A Call to Action: MPG Meter Disclosure

I owned a 1997 BMW 528 and you could get into the computer and set the guage to increase or decrease the calculation by a certain percent. It allowed you to dial it in very precisely. I'm surprised to see that you can't do that with the Prius.

REV

 

Between the cars my parents and I have owned w/trip computers (Prius and an 02, 04 and 07 Nissan), I've never seen any ability to adjust the readings from the trip computer.

 

All right, so let's take a whack at this.

The car manufacturers put those gauges on the cars, and, at least with the Prius, they all read high.

There was a major kafuzzle lo these many years ago when it was discovered that GM was setting the speedometers (and the odometers) high by 5-10%, and on purpose. Result was that the warranties were running out early.

They got taken to court, I rather forget by whom, and claimed, "But it's a safety measure! The speedometer reads 60 mph, but they're only going 55, so it's safer for everybody!"

This was countered by the National Bureau of Standards who said, "A pound is a pound, a meter is a meter, a second is a second, and you're not going to go running around on our watch claiming an MPH is whatever you say it is." The NBS (or whoever it was, national Weights and Measures?) won the day. The automakers of the time could have their Gaussian variations, but the mean durn well had to be on zero.

Imagine if the milk guys could claim a gallon is whatever they want. Pretty soon a gallon would be defined as something the size of a current pint!

So, the car manufacturers are playing silly buggers with MPG figures. I can live with inaccuracy. I don't like shifts of the mean. Let's set the Feds on them.

Or maybe just a class action lawsuit? Naw, I'd rather they just fix it than give a bunch of lawyers a hundred million for no good reason. Firmware fixes anyone?

KBeck.

 

Same thing with my 2001 Audi. Dealer diagnostic tool or an aftermarket one (VAG) could adjust it up or down to be quite precise. Must be a German thing.

 

Yup. It wasn't something in the owner's manual, it was kind of like hacking into a TV's maintenance program. Knowing the right butons to push in the right sequence. It wasn't hard to do once you knew how. I dialed that thing DEAD ON. If I calculated 22.7 ... that's EXACTLY what the in car computer said. I loved it.

I was kinda surprised there wasn't an option or at least a hack to do the same. It's very possible that there is, but we just don't know it.

REV

 

i had an early gen 2 (2004) a mid Gen 2 (2006) that was within 2% of being accurate and the car computer did NOT always read high. it was near 50/50. with the bladder in both a precise fillup was not possible so the tolerances mentioned above are Lifetime readings. the 2006 was driven 54,000 miles and it was only a .65 mpg difference which was just over 1% variance.

then i got my 2010. i posted a thread here in June 2009 questioning my Computer readings. Toyota probably responded to this and has improved the accuracy, but due to the early delivery of mine (VIN # 368) my accuracy is off by up to 10% (see signature) but its pretty consistently wrong which means that after "adjustments" for bleeding edge technology (my VIN on my Leaf is 258...maybe i should wait a bit longer?? then again NO F'ing WAY!!) states that the reading is "accurate" on the scale it uses (guessing 5,000 ft miles maybe?)

 

Huh? Where did you get this story, particularly the part about NBS and speedometers?

I received the class action lawsuit paperwork that whacked Honda and Subaru on their odometer errors. But your story conflicts with everything I've heard about speedometers, which is that they are required to be biased high. And that is consistent with all my modern cars.

 

Balderdash!

 

I'm enjoying this forum -- I don't think I have ever really checked the MPG in any car .... I have an idea .... but never really checked it.

As to speedometers -- in most jurisdictions they can never show a speed less than the posted speed on the gauge ... so they always run a bit high.

One of the reasons that may german cars have both types of gauges.

 

Actually, had it from my Dad back in 1963, when I was ten, so I do apologize if it was a little over the top. Just did the research, and here's the fair use quote, from Unsafe At Any Speed (Ralph Nader):

In 1963 the National Bureau of Standards (Department of Commerce) released the MacKay report, which showed with irrefutable exactness that for American automobiles, a mile is not necessarily a mile. For years, as some alert motorists know, Americans have been driving less than they think they have. The MacKay study showed that automobile odometers over-registered mileage on an average of 3.21 per cent, with some cars registering an error of over 5 per cent.​

Complaints about odometers have been registered for years with state agencies and the Federal Trade Commission. But state regulations defining the permissible margin of error were ignored by the industry and not enforced by the state administrators.​

Few practices can be more deceptive than tampering with the integrity of a measurement, whether it be miles, pounds, or inches. Few deceptions could serve such a variety of purposes. Car and tire warranties based on mileage run out sooner when odometers are over-set. Gas-mileage-per-gallon claims of manufacturers are overestimated or inflated, making easier the task described by Ford's Ray Pittman: "We fight for fractions of one per cent for fuel economy." A car owner could receive a lower trade-in value because depreciation is estimated partly on total mileage traveled. Over-set odometers tend to make the car owner think his vehicle is ready to trade in sooner, which helps feed the new car turnover. Finally, customers who rent cars pay for miles they did not drive. Based on the estimate of 1.25 billion miles traveled in 1964 by rented passenger cars, a 3.31 per cent overcharge, at the rate of ten cents a mile, would amount to an overcharge of almost four million dollars. An average 3.21 per cent premium on gross sales is a healthy fillip for the large, car-buying rental companies.​

The automobile industry learned of the National Bureau of Standards study of odometer performance in 1962. An Automobile Manufacturers Association odometer committee was formed to represent the industry in meetings with bureau officials who were working out new standards for the states so that odometers would have to register an average error nearer to zero. (Because a variety of conditions such as tire size, inflation pressure, weight, and road pavement affect odometer readings, it had been customary to provide for a plus-or-minus error tolerance range around zero.)​

The AMA odometer committee did not dispute the National Bureau of Standards findings. It stated that member companies ordered odometers from suppliers according to SAE specifications. SAE recommended practice J678b permits a five per cent over-registration error.​

But the AMA knew that the game was over, at least to the extent that it was played, and in December 1964 the Automobile Manufacturers Association informed the National Bureau of Standards that in 1965 manufacturers would install odometers that were set to the new bureau specifications.​

The AMA position, so long unquestioned by the public guardians of weights and measures, was wholly untenable from both an engineering and a moral viewpoint It was technically simple to produce more accurate odometers. Yet inquiries about the role of SAE as a ratifying participant in a fraud on consumers still elicit only the stock reply from SAE's New York headquarters that "the speedometer and odometer are designed and manufactured to be as accurate as possible."


There's more there, but you get the idea. I pretty much remember that my Dad told me there was a lawsuit. I did not make this up from the whole cloth!


KBeck
​

 

​

Thanks for the story. My memories don't reach that far back. This story parallels the much more recent class action lawsuits and settlements about warranty fraud from over-reading odometers, though much smaller errors than a half century ago.

But it doesn't address the speedometer side of the story. Those separate gauges have separate errors, and your earlier story conflicts with everything I've heard about speedometer errors. Because the Gaussian variance cannot be driven to zero, and because of various legal and industry edicts that the speedometer must not read low, the mean must therefore be pushed high. And that matches the consensus among those of us here who have measured this error. And many of us have found that the internal speed value known to the ECUs, as displayed on an engine monitor such as ScanGauge, is quite accurate. But the number displayed on the dashboard is different, and higher.

 

I have noted that European car speedometers tend to be spot on. Perhaps it's a reg, like DIN. Or it could be attitude. Having ABS standardized sooner could have helped because this made it easy to grab a 4-wheel digital average for speed and distance.

 

In my Cadillac and my VW, I used to track fuel consumption by pencil and paper to see how accurately the computers would track it. After some time, I stopped tracking it in both cars because it was usually off by only a tenth or two mpg, so that was good. So far, the Prius seems to be running off about 2 mpg on the high side. Even still, it's getting great mileage.

 

SAE J678 specification did allow odometers to read 5% high, it also allowed them to read 5% low. The reason for that was that speedometers were driven by small gears, usually with low tooth counts at the rear end of the transmission. Add to that variations in bias ply tires and a +/- 5% tolerance for production systems was reasonable. I had owned a few cars in that era and always checked my odometers and speedometers. My odometers varied one way or the other, and speedometers always read high, just like SAE recommended.

 

Here's my error record, all in favour of Toyo of course:

7.52
7.58
9.56
7.25
9.00
5.84
7.18
6.94
7.76
4.83
6.08
4.26
7.09
3.84
7.09
5.19
6.49
5.91
5.14
4.65
6.34
6.34
6.06
7.13
6.92
7.32
7.72
6.82
7.35
6.92


Bottom line, our Prius is very accurately inaccurate. I'm sure this could be rectified. So please, Toyota, stop BS'ing.

Our previous Honda Civic Hybrid's reporting of mileage was equally consistant, but slightly pessimistic, ie: it always under reported the mileage.

 

My actual vs. dashboard computed mileage is in my sig. I haven't updated that since the last two (Chicago winter) tanks, but that Toyota's in-car computer tells lies, is, no lie.

 

Lets face it, you would have to measure every gallon of gas you pump into your car to make sure your getting a full gallon! I myself will just enjoy my GREAT Prius!

 

In today's vehicles, it's just a program in an ECU that performs these calculations. If Toyota can "reflash" the system to fix an ABS problem (recall), they should be able to reflash the MPG calculation system to correct for the plus 3 to 4% error, which seems to be pretty consistent.