[ssrenity]

There are 3 parameters that can be changed via the Combination Meter System. If your Fuel Economy display is high or low (after manual calculation) you can adjust it. To enter the Dealer Customize Mode you MUST do everything within the prescribed time as described below.

3. DEALER CUSTOMIZE MODE
The combination meter can be customized with dealer customize according to the following procedure.
CAUTION:
Perform the steps described in 1) through 4) within 10 seconds.
1) Within 3 seconds after turning the ignition switch to ON, set the lighting switch to tail light or headlight position.
2) Press the trip meter knob five times.
3) Turn the lighting switch to OFF, and press the trip meter knob five times.
4) Set the lighting switch to tail light or headlight position again, and press the trip meter knob five times.
5) Move on to the “DEALER CUSTOMIZE MODE”.
• When the dealer customize mode operates, the LCD displays each adjustment screen.
• The dealer customize mode consists of three setting screens. {Avg.F/E Correction Screen}, {Ambient
Temp. Correction Screen} and {Clock Adjust. Screen} is displayed cyclically in this order every time the trip
meter knob is tapped.
• Holding down the trip meter knob while each setting screen is displayed can change the setting value.
NOTE:
When ignition switch is turned OFF or the vehicle is driven, the customize mode is cancelled automatically.
No. Customize mode Initial value Correction range
1 Avg.F/E Correction Screen (initial value+/-0%) Correction Range+/-10%
2 Ambient Temp. Correction Screen (initial value)+/-0°C (°F) Correction Range+/-3°C (°F)
3 Clock Adjust. Screen (initial value on) Correction range on or off

 

[chrisi]

Like most here, I've noticed my average fuel economy reading is optimistic, if this can get it closer to the real world it'll be great. I wonder what effect it will have on the miles to next fill up choice? I guess I'll find out.

 

[ssrenity]

I have experimented with the settings on my OB. It was reading 2-3 mpg optimistic compared to manual calculation. I first reduced the setting to -5 and that made the reading a bit pessimistic. I changed it to -4 and now it is very close to actual calculated mpg. Within a few tenths of a mpg. Winter fuel and weather has knocked down actual mileage from close to 30 mpg to about 26.4. I'll keep monitoring it and see what happens when the fuel changes back in the Spring.

 

[foxrider68]

Now this is the good stuff I like finding in these forums!

 

[vr4Legacy]

Thanks! Gonna try to figure out my average difference on MPG and see how well it works.

 

[hallux]

I have experimented with the settings on my OB. It was reading 2-3 mpg optimistic compared to manual calculation. I first reduced the setting to -5 and that made the reading a bit pessimistic. I changed it to -4 and now it is very close to actual calculated mpg. Within a few tenths of a mpg. Winter fuel and weather has knocked down actual mileage from close to 30 mpg to about 26.4. I'll keep monitoring it and see what happens when the fuel changes back in the Spring.

When you say that it's within a few tenths, are you forgetting it only changes in 0.3 increments? That alone can account for as much as a 0.3 MPG difference.

Unless there's a way to change how precise it is I'm not doing anything like this, the accuracy on mine varies based on the type of driving. I took a highway trip, the readout was only .25 off, if that, but my normal driving sees as much as 1 MPG off.

 

[ssrenity]

When you say that it's within a few tenths, are you forgetting it only changes in 0.3 increments? That alone can account for as much as a 0.3 MPG difference.

Unless there's a way to change how precise it is I'm not doing anything like this, the accuracy on mine varies based on the type of driving. I took a highway trip, the readout was only .25 off, if that, but my normal driving sees as much as 1 MPG off.

What I'm saying is if the computer readout is 26.4 the manual calculation will be within a few tenths of that +/-. I realize that the computer may be about to change to 26.1 or 26.7 so that is why I have tested this over several tanks of fuel. It is never going to be exact, but before I adjusted it the manual calculation was always 2-3 mpg less than the computer stated. Now it is very close. If yours is as close as you say, within 1 mpg, and sometimes .25, I wouldn't bother to change it either. Many of us who have been calculating every tank manually have found consistent optimistic computer readings 2-3 mpg higher than reality. I think Subaru made it adjustable for a reason.

 

[Bassman]

Pretty fortunate here, I guess. MPG and temp readings are pretty much right on with our car "right off the shelf." Clock is accurate too.

 

[30917]

The accuracy may have something to do with the winter/summer fuel blends too..Just 0.02 cts here

 

[vr4Legacy]

Does anyone know EXACTLY how the car measures mpg, or more concisely, fuel usage?
Does it use injector pulses and approximate fuel usage per pulse?
Does it have some way of measuring fuel as it is passes through the fuel lines?
Does it use a set standard for the amount of fuel that should be used per rpm?

It obviously uses the speedometer to figure the mileage part, which could account for quite a large variance, given no car has a 100% accurate speedo.

Just curious...

 

[upflying]

I would reckon most manufacturers calibrate the odometer and speedometer to be optimistic to read distances greater and speeds higher than actual.
This creates customer satisfaction that MPG is higher than reality. It also helps create an illusion of better performance. It also helps to keep the traffic cop away.
How Subaru calibrates the PCM is probably proprietary.

 

[Frumper64]

This should be a sticky - very good information

Jim

 

[mxfrank]

These gauges work off the ECU, by measuring the duty cycle of the injectors. Injectors can have small variations in fueling rate which end up having an observable impact on accuracy. But I think that the accuracy drops when the car isn't moving. I'm pretty sure my readings are better for 100% highway than for stop and go.

 

[vr4Legacy]

Thanks mxfrank, kinda what I thought. And yes I agree. When I take long trips with few stops my mpg is usually very very close to actual. But my wife drives the car daily with a very short commute, it can range as fars a 2mpg off.

 

[21staylor]

Saw this article with a response from a GM engineer.

Your Fuel Economy Gauge Is Fibbing - Edmunds.com

For what it's worth, I have a 2011 3.6 and my gauge has been over by 1 mpg averaged out over a year. It varies based on type of driving and by the tank but hasn't really varied summer to winter or vice versa.

 

[mpnret]

Saw this article with a response from a GM engineer.

Your Fuel Economy Gauge Is Fibbing - Edmunds.com

Am I missing something here? In the article Clark of GM blames the optimistic gauge on ethanol. Steve Manzor blames it on dirty injectors. Wouldn't both of these cause the guage to read lower rather than higher?

 

[mcaurilio]

Am I missing something here? In the article Clark of GM blames the optimistic gauge on ethanol. Steve Manzor blames it on dirty injectors. Wouldn't both of these cause the guage to read lower rather than higher?

Nope, I don't think you're missing anything. You're absolutely right, both of these would already be taken into account if the gauge calculations work as they say. Classic finger-pointing. Shady bastards, the whole lot of 'em. At least Subaru gives you a way to correct the calculations (maybe others do too, I honestly have no idea).

 

[21staylor]

Am I missing something here? In the article Clark of GM blames the optimistic gauge on ethanol. Steve Manzor blames it on dirty injectors. Wouldn't both of these cause the guage to read lower rather than higher?

Not sure about the dirty injectors being a real reason, but I could see it being true with ethanol. Since the gauge is only measuring fluid pushed through if that fluid didn't have the same energy content, the gauge would read higher than the actual mpg done by hand calculation...

 

[mcaurilio]

But that lack of energy content would be reflected in the less distance traveled, at least according to how the article states the trip computers work.

All else being equal, if ethanol-laden fuel is only 90% as effective as good-old gasoline, a car will only travel 90% as far on ethanol gas as it would if it were using real gas. So although the volume of liquid passing through is the same, the difference in distance would result in the trip computer already factoring in the difference between ethanol and straight gas.

So, in other words, the manufacturer's excuses for why the comptuers are inaccurate are complete hogwash.

 

[mpnret]

Not sure about the dirty injectors being a real reason, but I could see it being true with ethanol. Since the gauge is only measuring fluid pushed through if that fluid didn't have the same energy content, the gauge would read higher than the actual mpg done by hand calculation...

According to the article the amount of fuel being delivered is not actually measured. It is determined by the amount of pulses and duration of the fuel injectors. So for the same amount of pulses and duration a dirty injector would result in less fuel being delivered than thought for a given distance. This would result in a lower mpg on the display than the hand calculation. The hand calculation would actually know the amount of fuel used not a guess determined by pulses and duration. The same logic would apply to ethanol.