Hi guys. Please allow me to add a couple of comments. The first is I've been told numerous times that dash readings for fuel mileage can be totally inaccurate. As with temperature gauges, they're best used to notice when something unusual is going on, and not actual values. That's why some dash gauges don't even have numbers
To calculate the fuel mileage, software is involved that looks at road speed and "injector pulse width". "Pulse-width" is the percentage of time an injector is switched on during the intake stroke. That's programmed in at the factory as a starting point, then the Engine Computer makes small adjustments according to readings from the front, or upstream, oxygen sensor. Part of those adjustments are done instantly as you're driving. Those are called the "short-term fuel trim, (STFT) adjustments. Those can be read on a scanner, and are part of what is used to calculate the instant fuel mileage. When the computer sees it is always making the same corrections for the same driving conditions, it makes an adjustment to the "long-term fuel trims", (LTFT). From then on, it uses those as the starting point instead of what was programmed in at the factory. That means it needs to make fewer and smaller fuel trim corrections as you're driving. To calculate the instant fuel mileage, the computer looks at the current injector pulse-width to calculate how much fuel is going into the engine. For the average fuel mileage, something gets added in as one of the variables, then there might be some limit that was programmed in. Typically, that limit is miles, as in the last 10 miles or perhaps as much as the last 50 miles, but could also be since the readout was reset by you. To say that more clearly, the average fuel mileage may not be what you got on the last cross-country trip. It may be what you got in just the last 50 miles. That's where the vehicle model and year is another variable.
Manufacturers make software updates available all the time to address a whole bunch of variables and fixes, as they become known and necessary. It could be as simple as injectors were installed from a new supplier and they flow different volumes of fuel than what the software is expecting. That would be a fuel mileage reading issue, not an engine performance issue.
Another variable we usually forget about is excessively high fuel pressure. A plugged pressure relief valve or blocked fuel return line, (depending on the car model, year, and engine size), can cause that. On many models fuel pressure is adjusted constantly according to engine vacuum, (a measure of engine load), and is used to prevent an excessively rich coast-down condition. Fuel pressure is one of the variables the Engine Computer uses to calculate injector pulse-width. On a few models, they actually measure fuel pressure, but on most, it is supposed to be known values, so that is programmed into the software, then there's no way to correct for it if it's wrong, other than to watch the oxygen sensor readings. Based on those readings, the computer can only adjust fuel metering roughly plus or minus ten percent. If the needs go beyond those limits, the engine will get too much or too little fuel. Eventually the computer will see that, set a fault code, and turn on the Check Engine light. Since that light isn't on now, this isn't the cause of the poor fuel mileage, or not all of the conditions have been met that are needed to set that code.
To get a better picture of the actual fuel mileage, once the engine is warmed up and you're cruising at a steady speed, reset the reading, then it will start fresh with the current fuel mileage. The numbers will update rather quickly at first giving you the ability to see what affects it. That could be struggling to get up a hill, fighting a head wind, excessive speed changes like for curves or city driving, etc. I also have two friends who are constantly on the gas, off the gas, on the gas, off the gas, at 55 mph. They aren't aware of it, but the rest of us feel it.
Depending on what the dealer looked for already, they can read the fuel trim numbers on a scanner to see if the Engine Computer is requesting additional fuel or if it is requesting a reduction in fuel, (shorter pulse-width), but not achieving that. Live oxygen sensor readings would show that as running too rich. If that sensor reports it's running too lean, the computer will increase pulse-width to put more fuel in during each injector pulse.
One thing I haven't seen mentioned yet is dragging brakes. To identify that, stop on a slight incline, shift to neutral, release the brakes, and see if the car creeps downhill on its own. If it does not, Ken or I can describe how to pursue the diagnosis if you want to do it on your own. This used to be pretty common on older Chrysler products. They had a metal bracket crimped around the middle of the front rubber flex hoses. Over time, rust built up inside that crimp and constricted the hose. You could push brake fluid through that restriction, but then it couldn't release on its own. That kept that brake dragging, and that usually got worse as the brake heated up, causing the brake fluid to heat up and expand. That applied the brake even harder. This was a time issue, not a mileage issue. It was a simple fix to open that crimp up a little, often without even removing the wheel. Many import vehicles use the same type of bracket in the middle of the hose, but they are usually round barrels, or tubes, instead of a bent strip of metal. For those, the only solution is to replace the hose.
Another thing to consider is whether or not you use the cruise control, and this can affect fuel mileage differently on different car models. The perfect example is what I found with my 2014 Grand Caravan on a 200-mile road trip. The average fuel mileage reached 31.9 mpg and was still rising, but I ran out of highway. The instant and overall mileage dropped a lot when I tried to use the cruise control because the highway had a lot of hills. The transmission down-shifted a few times to keep the speed up, and it down-shifted again a few times going downhill to hold the speed down. The cruise control made the engine work unnecessarily hard to maintain speed going uphill. Instead, I followed a string of large trucks in the slow lane so no one behind me would be upset with my speed changes. I let the speed drop off going uphill, and it was free to pick up speed on the other side going down. Overall, I pretty much kept up with traffic for over 200 miles.
I've never come close to the mileage around home. I found the average reading will drop a tenth of a mile with just one minute of idling at the end of the driveway or when stopped in traffic. It also drops one or two tenths just getting back up to speed after a stop sign. I suspect there's a conspiracy here. They put up more and more stop lights to make me waste gas so they can collect more gas tax!
Now that the temperature has dropped to the 40s and 50s, the average milage has dropped noticeably. I park this van for the winter so I don't know what those temperatures would do., But it did surprise me how much that little change affected fuel mileage.
As Ken Suggested, I would calculate fuel mileage manually, but over a half dozen tanks of gas. That will mitigate the variables, mainly how much gas you stuff in at each fill-up. I'd be willing to bet you'll find the mileage is higher than what the dash gauge is showing.
Wednesday, November 19th, 2025 AT 3:30 PM
