This is absurd...

[Martinjmpr]

how exactly do you figure out mpg?

 

i'm not really worried but was curious.

 

Fill the tank

 

Set the trip odometer to 0

 

Drive around until the tank is empty

 

Fill again

 

Divide miles driven by gas consumed, = MPG

 

It helps if you use the same gas pump each time, as different pumps have different "shut off" settings. Also, I've noticed on my 99 OBW, the filler neck "burps back" quite a bit, so if you pull the filler out after the first time the gas shuts off, you will be a couple of gallons short of full. The important thing is consistency: Whatever you do at one fill up, do at all of them, that way your mileage figures will be more accurate.

 

BTW, I get around 22 in our small town and 26-27 on the highway in Winter (running 85 octane at 7200' elevation), that's with 81k on the odo.

[Spazz698]

My mileage has actually been way up lately. Taking the roof rack off and putting a few extra PSI in my tires definitely helped. I was reading the other day that the recommended pressure on my door is a little low for modern tires and such. Now instead of 30/29, I've got 34/33. I could probably go higher.

 

On a trip from LA to the Bay Area, I managed to get 25mpg with three passengers and our luggage for a weekend, driving 90mph (seriously that's how fast people go on that stretch of the 5).

 

 

From here:

 

That would explain why my tires were sitting at 30 instead of the 44psi the tires say, think I even got an extra 50 mi out of my tank too

[LosDiosDeVerde86]

That would explain why my tires were sitting at 30 instead of the 44psi the tires say, think I even got an extra 50 mi out of my tank too

 

going any higher than 35, and you'll start to get center wear. you only want to go max (44) when you're really haulin a load.

[SevenSisters]

Don't know how hard it is to reach on your car, but drive around until it's hot or heat up the manifold with a propane torch. Buy a special O2 sensor socket.

[M@dM@x]

I'm sure this may have been metioned, but welcome to winter gas. I usually lose a few MPG on all of my cars during the winter. In a few weeks we will actually have real gas again.

[medfurd]

The O2 sensor voltage output must flutuate rapidly between 0.1 and 0.9 volt and must cross 0.5 volt at least 5 times in 10 seconds.

 

You're the only one that has mentioned this. O2 sensors aren't just throwaway items like air filters. Sure they go bad but you can check. And they don't usually fail gradually. They either work or they don't. But you don't need Palm software, just get a multimeter at Autozone and test the voltage with the parameters frag gave. You could have a shop check it too, it's simple, but they might try to sell you a new one anyway.

[frag]

You're the only one that has mentioned this. O2 sensors aren't just throwaway items like air filters. Sure they go bad but you can check. And they don't usually fail gradually. They either work or they don't. But you don't need Palm software, just get a multimeter at Autozone and test the voltage with the parameters frag gave. You could have a shop check it too, it's simple, but they might try to sell you a new one anyway.

 

Mostly right except for the part underlined. A few of us here have had bad gas mileage that was cured by replacing an 02 sensor that was'nt giving a CEL. Like, I think, Commuter once wrote, they often become «lazy» (let's say crossing 0.5 volt 4 times instead of 5) after a few year and/or 100 K miles and begin causing an overrich situation without triggering a CEL. That's why a precise test is necessary to detect that situation.

Most of the time, if your 02 sensor is 100k old, if you have abnormal soot in the tail pipe and if gas mileage is coming down without any other explanation, it's a sign the 02 sensor has maybe become «lazy».

[medfurd]

If one's rear O2 sensor failed the same test, would you replace it? B/c I know the front one is supposed to be more important.

[frag]

If one's rear O2 sensor failed the same test, would you replace it? B/c I know the front one is supposed to be more important.

 

The test for the rear sensor is not the same. I think the rear sensor's reading should stay stable approx in the center of the range : 0.5 - 0.6 volt. Since the upstream one is doing all the compensating, the rear one only measures the result which should not be fluctuating.

[Al_SemC]

And they don't usually fail gradually. They either work or they don't.

I disagree. They get lazy over time. Their response time gets higher and the maximum signal swing they generate gets smaller too. While a new sensor reads 0.1V for lean and 0.9V for rich and does it in 350ms or so (when hot), an old one might read 0.3V for lean and 0.6V for rich and does it in 450ms (when hot). This causes the ECU to incorrectly adjust the fuel trim, decreasing fuel economy, causing rough idle and missing, etc. Basically, it throws off the automagic tuning. My source is someplace on Autometer's website. I'll update when I find it.

 

EDIT: Here it is.

 

 

The test for the rear sensor is not the same. I think the rear sensor's reading should stay stable approx in the center of the range : 0.5 - 0.6 volt. Since the upstream one is doing all the compensating, the rear one only measures the result which should not be fluctuating.

The post-cat sensor should be more stable (flit around less) than the pre-cat one and it should hang around stoichiometric (0.5-0.6V), as frag said. If you find that it's unstable or it's reading some whack assed measurement all the time, either it's dead or your cat is. This generally causes a CEL.

 

You should make sure that your O2 sensors are working properly. Running rich eats gas, blackens the back of your car, makes it run like hell, accelerates the death of your O2 sensors, and destroys your cat. Running lean causes detonation which can, over time, blow the crap out of your engine. Both not cool.

 

As a final note, these things are generally easier to read with either an oscilliscope (those of us that have them) or an analog voltmeter (one with a needle). Very rarely do people have digital voltmeters that update their reading often enough for you to get more than a rough idea of what's going on. With the oscilliscope, you can watch the actual trace and get a nice sensor output vs. time graph. With the analog meter, you can watch the needle sweep back and forth and use that to approximate what it's doing. Otherwise you're just looking at rapidly changing numbers.

[Guest jclay]

Going back to the whole fuel treatment issue, I'd say thats a good place to start as well. Also, check the fuel injectors. Mine started to leak due to rotting o-rings. Less fuel made it to the cylinders=less MPG

[garner]

I went ahead and bought some new plugs

 

didn't buy Bosch by any chance did you? Use only NGK plugs in Subarus. I dropped 6 mpg on my 03 TS after foolishly thinking Bosch platinums were work $5 each, swapped back to NGK and reset the computer, prest back up to 28-30 (unless I did like a maniac)

 

O2 sensors are really important too...

 

garner

[FindingForester]

Low-emission "winter" gas has a lower heat of combustion than straight gasoline does. That is, it yields less energy when burned. Cold air also reduces gas mileage, including that it will take longer for the engine to reach normal temperature, so you have a double whammy against gas mileage in winter.

[shortlid]

I have heard O2 sensors just get tired after awhile.