Toyota why is the mpg readout not accurate?????

???

 

Most the things you mention are irrelevant. Toyota could have left out that significant optimistic bias with no expense.

 

1. True_fuel_consumption ~= indicated_miles / indicated_MPG
2. True_MPG ~= True_miles / True_fuel_consuption

This allows me to drive larger diameter front tires in our 2003 Prius yet know the true MPG has not changed. Larger diameter tires reduce the indicated_MPG and indicated_miles. But both GPS and road mile markers have showing the same 6% calibration constant. This also explains the reports of new tires having lower, initial mileage that others have reported as new tires have slightly larger diameter.

Bob Wilson

 

From past discussions and observations, I'd come to a general understanding that the rolling circumference of our tires is very weakly connected to diameter. It seems that under normal loads and inflations, the contact patch is flat, not round, so the familiar circle equations fall apart. Rolling circumference is more closely connected to the length of the steel belts just beneath the tread, than to tire and tread diameter. That is why tire specifications can have a comparatively static 'RPM' (revolutions per mile) figure. But I don't recall you being a part of the same discussions I was in. Do you have any different insights to add?

Note to self: Instead of calibrating the odometer against 100 miles of Interstate Highway mileposts only once per tire set, do it repeatedly as the tires wear down. If the circle 'diameter' model is correct, the calibration constant should shift about 0.25% for each 1/32" of tread wear, or nearly 2% over the tire life. If the 'belt length' model is correct, the shift should be negligible.

 

I found the revolutions per mile to be the best metric from the Tire Rack specifications. Happily, we can read out MG2 rpm and it has a fixed relationship to revolutions which the Odometer and trip meter convert in indicated distance. So I use GPS and highway mile markers to figure out the calibration constant. The "indicated" is misleading.

Correct! Before I did my 1,000 mile tank, I checked the calibration constant against GPS measured distances and confirmed the calibration constant was slightly under the 6% originally measured. I used the more recent calibration constant when claiming my 1,000 mile tank.

Note that a GPS really takes several minutes to become truly accurate and then there is the variability built into the system. Longer distances and loops that return to the starting point tends to mitigate the GPS 'wander.' Mile markers are my gold standard.

Bob Wilson

 

"On the fly" MPG calculation is problematic at best. There are too many changes and factors to calculate for to know what your actual MPG is at any given moment, and your "average MPG" when you shut down is probably based on an averaging of data points saved during the trip...all of which might have been inaccurate.

Compare this to the simple math of "miles driven / gas put into tank".

Motorcycles are notorious for "off" MPH readings on the speedometers. Based on wheel size, tire size, gear teeth...all of these things throw off the calibration of the speed sensor. Bike makers compensate by having it read FASTER than you are actually going. Once I found that I had a 1/30th mile error in what my speedometer read, I bought a "Speedo Healer" which uses a calculation to recalibrate the reading so it is spot on.

That's just MPH. I don't think getting accurate MPG readings in real time would be as simple a fix....especially with a hybrid system that works to limit gas consumption. A straight ICE would be easier to calculate fuel consumption rate because it's the only means of propulsion you use.

 

We'll have to agree to disagree. My measurements have shown:

• MAF and injector timing -> very close agreement in fuel consumption after the engine goes closed loop, ~45-50 seconds regardless of outside temperature. This is how long it takes for the O{2} sensors to begin following the fuel short-term trim.
• indicated speed -> fixed ratio of MG2 rpm, regardless of tire revolutions per mile.

With accurate fuel consumption and a calibration constant for the tires, calculating an accurate, instantaneous MPG is trivial and quite useful when combined with GPS ephemeris of longitude, latitude, and altitude. Add the engine metrics of rpm and MG1 torque, it is trivial to map BSFC for any given trip and model what works and doesn't work in Prius, efficient driving.

I agree that if one takes the ordinary Prius metrics, they can be off. But that is often due to failure to measure the calibration constant for the true revolutions per mile of any given set of tires. There is one perverse aspect, an undersized tire with more revolutions per mile will give the illusion of higher MPG. The tell is the indicated mph will also be higher so traffic will appear to be passing all the time. But at unit offset errors, it is easy to not see the error especially at urban traffic speeds and not that easy at highway speeds.

So we'll just have to agree to disagree but I am planning to putting larger diameter, fewer revs per mile, tires on my wife's car when the current front set wear out.

Bob Wilson

 

I was never sure how the readings are calculated. I know in my boat, a fuel-flow meter was used, and was extremely accurate over time. Combined with the GPS data, it was quite useful.

 

Calibrating instruments is a habit that comes from earlier years when everything was analog. A good habit, it has been integral to my Prius studies. But what really brought it home were the tire specifications at Tire Rack when shopping for Prius tires.

Bob Wilson

 

A little OT, but do you rotate your tires? I forgot who posted that they don't.

 

No, I use four-wheel alignment and rotating tires screws up the feedback on whether the alignment is correct. So I'm using the European approach.

BTW, this also support tire experimentation without having to buy more tires than needed.

Bob Wilson

 

I logged some 10k miles with my CT 200h (a G3 Prius with a fancy suit) and also found an average long term 5.5% over estimate in MPG. I also checked rolling averages and found that with high probability any two successive tanks gave an over-estimate IIRC between 4 - 7%.

Are people really having trouble knocking 5% off their meter readings ?! For the truly arithmetic challenged I suggest subtracting 2 - 3 mpg.

 

Here's my last few tanks. All the errors are on the optimistic side, in case you haven't guessed. The 11.9% is an aberration, I'd disconnected the battery to do a brake job.

Sure you can apply the Toyota exaggeration factor, but come on... This is not due to limits of accuracy, or the errors would be falling on BOTH sides.

 

Yep.

My impression is that the meter is pretty darn precise, with a ~ 5% positive bias.

 

It's kinda interesting that the last few tanks, with lower liters per 100 km (higher mpg), also have lower error.

 

That doesn't sound right at all.

ALIGNMENT addresses how the wheel sits when mounted on the car. You don't mess with the wheel, you mess with the hardware holding it. Unless you have different tires on different wheels or different wheels on different positions, moving wheels around via rotation DOES NOT change these settings nor should it create a problem.

If you ARE getting an alignment problem from rotating your tires, you likely have one or more defective wheels/tires or a wheel out of balance.

 

^^^ I read the wear patterns and long term use makes the pattern more pronounced and unique to that wheel. In contrast, rotating the tires means the first wear pattern is 'additive' to the second position. I simply prefer to know the wear pattern at each wheel.

In the next year or so, I'll have to replace the front tires but sad to say, the same tire is no longer available. So I'll have to find something else but I know the metric that worked, ~850 revs/mile, so it won't be that hard.

Bob Wilson

 

Maybe that is so, but rotation ensures even tire wear so you don't end up with mismatched tires because all 4 don't wear down at the same rate.

 

Different GOALS yield different APPROACHES.

 

I think Bob is saying that rotation muddles and hides of the evidence of specific misalignment issues. He'd rather keep the evidence plainly visible so that he can better track and fix them.