calculated vs. computer MPG - Please post your results

After almost a month, finally emptied my 2nd tank. There is a downside to a fuel efficient car

The problem in Europe is that consumption is shown in l/100 km with a resolution of 0.1 l/100 km. Pretty crude.

The first tank showed 3.9 l/100 km on the MID, 4.27 calculated. But I filled the tank to the top of the filler tube. (Yeah, I know, it is better not to do that, I'm gonna kick the habit). The dealer probably stopped at the first click. With 17 km on DTE I managed to squeeze in only 39.48 l.

The second tank showed 3.7 l/100 km on the MID and 3.91 calculated. I filled it exactly the same as after the first tank to get an accurate reading. With -15 km DTE I got in 39.57 l.

The 3.7 however was a high 3.7, 30 km before filling up it still showed 3.8. Assuming the MID should show the rounded value (as opposed to the truncated value), it was probably 3.74 or 4.5% off, pretty much in line with the other reports and slightly worse than my Gen II.

Now to the result. 3.91 l/100 km is the best tank I had, beating my previous Gen II record of 4.09 that was set under ideal circumstances. My best 3 tank average was 4.23 l/100 km.

For me, yes, the Gen III does live up to its promises regarding fuel consumption.

Conversion table:

	Column 1	Column 2
0	l/100 km	mpg
1	4.2	56.0
2	4.1	57.3
3	4.0	58.8
4	3.9	60.3
5	3.8	61.8
6	3.7	63.5

 

Time for another update, as I've added another 1,000+ miles (probably more) and a couple of tanks of gas.

I still see the MID's MPG at about 5% higher than what I get from the normal calculations. The two most recent tanks were much closer. Perhaps this is because I was driving continuous highway miles (mainly) with as little as two or three stops between fill-ups, and as little as one day.

Here is the latest chart I have showing the percent difference (Computer over Calculated) for the lifetime of my 2010:

 

The 5% variance is same as Japanese owner's observations and a Toyota engineer's comment for the display target.

Ken@Japan

 

I would say that I now have enough data to say the 5% variance is close to being reliably established. Certainly, one year's worth of data would be more reliable; and one year's worth from a good sample set (25+ cars, I'd guess?) would truly establish not whether there is an issue but exactly what level of variance exists.

The following is the raw data after 6,103 (6,102.9) miles. Because I did not know the actual fuel level at purchase, I only use the miles starting after my first fill (at 311 miles). Therefore, the numerator I use at 6,103 miles is 5,792 miles. The denominator is, of course, the gallons computer-calculated or purchased across only those 5,792 miles.

Multi-Information Display (Computer): 5,791.6 miles, 101.777 calc. gal. for 56.905 MPG

Fuel Purchases Data (Calculated): 5,792 miles, 106.774 gal. for 54.25 MPG

The MID mileage is, across the 5,792 miles of measured data, 4.90% greater than the calculated mileage.

 

I guess the silver lining is that is appears to be consistent @ a relatively consistent speed. Therefore, I will always subtract 5% from my MID estimates on my regular commutes.
I do hope Toyota provides a fix in the future.
Peter

 

On my first tank where I filled it up, I got 45.7 mpg on the display, and 43.7 for 419 miles of driving. That's close to a 5% difference. Mix of city/highway (40/60 I'd say).

 

Ok, just did my 2nd fillup (first done by dealer and I was not completly sure of the mileage). My first was 367 miles added 8.235 gallons for 44.57 calculated - display said 46.7. 2nd fillup was 403.8 Miles 8.871 gallons for 45.52 calculated - display said 48.3.

I am no expert at driving the Prius (obviously) but have been running the whole second tank on ECO mode (no PWR mode) and trying to do some pulse and glide. So I am a little disappointed that I did not go up but 1 MPG on the second tank. The first tank was almost all in regular mode with some mixed PWR and ECO mode.

However most of my driving is small trips (I work from home so no commute).

This is still more than double what I was getting on my Rav4 (2001) (average of 22 MPG) so I am still glad I got the car (not so happy to have a car note again though).

 

I'm getting slightly different results, but I have an unusual commute. I drive 150 miles on 1-95 each day with only 34 or so being city driving. Bright side is the engine is warm by the time I get to I95. My 5 tanks are distinctly showing the MID display getting CLOSER to the actual calculated MPG. The first was off by 4MPG, over 460 miles and just under 9 gallons, the second was a little better and so on until the 4th one that read 59MPG on the MID and over 57 calculated! At the same time the car has been breaking in and I've been increasing my tire pressure between tank fills so that most probably plays into it. I fill up 3 times in two weeks so I should have some results for the latest tank early next week.

BTW, I am running 44 PSI front and 41 rear and I really like the ride! It is more comfortable imho, and less jarring over freeway turds (lane divider bumps). Other than the engine weight being in front, is there a special reason we are told to inflate the front to 2-3 PSI higher than the rear?

I also have been using the eco mode almost exclusively. I tried Power mode but it was too much like the GO pedal was an on-off switch! Reminded me of a 300Z I drove once. And no Cruise Control, it really sucks gas on the CC because it never seems to find the sweet spot.

 

And I thought it was just me. My Gen II had some swings ... which I normally attributed to differences in gas pumps at various gas stations ... but over 20,000+ miles, the numbers stayed pretty close. My Gen III so far at 3,000 miles ... the computer mph has consistently been about 2mpg better than what I actually calculate.

So far, my Gen III is getting almost the exact same calculated mpg as my Gen II ... on a tank by tank basis. I expected better but the weather here in Dallas has been hot ... which hurts mpg ... and I have the 17" tires, which I am guessing will cost me 1 or 2 mpg.

 

My last tank, 50.7 per computer, 47.9 computed. This variance of about 5% computer overstatement is similar to what I get from my 2005 Passat TDI wagon's computer versus measured consumption.

I am in 110º weather, use a/c 100% of the time, 50/50 surface street versus freeway, all city driving. I engage in no hypermiling antics and although I do not do jackrabbit starts, I keep up with traffic from the light.

 

I have managed to read the whole thread only now.

First comment: considering that the car does not know how much fuel you put in (no it doesn't, it doesn't read the receipt; the fuel gauge is an estimation and to make a 0% precise fuel gauge is really expensive), a 5% error on fuel consumption estimation is really not that bad. I personally think it is acceptable and as an engineer it does make sense in terms of system requirements. It probably averages off over time.
Second comment/opinion/theory/guess: 5% error on an mpg in the US vs L/100km like in EU, does look psychologically bigger.

Example:
In EU, the combined declared fuel consumption of the Prius with 17" wheels is 4L/100km = 25km/L = 59mpg. (BTW: in my experience they are right - after 7 years driving a Polo, my average is 5.6L/100km in town - dead on the city average declared and tested according to EU norm by VW when I bought it - I was *really* surprised to see that it matched that well).

If the error is constant, regardless of the units used (ie. L vs gallon; miles vs km.), the 5% error on the EU MFD would look like 4L/100km/1.05 = 3.8L/100km = 26.3km and that is not a big difference (1.3km difference; psychologically not much).
In the US instead 5% would look like: 59mpg*1.05=61.95 or about 62 = 3 miles delta = 3 mi *1.609 = 4.8 km - wow! that is a lot...

Because calculations in the US are based on different units, could it be that the errors are about 4 times bigger? E.g. 1 gallon = 3.78 liters, which amplify how 5% error looks like. For me 1.3km is nothing - 3 miles for US drivers seems to be a lot.
Well, in both cases is always 5% difference.

Considering that in the EU the MFD shows 3.8 or 3.9, but not 3.85 (and I would not care less to see 0.5L/100km difference in fuel consumption), if it shows 3.8 rather than 4 I will not panic about it. Moreover, MFD are very expensive to make them very accurate - all car manufacturers try to save money on the sensors used for these things (i.e. see external temperature - ever checked the error on those ones???) and have to compensate like crazy with software.
An ex-colleague of mine has been working on a company making the programming of the microcontrollers for MFDs of BMWs and VWs here in Munich and she told me that the sensors used for the temperature were so bad in terms of precision (yes, even on BMWs!), that they had to do all sort of tricks to try to compensate the error introduced by the sensor itself.

I wouldn't freak about a 5% error on mpg - you are probably getting the same on external temperature and reported speed. On all cars. After all, regardless of what the MFD says, the truth lies in how many miles/km you drive and how much fuel you put to drive them. The MFD can say whatever it wants.
And another question - how does the MFD in the Prius calculate the FE? what sensors, placed where, are used? how does it take into consideration the fuel used when charging the battery while not moving the car, when using only the battery to move, to coast, etc.etc.etc.? Where and how is the fuel consumption controlled? using what method/sensor? at the tank? at the injectors?

One last consideration: declared fuel consumption, whether calculated according to EU or US norms need always to be taken with a bit of caution. I use them to give me an idea of how a car will perform compared to another one from another manufacturer. At the end of the day it depends on how you drive a car - all manufacturers will put more or less the same mistakes (intentionally or not).
And please don't get me wrong on the following. But considering that in the US you are not driving diesels, you invented gas guzzlers, SUV and Hummers and usually drive 2.6L liter cars that get 10km/L - 25 mpg on highway at constant speeds! I would really not discuss/freak too much whether the Prius, that finally gives you access to what elsewhere on the planet is normally accessible (smaller engine cars and clean and powerful Diesels), is doing 5% more or less than what declared. Remember that in EU we pay 1 gallon gas 5€, you freak out when you pay it 2.5$!!!

 

If you are talking about a +/- 5% error, I could understand your position, though it still seems a bit high to me.

Unfortunately, it does not appear that the error follows much of a bell curve centered on zero.

I would like to think this is the case, and I won't be able to fully refute your assumption for a long time. Still, I have driven 6,500 miles and filled the tank 14 times, so I think the average is "settling" in at a fairly predictable place: 5%. The highest overstatement was greater than 7%, while the lowest overstatement was just under 3%. Remember: I have not yet had an understatement, and I haven't heard from anyone who has had regular (much less repeated) understatements.

Here is the latest update of my chart:

 

I just did my second fill-up. The actual mileage was 61.6 and the display said 62.7, a difference of 1.1 mpg or about 1.8% of the actual mileage. My first fill-up showed the display to be over by 2%. So I for one am not seeing the big error that some others are reporting. The display seems pretty close.

 

Vacation trip, 52.0 MPG on display, calculated to 48.8MPG

 

I am seeing ~ 2 mpg less when manually calculating economy on my 2010 III.

 

Can you offer any specifics?

Are you recording both the MID readout (in your car) and the calculated MPGs for every tank? If so, then please share this information.

 

Looking at how this thread is unfolding, I would like to see more and more people share there actual data. If Toyota has actually programmed the computer to produce higher-than-actual numbers, I'd like to have sufficient data to bring this to their "attention."

Perhaps it isn't that big a deal, but the Prius has such high MPG numbers that there is no reason to exaggerate the claims.

 

I posted the following on another MPG thread, but I thought I should repeat it here (sorry for the duplicate post!)

I am finally getting around to posting my first two fill-ups. I went on a weekend road trip from Syracuse NY to RI and back and filled up my car twice.

Fill-up #1:
after 200+ miles in-town and the rest highways (including the hilly Berkshires through MA)
436.8 mi, 48.4MPG MFD, 46.9 at the pump. This was on the free dealer-filled tank.

Fill-up #2:
Mostly highway miles, including the hilly Berkshires.
474 mi, 51.6 MPG MFD, 51.09 MPG at the pump

I wasn't "hyper" about hypermiling and I used the AC only occasionally. I am very happy with the mileage so far!

I am still working on tank #3 and am finding that my mileage is hovering in the 47MPG range for mostly around town. I do live in a hilly area and I had the car undercoated to combat the salty roads in the winter here, so I suspect my mileage may not hit 50MPG around town until I improve my hypermiling skills. LOVE this car!

 

Fill-up yesterday. Indicated 51 MPG. Actual 48 MPG. Computer is definitely in error...