Is there a larger gas tank form another car that will fit into my 69 cutlass ? It seems to me that a 17 gallon tank is just not quite big enough to go on long a cruise and not hav to worrying about filling up!! Gas mileage just ain't that great!! Lol!

 

Most cars back in the day were designed to go 250-300 hwy miles per tank. Whats your MPG?

 

Just to be clear, he's talking Imperial gallons. The capacity is 20 U.S. gallons. I doubt a larger tank would fit under there or be as safe.

Of the 20 gallon capacity, probably about 18 are usable. If the car gets 15 mpg combined city/highway, that's a 270 mile range, which, as oldcutlass points out, is about what you'd expect from these cars back then.

 

It's only about 9 mpg so that's about 150 miles . I just rebuilt the motor and it pushing 390 hp from a 350 bb.

 

I learned back in school that all 20 of mine were useable. When it was on E, the dang gauge would rise up to the 1/8 mark after sitting in the sun for a while. I'd get back in after a day in class and think I had enough gas to get home but ended up on the side of the road!

How did you make that?

 

$$$$$

 

I question this. Where is the fuel pickup pipe? In these cars it generally extends down from the top of the tank and ends about 1/2-inch above the bottom of the tank. It can't go completely to the bottom as there needs to room for the sock filter plus room for the gas itself to get to the pipe. This means that, once the level of fuel drops below the bottom of the pipe, the car will run out of gas, even though there is a half-inch of fuel still sloshing around the bottom of the tank. Depending of the area of the bottom of the tank, there could be up to a couple of gallons of fuel still in there when the car sputters to a stop due to lack of fuel.

At least, this was the situation in my '67 Delta 88 when I removed the tank to replace the sending unit and made some measurements when the sending unit was off the tank. I'm guessing its similar for the A-body cars. The area of the bottom of the tank might be smaller, so perhaps it's more like a 1 to 1.5 gallons of fuel stiill in there when the fuel level drops below 1/2-inch, but I don't think any fuel tank runs 100% dry when the car "runs out of gas." That would mean the fuel pickup pipe would be in the bottom of the tank and susceptible to clogging given the crud that invariably builds up in a gas tank over years of use, and that would be a bad design.

 

The gas tanks in these cars are 20 gallons, period. There is no 17 gallon tank made, despite what RockAuto may show in their catalog. We discussed this exact issue in a prior thread. I've always been able to get a full 20 gallons out of an A-body tank with no issues.

As for a larger tank, no one makes one. First, it would hang down below the frame rails, which is asking for trouble, and second, there is no manufacturer who wants to accept that legal liability. The factory tank configuration and size was tested against the federal requirements in place at the time the cars were built. Any change to that design that isn't tested against similar standards is just asking for a multi-million dollar lawsuit. Your only option is to add a racing fuel cell in the trunk.

 

Well, run your '68-'72 Cutlass out of gas, then fill it up and see how many gallons it will take before the gas nozzle shuts off. Mine took 20 gallons on multiple occasions back in the '80s.

 

Have to agree with Joe. I fell for the 17 gallon "myth" for my Spectre tank based on how much I could put in it when the gauge was getting close to E. In reality, I think the sender is fairly inaccurate and there is still a fair amount of fuel in the tank even at E.

Kind of like the sales literature that says SBO headers fit a Cutlass S but not a Supreme....

 

This makes sense to me. I figured out back when I ran the tank dry a couple times that it was ~ 15 gallons from E to F (~2 gallons per each 1/8 mark) and there was ~2 gallons left below E and about the same above F. I verified this a couple times by knowing the mileage and how far I drove after the gauge read E before the engine stopped. I was the epitome of a poor college kid.

 

Understand that the fuel gauge just isn't that accurate. Typically if you completely fill the tank, the needle goes past FULL and stays there until you drive about 50 miles or so. The sender isn't linear over the F-to-E range either. The indicated level drops slowly at the beginning of the range, speeds up around half a tank, then slows again towards E. The needle goes past E and there are usually a couple of gallons left with the needle on E. Also keep in mind that the float on the sending unit is supported by a thin wire arm, which can get distorted during installation.

 

It's entirely possible to design a fuel sending unit/gauge combination that has a linear response. But the question is would people even want it? People are accustomed to fuel gauges that go beyond full and stay there for a while, and a fuel gauge that goes below E before you run out of gas. I don't think that people would like having a car that is full when it is on the F, or runs dry when the needle is right on the E.

 

No question, but my point is that the design of the sender that Olds used is far from linear. Ignore the fact that it goes past the E and F levels and just consider the fact that it uses a rotary pot to sense a linear level movement. When the arm is near horizontal (about half tank) the resistance change is pretty linear with level. As the arm rotates up or down, you're on a sine curve and resistance change is non-linear.

 

Agreed.

winding a rheostat with accurate "taper" isn't really that big of a problem. there are other things that can really screw up linearity. such as:

1. the gauge itself has non-linear response. so the rheostat has to be wound in a non-linear fashion that's complimentary to the gauge's non-linearity to make the system usable.

2. tank shape irregularity. even if you had a linear sender and gauge, tanks have irregular shapes that aren't perfectly symmetrical. tank shape irregularity will prevent a linear change in fluid level, which is what the sender looks at. so even if you had a perfectly linear rheostat, the irregular shape of the tank keeps the sender/tank combination from having a linear response. when you're designing a 3 part system (tank, sender, gauge) you have to design all three to work together. then if you change one of the parts things can go all to hell.

3. incomplete sweep of the sending unit. the motion of the sending unit actually traces a "chord" across a circle. most sending units have a float arm that is too short, which results in the "stuck above F" and "goes below E" problems. fixing this requires physically modifying the sending unit.

I don't disagree with your point at all. I was trying to ask the question about whether people would even want a linear system, after so many years of dealing with non-linear ones. if someone were willing to give people a perfectly linearized fuel sender/gauge system, would they like it or hate it?

I'm thinking that if someone sold something like that, that the people who bought it would be unhappy because it's too accurate. People don't like to run out of gas when the needle is right on "E". They've been conditioned to be able to drive below E, and if they had a really accurate gauge, they probably wouldn't like it.

 

That's a real good point. (Warning, thread about to be taken well off topic...)

I've got a newer daily driver that has a trip computer that shows miles to empty, as well as instantaneous and rolling average MPG. I'm ASSuming that the computer calculates miles to empty by taking some recent average MPG (it definitely isn't the long-term average that's shown on the computer) and divides that into the tank level reading from the tank sender. The result is that the miles to go varies widely, based not only on the recent average MPG but also the non-linearity in the level sender. I also assume that the instantaneous MPG reading on the computer comes totally separately by integrating injector flow rate and on-time and dividing by MPH. It's interesting to watch how these data points vary so much relative to each other. (Hey, when you're stuck in traffic for 90 min each way, you have to amuse yourself!).

 

On mine and most all other fuel gauges, the needle will float around with the fuel sloshing in the tank. My take is when it stops moving around its time to put fuel in. Pretty simple philosophy. The newer cars meet your criteria Bob, the tank the sender and the gauge work in harmony and are much more accurate to the old systems. Our old systems are 1 step over slipping a stick in the tank to measure the level.

 

All good points. What I should have said is I found my tank took 15-16 gallons to get from E to F and I knew there was ~2 gallons left when it hit E, so I could determine how many miles I could drive when the needle was on either F or E.

 

I think the non-linearity of the tank/sender/gauge became a thing of the past with the advent of in-car computers. I had an on-board trip computer in my BMW in the 1990s and it was dead-on accurate in determining the number of miles to empty. All that's required for this sort of thing to work is to create a calibration curve to compensate for any non-linearities in the tank shape and sending unit's response. I think that just about any of today's cars with a trip computer should have an accurate fuel gauge.

When asking about whether people would like or dislike accurate gauges, I was thinking about the classic car population, not modern cars.

 

That is decidedly not the case with my 2002 daily driver, as noted above.

 

Joe, could it be that your driving habits are what's non-linear???

 

That's a given.

I tell people that all my cars have a digital throttle (it's either OFF or ON).