twohoos

So, at 61K miles I switched my dash over to km to catch the imminent 100,000km rollover. When the big moment happened I stopped for a quick cell phone pic, and immediately switched back to "real" mileage. Out of idle curiosity, though, I checked to see that the numbers agree: 100,000km is precisely 62,137mi after all. So I was a little bemused to see my odometer displaying just 62,112 miles - fully 25 miles lower than it should be (or equivalently, the metric reading was 40km too high). Now I know full well that neither of those numbers is "right" (Honda odos are notorious for over-estimating mileage, but I've made my peace with it lol). Still, I thought they'd at least be *consistent*!

Now, after a bit of thought about how the odo/speedo work, I decided my discrepancy *could* be normal (i.e. a "design feature"). But, my car has had some "issues" that might have affected the odo, so I can't be sure just yet. To check my theory, I need your help.

Could you kindly check your odo and post the following info here:

1) Model year and type (AP1/AP2)
2) Odometer and drivetrain: ORIGINAL* or MODIFIED** (see below)
3) Main odo in MILES and KM (no rounding! that's the whole point lol)
4) Trip odo in MILES and KM, including tenths (either Trip A or B, but ONLY if it's been reset within the last 620mi/999km*** (see below)

*Post "ORIGINAL" only if your car's dash, trans, and diff are the original parts that came with the car, AND have always seemed to work properly. (Clutch replacement is ok. )
**Post "MODIFIED" if you've ever A) replaced your dash, B) installed trans/gears and used a "yellow box"/"yellow jacket" device, OR C) had any issues or repairs relating to the odo/speedo or vehicle speed sensor (VSS). You can give a quick description if you wish.
***The km trip odo rolls over when the miles odo is still in the 600's, so to make sure the Trip reading is sync'd it has to have been reset recently

Example, for my car:
1) 2000 AP1
2) MODIFIED (AP2 trans, yellow jacket)
3) 62,112 mi / 100,000 km
4) 59.3 mi / 95.5 km

rrounds

My car
1) 2006 AP2
2) All Stock
3) 118570 mi / 190897 km
4) 283.3 mi / 456.1 km
118570 = 190820
283.3 = 455.9

ROD

cosmomiller

My car:

1) 2009 RYP AP2. (Yellow makes a difference due to the Coriolis effect)

2) All stock.

3) 57,787 miles / 93,037 kilometers / 50,215.5 nautical miles

4) 131.6 miles / 211.9 kilometers / 114.4 nautical miles

5) 89 degrees F / 31.6 degrees C

57,787 miles = 92677 klicks.

Difference of 360 klicks from true calculation.

twohoos

Lol cosmo check your math
57787mi = 92999.2km

It appears the 3 of our cars are remarkably consistent: miles accrue at a smidge under 99.96% of the rate we'd expect based on the kilometer reading.

The fact that the "conversion value" seems to be the same for both AP1 and AP2 throws a wrinkle in my theory though. :/ Still, let's get a few more posts from other cars and see if the trend is the same.

AP1 folks, chime in!

gaus

Very interesting. Great thread.

I have an S2000 which was made for Germany and it does not allow me to
change the display to miles. This car was developed in Japan which uses
the metric system. Therefore I think it is reasonable to assume that the car first compute
speed/distance in kilometers and then simply convert it to miles.

Wikipedia tells us that there are 1.609344 km in a mile. I used trip meter reading provided so far in the thread
to find the conversion factor used in each car. This is what I got;

twohoos 1.61045

rrounds 1.60995

cosmomiller 1.61018

It looks like Honda is using different factors in different model years.
Perhaps in some models they went with 1.609 and others they decided to
round it to 1.61.

Needs more data. Owners in US, please give us more data points!

Trip meter includes one decimal point so it provides more accuracy than the odometer.

Gaja

zdave87

Different years had different sized tires, so that might account for the difference.

twohoos

Sorry, I shouldn't really say "conversion" factor because I don't think that's how it works: the dash computer is NOT just translating km to miles or vice-versa. Instead I believe that there are effectively two separate, independent odometers - one for miles, and one for km. (Actually, SIX separate odo's, if we include the trip counters!) The two odo's are both looking at the VSS signal, but they independently convert that to their own distance measure. Now, each conversion is an approximation since the VSS signal has limited precision; but because of the different units, each conversion has a slightly different approximation error - that's what causes the apparent discrepancy.

I can go into more technical detail later, including how the VSS signal is processed, and the legal/regulatory issues that would force Honda to have separate mile & km odo's. In other words, this is not a case of lazy engineers who just rounded off a conversion factor.

Chuck S

Remember significant digits from high school physics? The main odometer measurement accuracy is even miles, not 1/10th or 1/100,000 like the conversion so that's the accuracy.

My unaltered '06

Main 50,698/81,625
B 72.2/68.0
A 7,397.0/11,970.0

Main 99.9575% = 100%
B 99.8737% = 100%
A 99.9588 = 100%

-- Chuck

twohoos

^Thanks for the data Chuck! And yes, I'm familiar with both measurement precision and measurement accuracy.

gaus

twohoos,

Why do you believe that there are effectively two separate, independent odometers?
What are the legal issues that would force Honda to have two separate odometers?
Besides S2000 what other cars display total distance driven in two different units?

Thanks
Gaja