on April 7, 2007

1. You could build a very accurate meter, it would also cost quite a bit.

I've never understood this. Why can't they just calibrate the display differently?? Like, put "1/2" where "3/4" is now, and on down the scale. So it moves inconsistently over the course of the tank; that shouldn't be any big deal. Or are they very inconsistent from tank to tank? It seems like it's not SO different from tank to tank that a re-scaled meter would be worse than the current setup. And it would be such an easy fix!
posted by rkent at 11:17 AM on April 7, 2007

I get at least 40 miles after the low fuel light comes on.
I get 60. Don't ask how I know that.
posted by Opposite George at 12:56 PM on April 7, 2007

Actually, most modern cars with a body computer of some sort do exactly what rkent suggests - have a table to linearize the output of the fuel level sensor. They also calibrate the amount of fuel remaining at a display of empty.
posted by rfs at 1:09 PM on April 7, 2007

Also remember that there's more leeway at the bottom of the gauge than the indicator typically gives credit for. Clarkson's 800-mile trip to Edinburgh and back without re-fuelling shows that even a digital sensor on a high-end car will tell white lies about an empty tank -- especially in a diesel, where running dry has far rougher consequences than a petrol vehicle.

I do wonder if there are regional differences in the calibration: it's seems less typical in Europe (or anywhere with expensive fuel) for drivers to fill the tank to the brim.
posted by holgate at 1:48 PM on April 7, 2007

do gas seem to burn faster in the lower 1/2 of my tank?

If you could accurately measure the mid-point of your fuel supply, in theory the second half should give you slightly better mileage than the first, because your vehicle is that much lighter. This assumes all other factors are equal -- for example, you're driving around a track at a constant speed and constant atmospheric conditions.
posted by beagle at 2:26 PM on April 7, 2007

Not really -- just get the tank capacity out of the owner's manual and then subtract out what it took to fill it back up. Just sayin.

Except you are also filling the plumbing to and from the tank, and the owner's manual may not be completely accurate.
posted by Kwantsar at 4:57 PM on April 7, 2007

Oooh! thanks for asking this, I'd been wondering about it -- my Corolla does the same thing. I thought it was just Toyota protecting me from myself!
posted by bitter-girl.com at 9:37 PM on April 7, 2007

My 20 year old BMW has a non linear set of measurements on the fuel indicator. That is to say, it has markers for 10L, 20L, 30L and 70L. The needle barely moves until it hits about 40L, then you can almost watch it move down.
posted by tomble at 2:57 AM on April 10, 2007

I'm not sure if this is an urban legend or not, but I heard that a car company (or car companies in general) developed very accurate fuel gauges that gave a much more accurate reading, but then had to change back to the older, less accurate gauges due to public demand. Basically, the innaccurate one drops slowly for the first half a tank then fast for the second half, which (at first) makes you think you're getting better fuel economy than you really are. The more accurate gauge tells the truth and moves at a steady speed, thus moving faster than the innaccurate one at first, giving the impression the car is getting a lower mpg. So they keep the gauges innaccurate because people like the (false) impression they get of better fuel economy.
posted by EndsOfInvention at 4:04 AM on April 11, 2007

...but then had to change back to the older, less accurate gauges due to public demand.

The public demand being "hey suddenly we are getting worse fuel economy! wtf!", not "hey, I want an innaccurate fuel gauge!".
posted by EndsOfInvention at 4:06 AM on April 11, 2007