Your Fuel Economy Gauge Is Fibbing: Edmunds Testing Finds Overestimated MPG Is Common

I need to remember this for the next time TDI fanboys talk about beating Prius.

 

I found our previous '06 Honda Civic Hybrid in-dash fuel economy indicator slightly underestimated fuel economy, ie: if it said 4.5 liter/100km, calculated would usually be 4.4, sometimes 4.3.

 

The gauge on my Insight-I has been accurate within 0.7% over >60,000 miles: LMPG reading is 84.6mpg, hand calculations give 84.04mpg. So yes it slightly overestimates mpg overall. Tank by tank, the two methods may differ by up to 1 mpg either way.

I found most of the "explanations" from the mfr's to be ludicrous handwaving. GM's claim that ethanol content affects mpg calculations due to variations in energy content??? Huh?? We're measuring miles per unit volume irrespective of energy content. Yes the mpg you get will vary due to ethanol content, but the two numbers should agree with each other.

It's surprising that nobody mentioned two likely actual causes of the discrepancy:
(1) pump accuracy - if the pump overcharges you slightly by claiming it pumps more gas than it actually does, your hand calculations will be lower than the real mpg. I would expect that pumps would err in this direction.

The fact that the discrepancy in my Insight-I is so small suggests (but does not prove) that gas station pumps may be fairly accurate. But I use first-tier gas (mainly Shell) in CT and MA. Perhaps pumps at cheap-gas stations or in other parts of the country are less accurate??

(2) fill accuracy - you may not refill the tank to exactly the same level that you did when you started the measurement. Possibly the car is tilted a bit differently than it was at the fill, perhaps the pump (or you) just clicks off at a different point. In this cast the discrepancies would be random.

What makes the most sense is to just use the dash display as an indicator and track mpg with hand calculations. The advantage of the dash display, which nobody pointed out, is that it allows you to see how you're doing between refills: you don't have to wait days or a week to find out how your driving and conditions affect mpg.

 

Sure .... they certainly would not underestimate the fuel efficiency.
My Dodge is right at 5 mpg overstated.

I assume is is difficult to engineer a simple, least costly, accurate system. After 2 years of documenting the displayed verses calculated, the difference each fill are inconsistent.

 

they do not measure fuel directly, rather use Mass Flow sensor reading to estimate amount of fuel burned. Ethanol contains oxygen, thus requires less oxygen to burn. This means more fuel will be burned per cubic feet of air.

 

Interesting. My fuel consumption display has been within +/- 1mpg for most of the time I've owned my car and most owners of the 2009+ TDIs are reporting similar results.

 

I noticed for me, it's harder to keep the gauge on the Prius consistent as the gas tank gets emptier. Maybe it was just an anomoly when I noticed it, but my avg MPG stayed relatively consistent until I got to around 4 pips on the gas guage lately, then it became harder to keep it around the same. It should take a lot to get it to drop 1mpg after you've put about 250-300 miles on the tank. But I could get that drop after 20 more miles of driving, and it's the same route every day.

I wonder if I should try to track fill amts vs display difference to hand calc difference. Do some varying 1/2 tank fills, 3/8 used, 3/8 left, drained to the final little pip. It would seem quite odd that this would affect things that much, though.

 

Our Prius mpg display (actually liters/100km, same thing) is running around 7% optimistic, year-to-date since I started calculating in January. At least it's a very consistant error, LOL.

 

My error has been pretty constant too. My mileage is also fairly consistent, so I've adopted a "subtract 3.5 mpg" from the displayed rate as my "true" MPG while driving.

There have been a few instances where the pump didn't bear this out, but this was highly likely due to a fill difference at the pump, because, invariably, the next fill up will be off as well (and if you averaged the two together, they are spot on to the expected value).

I feel guilty when I show off the instant MPG reading (and I plan on doing that this weekend) and feel the need to explain that the mileage really is 3.5 mpg less.

I'd like for it to be more accurate, but at least it is pretty precise, so I can adjust the displayed value (mentally) to be accurate.

 

Are you referring to a specific system? Some systems count injector pulses, which would deliver the same amount of fuel regardless of the type of fuel.

Tom

 

I log and track the dash display reading AND the actual miles/gallons at each fillup. My 2004 Prius display showed a 1.3 favorable mpg number (2.6%). My 2007 shows a 1.4 favorable mpg (2.8%). If you know that bias figure, then the dash display is pretty accurate for each tank. It is more consistant than the miles driven/gallons on fill. But you have to track the miles driven/gallons filled over many tanks to accurately assess the bias.

 

To those who know how accurate your display is vs the real consumption I ask:

What brand/model/size tires? Do you know the official Revs per mile (RPM) of that tire?

I ask because changing your tires changes the accuracy of the speedometer/odometer and presumably will affect MPG calculations since the miles are different "lengths".

Going further my working assumption currently is that the computer assumes 845 RPM tires and all calculations are off depending on the ratio of actual RPM to expected RPM.

I started making a list of OEM tires and revs per mile in another thread, still looking for EU/UK tires for the list but I'm pretty sure the US choices are all here.

15" Rims (standard aluminum rims in the US 2004-2009)
185/65/15 (2004-2009 Standard U.S. OEM tire size for Goodyear Integrity at 855 RPM)

15" Rims (standard alloy rims in the US 2010+)
195/65/15 (2010+ OEM tire size 1 Goodyear Assurance Fuel Max at 836 RPM)
195/65/15 (2010+ OEM tire size 2 Bridgestone Ecopia EP20 at 833 RPM)
195/65/15 (2010+ OEM tire size 3 Yokohama AVID S33D at 829 RPM)

16" Rims (Touring Package only in US 2007-2009 magnesium rims, different in UK & EU?)
195/55/16 LRR US (2007-2009 OEM USA Touring package size for Bridgestone Turanza EL400-02 at 854 RPM)

17" Rims (2010 model 1229 aka "package V" alloy rims, 2011 "Plus Performance Package" alloy rims)
214/45/17 US 2010 could have one of two OEM tires Turanza EL400-02 at 845 RPM which is NOT LRR at that size or Pilot HX MXM4 which is LRR (GreenX) at 844 RPM