1999 Dodge Caravan Gas Mileage Off

  • 6 CYL
  • AWD
  • 170,000 MILES
3.8L V-6 Grand Caravan ES AWD owned since new and always maintained and oil changed religiously. Runs like a top, clean exhaust, etc. And yet the fuel mileage has gradually declined by about 3+ mpg since new (was good for an easy 16/24, now about 13/19). I burn no oil between oil changes and power is good, so don't think engine wear is a factor. Transmission rebuilt 17K miles ago and still shifting nicely with no slippage. Filters are clean. Plugs replaced about 35K miles ago. Injector body decarbonized per recommendation of dealer to no affect.

Dealer says this is normal for higher mileage vehicles. Maybe for my 1965 Chevy with worn out cylinders, but I'm skeptical in this case. What are the other lilkely culprits if the cylinders are assumed to have little wear and considering it runs perfectly? The exhaust is still all original and I sometimes wonder how the cat is doing, but so far not affecting the way the car runs.
Do you
have the same problem?
Friday, July 23rd, 2010 AT 1:41 AM

1 Reply

Hi muthaiga. Welcome to the forum. The best way to approach this is to find a mechanic willing to take the time to connect a scanner to read the long and short-term fuel trim values. I'm not sure about aftermarket scanners but I know the Chrysler DRB3 will display the numbers.

The Engine Computer starts out commanding fuel metering based on values programmed in at the factory in "lookup tables" that cover every combination of sensor readings. From that starting point, once the oxygen sensors are warmed up, they report on unburned oxygen in the exhaust stream. Basically, the computer uses that information to add or subtract fuel from the pre-programmed values to meet the desired conditions right now. That's the short-term fuel trim numbers. If they are positive, the computer is requesting extra fuel. Negative numbers means the computer wants slightly less fuel. These numbers are constantly changing and updating as you drive. After a while, when the computer sees it is always making the same adjustment to fuel metering, it will move those numbers into the long-term memory. Those numbers are what will be used in the future to run the engine instead of the factory pre-programmed numbers. The long-term numbers can only modify fuel metering by plus or minus around 10 percent from the factory numbers, and the short-term fuel trims can only modify the fuel delivery plus or minus 10 percent around the long-term numbers. The key to all of this is those oxygen sensors.

When viewing the fuel trim values, if they are very high positive, you have to determine if the extra fuel is needed for a proper mixture or if the computer is responding to inaccurate sensor data. A good engine performance mechanic will force the mixture to go lean by introducing a vacuum leak, and will force the system to go rich by injecting propane, or by removing and plugging the vacuum hose to the fuel pressure regulator to make pressure and the resultant flow through the injectors increase. Then he will watch the short-term fuel trim numbers to see if the computer compensates correctly. The scanner will also display "oxygen sensor cross counts". That is how many times per second the sensor switches between "lean", "rich", and "centered". While a centered mixture is ideal for best engine performance and lowest emissions, the switching between too lean and too rich is necessary for proper operation of the catalytic converter.

Some other things the mechanic will look for is a cylinder misfire. If spark to one cylinder is lost intermittently, there will be unburned fuel and oxygen in the exhaust, but O2 sensors only respond to the unburned oxygen. The computer will interpret that as running too lean and will command more fuel, in your case, to the three cylinders on that bank. No matter how much extra fuel might be requested, there will always be that unburned oxygen in the exhaust.

If the fuel trim numbers are high negative, the computer has seen the rich mixture, (actually the lack of lean cycles since it can only monitor oxygen in the exhaust), and is trying to subtract fuel. A restricted air filter will cause you to have to press the gas pedal just a little further than normal. The throttle position sensor will tell the computer to expect the engine to need more fuel which it delivers, then it subtracts the extra fuel. Another less common problem is high fuel pressure. Too much pressure results in too much fuel flowing through the injectors during each pulse. Unlike the very trouble-prone pressure regulators on GM trucks, regulator problems on Chrysler vehicles is practically unheard of. However, a loose or cracked vacuum hose going from the intake manifold to the regulator will cause pressure to go up. Manifold vacuum pulls the fuel from the injector. When vacuum goes down as in moderate acceleration, fuel pressure is increased to give more push to the fuel to maintain a steady flow. If pressure is too high and too much fuel sprays from the injectors, the computer reduces the injector on-time to reduce the amount of fuel, but remember, it can only reduce fuel by around 10 percent. That might not be enough modification to compensate for the high pressure.

One rare problem that can be hard to find is a tiny exhaust leak ahead of the catalytic converter. Between the pulses of exhaust flow, the momentum creates small pulses of vacuum. That very small vacuum can draw in outside air through the leak and by noticed by the O2 sensor. That will cause the computer to think the mixture is too lean and it will command more fuel. This is a very rare problem so don't go tearing the exhaust system apart until the more likely causes are investigated.

Another overlooked problem is dragging brakes, particularly the front disc brake calipers. Years ago there was a lot of trouble with sticking pistons in the calipers, but they don't cause so many problems now. Stop on a small incline, put the transmission in neutral, and see if the van rolls downhill on its own. If it does not, a caliper is sticking. One cause that is becoming more common is a constricted hydraulic hose due to rust buildup inside the mounting bracket in the center of that hose. Usually there are other symptoms such as a higher and harder than normal brake pedal and most people notice the over heating brake, but that's after the problem gets pretty bad.

At the mileage you listed, you can also suspect a worn / stretched timing chain. That will retard camshaft timing and reduce engine power slightly. You won't notice the drop in power but you will be pressing the gas pedal a little further than normal to make up the difference. The front engine cover will have to be removed to inspect the chain to see how loose it is.

One final thing to keep in mind is the different fuel formulations required in different parts of the country at different times of the year. Thank your hair-brained legislators for that. One person in California posted a question about a year ago relating to getting 16 mpg with their truck since they bought it new many years earlier. It wasn't until they traveled to a different state that they noticed the fuel mileage went up to more than 20 mpg. In southern Wisconsin, it is common knowledge that the reformulated fuel is to be avoided whenever possible. Besides making people sick, everyone knows it causes a drop in mileage.

Since you noticed a gradual drop in mileage over a period of time, my vote is for a stretched timing chain. Late valve timing will result in a slight increase in low end torque and the engine will idle very smoothly. Other than the drop in fuel mileage, no other problems should be apparent.

Was this
Friday, July 23rd, 2010 AT 5:12 AM

Please login or register to post a reply.

Recommended Guides