[Jeffand]

Thier has been some talk going arround that the actual gas milage of the Prius is different than what the display says.

Since the the gas tank bladder makes it difficult to get accurate gas fill amounts. It will be necesary to use multiple consecative fills to calculate the gasmilage. Be using multiple fills it will reduce the affect fill errors have on the milage calculation. This should produce more accurate numbers.

We will need miles traveled, gallon used to fill tank, and Prius computer display milage. For four consecative tank fills.

 

[Richard]

My display has told me about 1 mpg more than what I typically calculate. I wonder if it's not taking into account gas that's used for the air conditioner. What happens to that fuel, anyway? Does it directly power the A/C, or does the A/C draw power from the engine?

 

[mdacmeis]

Toyota's official statement is the displayed MPG is very accurate.

That said, my fuel records track the trip mileage, gallons pumped, MFD displayed average mileage, and the calculated trip mileage based on gallons pumped * MFD average. My last 3 tanks are:

calculated miles ------ actual trip miles ------ MFD mpg
268---------------------------269----------------------48.4
452---------------------------449----------------------47.5
185---------------------------187----------------------46.8

Each fill-up was done without knowing or considering how much fuel was required to fill back up. As you can see, the mileage in my case is indeed extremely accurate. Also note these 3 fill-ups are widely varied with respect to the amount of fuel needed, yet I am getting consistent fill-up results.

 

[Jeffand]

The Prius has a unique electric inverter air conditioning system. Instead of running off the fan belt, the new inverter air conditioner is fully electrically operated. Not only does this improve fuel efficiency, it ensures passenger comfort, even when the gasoline engine is off and the car is being propelled only by the electric motor.

 

[adamtraidman]

Jeffand - Thanks for explaining that, I was wondering how the AC stayed so nice and cool when the ICE was shut off.

Adam

 

[Richard]

The Prius has a unique electric inverter air conditioning system. Instead of running off the fan belt, the new inverter air conditioner is fully electrically operated. Not only does this improve fuel efficiency, it ensures passenger comfort, even when the gasoline engine is off and the car is being propelled only by the electric motor.

Does this mean I'm not using gas when I have the air conditioning on? Because if so, man, that guy's staying on all the time.

 

[BIF]

I just recently drove over 500 miles, got down to 1 mark (not blinking), and decided to fill up.

The tank took 11.45 gallons before the pump clicked off. Yikes! That was a close one...the tank only holds 11.9 gallons.

And y'know, I was in a strange town going to a store, and I unwittingly stopped at the LAST GAS STATION to the place. It was a good 12-15 miles from that gas station to the store and back. I would have certainly been running on fumes had I tried to take that last leg without filling up!

 

[RSnyder]

Does this mean I'm not using gas when I have the air conditioning on? Because if so, man, that guy's staying on all the time.

Sorry, even the Prius can't beat the laws of thermodynamics. The energy to power the A/C come from the high voltage electrical system. This comes either directly from the gas engine spinning the generator(s), thus using gas, or from the high voltage battery, which must be recharged later, also by using gasoline. Since the battery only returns 80%-85% of the energy you store in it, the latter case ultimately has an additional gas tax. But it's still a win over the 2001-03 Prius because the gas tax for running the engine when it's not needed to move the car is huge.

 

[Richard]

But...but...but I want to keep that guy on all the time. Please, just nod and say yes, that will be like permission to keep that guy on (you guessed it) all the time.

 

[BIF]

<nodding my head "yes">

I can run my AC when I'm parked or at a traffic light, and the hybrid battery's rate of discharge is not usually appreciably accellerated so much that the ICE has to turn on while I'm "standing."

When the light changes, or when I leave the parking lot, the ICE always turns on, even if I'm moving slowly. Even if the AC is off, the ICE always turns on. So to me, this means that I'm burning gas no matter what I do with regards to the AC. So use it, man! Or get used to changing shirts each time you get to your destination!

 

[AgentElrond]

Toyota's official statement is the displayed MPG is very accurate.

I'm somewhat inclined to disagree with Toyota. :wink:

Depends on the car though. My wife's 2003 has differed anywhere from 0.6-6.7 mpg OVER actual, without any particular pattern. My 2004 however came in at 5.4 mpg over actual after the first tank, and has dropped to be "more accurate" (?) with each tank, down to 0.6 mpg over actual on my last fill up.

I might be willing to believe differences in refuel level, except then I would expect semi-alternating high/low readings that "average out" over time; neither car has ever reported LESS than actual.

 

[RSnyder]

But...but...but I want to keep that guy on all the time. Please, just nod and say yes, that will be like permission to keep that guy on (you guessed it) all the time.

When my wife is in the car, the AC has to be on all the time (in the summer). And Pikachu is a 2001, so the engine runs whenever the compressor runs. So yeah, I'll give you permission :wink:

 

[johnson487682]

Toyota's official statement is the displayed MPG is very accurate.

I'm somewhat inclined to disagree with Toyota. :wink:

I agree with TechnoMage. I think improvements were made in the '04 model's displayed mpg accuracy, but many people have reported that the NHW11 models consistently display an overly-optimistic mpg.

I experimented with mine shortly after I bought it. Over 10-12 tanks, I recorded tank mpg according to the display and according to the pump. The two average values fluctuated quite widely for the first few tanks, due to the variability induced by the bladder. However, after the final tank in my experiment, the difference between the two averages had settled to a quite stable 5.6% error, with the display reading higher than the pump value.

The difference between the two readings is a function of the measurement method. Pump measurement is legally required to be accurate, because it uses volume from the pump (which is calibrated and tested regularly by the government) and distance from the odometer (which is required by the government to be accurate).

Display measurement is based on fuel injector data: given that the fuel injector sprays fuel at a fixed rate when it is open, total up the amount of time it has been open, and multiply by this rate to get the amount of gas used. However, this rate has tolerances such as +-3% (i.e. some injectors will spray less, others more). It has been hypothesized that (for marketing reasons) Toyota chose to adjust the computation algorithm so that all cars will report at least the accurate value; maybe more but never less. So if the tolerance is indeed +-3%, the displayed value for any given Prius should be 0% to 6% high.

Toyota can still call this "very accurate", because an average error of +3% is still very good by most standards. Most measurement systems (digital or otherwise) use a few op-amps and a few 1% resistors to digitize the input data, and these 1% errors accumulate through several input stages, so a 3% error isn't unusual.

What does this all mean? The displayed average values are useful for comparing one trip to another or one tank to another (i.e. relative measurement), but they are not useful for long-term absolute measurement, which is probably why the '04 resets the displayed average every time you fill the tank. The only way to be sure of your lifetime mpg is to use pump measurement.

Douglas (2002 Silver, Wisconsin)

 

[mdacmeis]

Toyota's official statement is the displayed MPG is very accurate.

I'm somewhat inclined to disagree with Toyota. :wink:

Depends on the car though. My wife's 2003 has differed anywhere from 0.6-6.7 mpg OVER actual, without any particular pattern. My 2004 however came in at 5.4 mpg over actual after the first tank, and has dropped to be "more accurate" (?) with each tank, down to 0.6 mpg over actual on my last fill up.

I might be willing to believe differences in refuel level, except then I would expect semi-alternating high/low readings that "average out" over time; neither car has ever reported LESS than actual.

This is referred to as "fuel fill variability." You cannot judge accuracy over a single fill-up. There is too much variation as a result of the pump, the bladder, etc. You need to total the fuel used over about 10 fill-ups, divide your total mileage over this same time by the total fuel pumped, then make an assessment. You can track the computer by dividing your trip mileage by the MFD MPG for each fill-up (assuming you reset at each fill-up), thus calculating the amount of fuel the computer thinks has been used. Total that over time and compare that to the total gallons pumped over time. I, for some reason, have been able to consistently fill without much variability. I am sure this will change when winter comes.

Note: If you pump 9.5 gallons into the tank, instead of the 10.0 gallons you could have pumped, this is only a 5% error. At an average of 45 MPG, this would translate into a trip mileage error of 22.5 miles over a distance of 450 miles. The MPG is thus calculated as 47.4 versus the 45 actual, which is also likely what the MFD displayed. As you can see, fuel fill variablility will always result in proper filling of the tank or underfilling, and never overfilling. As a result, the mileage you calculate can be higher or lower than what the MFD displays, until you average the results over time, which helps average out the periodic under-fillings.

Lastly, the MFD calculated the fuel used by multiplying the time the fuel injectors are on by the lbs/hour of fuel flow they are rated. The injectors are pretty consistent, however a 1 -2% under or overflow is not uncommon. The powertrain control module adjusts the injector on-time to address emissions and performance affects from this variation, however these adjustments are not reflected in the average MPG, since the change in on-time does not alter the inherent flow error. This is why there is some variation from vehicle to vehicle. Combined with the fact that each auto manufacturer tends to use different software to calculate the MPG numbers, comparing car to car can be misleading. The intent was never to be dead-on accurate, rather to be close, within about 3% of actual useage, limited by injector flow and calculation errors.

 

[Richard]

But...but...but I want to keep that guy on all the time. Please, just nod and say yes, that will be like permission to keep that guy on (you guessed it) all the time.

<nodding my head "yes">

...I'll give you permission :wink:

Thanks, guys! I heart Auto A/C.

 

[jshore]

I've been wondering about this accuracy issue recently, so I went to John's website and downloaded his spreadsheet data for both his 2001 and 2004 Priuses. He had a column which calculated his total calculated lifetime MPG according to (total gallons according to gas pump)/(total miles according to odometer). I added a column which calculated lifetime MPG according to (total gallons as inferred from screen MPG readings)/(total miles according to odometer). [For a particular fill-up, you get the gallons as inferred from screen MPG readings by dividing the miles travelled by the screen MPG. Then, you just keep a running total of this column to get the "total gallons inferred from screen MPG readings".]

What I find is that by the time he had traded in his 2001 Prius just shy of the 60,000 mile mark, his lifetime MPG calculated by using the gas pump method was 2.1 mpg lower than that calculated using the screen method, which meant the screen had averaged about 4.6% high of the gas pump estimate.

For his 2004 Prius at its current mileage just a little shy of 20,000 miles, I find that gas pump method is running 1.45 mpg lower than that calculated from the screen, which puts the screen almost exactly 3% too high.

So, there you have it for one particular 2001 and 2004 Prius for which we have great data. (Thanks, John!!!)

On my own Prius, I am running ~2 mpg lower via the gas pump measurement than via the screen, but that is over only a little over 2000 miles...And, the numbers have been converging somewhat with each tankful, leading me to suspect that part of that discrepancy may be due to the dealer not having filled the tank that well the first time. (In support of this notion, I also seem to remember that I lost my first "bar" on the fuel gauge earlier on that tank than on subsequent tanks of gas.]

Of course, there is another source of error in the odometer itself...This whole discussion of the overall accuracy of the MPG measurements is predicated on the notion that the odometer itself is quite accurate since both measurement techniques use that data as input. Anyone have any idea what they are good to, considering issues of varying tire pressure, tread depth, and the like?

 

[jfschultz]

Does this mean I'm not using gas when I have the air conditioning on? Because if so, man, that guy's staying on all the time.

Sorry, even the Prius can't beat the laws of thermodynamics. The energy to power the A/C come from the high voltage electrical system. This comes either directly from the gas engine spinning the generator(s), thus using gas, or from the high voltage battery, which must be recharged later, also by using gasoline. Since the battery only returns 80%-85% of the energy you store in it, the latter case ultimately has an additional gas tax. But it's still a win over the 2001-03 Prius because the gas tax for running the engine when it's not needed to move the car is huge.

My wife and I had a disagreement on when a store opened and we got there about a half hour early. (She was wrong he he.) We we pulled into a parking space and I hit the park button and the parking brakes and we just sat there in air conditioned comfort while waiting. The SOC slowly went lower and lower then the ICE started up and ran for about 30 seconds and cut off. This happened one more time before the store opened and I turned off the car.