I was just wondering whether most people calculate their mileage by hand or read it off the screen. I drive a Prius and find that there is often a difference. Thanks.

 

I just use the display because, frankly, it isn't a huge deal to me. People who keep track of this and compare for the Prius notice that over time the difference is usually about 1 MPG with the display being more optimistic. That's usually about a 2% error.

P.S. I fixed your signature. You forgot to put your car ID in the code

 

I do both.
In the database I enter the calculated. But if you look in my notes I always identify what the screen said as well.

 

Same here.

 

Calculated. In as much painstaking detail as I can get :-)

I use the lifetime odometer reading and petrol receipt to three digits after the decimal as my raw data.
It actually never occured to me to use the screen, since I have always considered it something of an estimate -- just like the gas gauge. I'm now sorry that I didn't track it, so I would know how much it varies by. I'll start up if the question starts to bug me.

 

Hi All:

___I always use total gallons from the pump and miles from the odometer for all my tank entries. The odometer is your legal mileage and the pump is a legally regulated system. It may not be exact but filling up from almost empty to the filler neck for years, most pumps match about what I would have expected … The various FCD displays in our automobiles are not a legal record by any stretch.

___The following is in regards to AH NAVI owners as that is where the whammo-dyne TC/FCD display is

___I know Tink uses the NAVI units calculations in his AH saying it is more precise then the odometer and I tend to disagree. Unless the NAVI unit is completely booted and has a hold of at least 3 solid birds, it doesn’t know where the heck you are all the while you are tooling down the road. I see this every morning when leaving my home and for that first 45 seconds or so, it is registering me in the middle of my subs lake or on the other side in another section. I have breadcrumbs enabled and see the error induced with a short amount of travel through my subs lake and fields etc. When leaving work on a non-digitized road, the breadcrumbs are stacking up and although they do overlap, they are also spreading out. In other words, they do not fall on top of one another even though I am traveling down the one and only road available … This is not just the std. < 15’ of error that the GPS units hold but real error of maybe 20’ at its worst after tens of entry/exits. I will take a pic someday to show this. Whenever you do not follow a particular pre-planned route, the NAVI will override your true travel route for a few seconds with its built in/programmed in “Snap to Road” feature and once it finally figures out you are going straight instead of following its direction as an example, it will “Snap to Road” (the one you are really on) and you continue on its merry way. Whenever the GPS is slightly off, it may have you traveling on an access road vs. the Interstate. I have only seen this once so far but again, the GPS unit finally figures out that you are 15’ over and it snaps to road (the Interstate) and once again adds another small bit of error. The GPS units are to imprecise even if the digitized roads are dead on imho. The Honda OEM NAVI’s are tied to the car in the case the GPS unit loses track but only for a short period. I do not think the system takes the track into account when it supposedly does have a lock or is originally seeking one during or just after boot …

___I also keep a log of the FCD’s vs. the actuals on the cars that have them but only report the legal calc’s myself. The Insight’s FCD was very close to its actuals on almost every tank with usually < a .5% delta between the two. With a vapor recovery fill, it would show a huge discrepancy to the downside and to the upside on the next regular tank fill (75 mpg calculated vs. FCD = 105 and on the next std. fill would show a 105 with the hand calcs at 135 as an example). When this happened, I would average the two tanks and it would almost match up the FCD display for both tanks once again. Look at the “Achievement Record’s” tank of the Insight at just 106 due to a 2 tank averaging as an example when the record tank - actual was up around 109.0. The lmpg display of the Insight was usually dead on. Dead on overall as there was a small period when the display would update with its normal .2 or .3 jump. The LMPG was registering 92.5 mpg for a touch over 2,000 miles and my calc’s were 92.5 mpg as well. Its final trip at 103.1 over 1,032 miles brought up the calc’s from 92.454 to 92.767 mpg but the lmpg FCD display had not yet locked in 92.8 … Another 300 - 500 miles of highway would have locked it in I am sure.

___I do not know if the HCH’s, Prius II’s, Escape HEV’s, or RXh’s can be recovery filled like the Insight could be so it is really a moot point if they cannot.

___Here is a short compilation of the distance and FE errors produced over 7 tanks in the PZEV Accord w/ NAVI in case anyone is interested:

“Odometer” vs. “NAVI display”

Miles traveled

789.4 - 788.5
860.6 - 859.0
767.2 - 766.3
829.0 - 828.4
802.9 - 802.2
833.3 - 832.3
859.8 - 857.9

”Odometer and Pump calculated” vs. “NAVI display”

MPG

42.519 - did not record
45.362 - 44.3
44.894 - 44.3
44.707 - 44.0
44.374 - 43.2
45.315 - 44
46.270 - 44.7

___Because I am forcing more Auto-stops now given the traffic nightmares I have seen this week, the discrepancies will only grow given the NAVI’s inability to boot immediately while I am slowly accelerating and/or decelerating in stop and go traffic jams after a reboot.

___With all of the above, the odometer and pump is the only “legal” way to calculate one’s FE imho.

___Good Luck

___Wayne R. Gerdes
___Waynegerdes@earthlink.net

 

Morning, xcel. Great post – I’m, unfortunately, playing with an analytical handicap. And although I’m going into the deep, dark forest with a pen light, here goes:

Some of my thoughts/arguments are based on the Garmin I previously used. Unfortunately, the Honda GPS gives much less information, such as how many satellites it’s locked into. It’s amazing how quickly I forget things, but I *think* the Garmin usually had at least 9 satellite locks at any given time. What the Honda manual does say is that it has 24 satellites available, and although I know they can’t pick up all 24 at a time, I think 9 would certainly be probable – likely the same 9 that Garmin picked up.

Regarding the missed placement on the Navi map, this is more a function of a mis-mapped road, which I also have in my residential area, and a couple other places. For the most part, it’s accurate, but every once in a while…. Speaking to the Garmin again, the mileage calculation is not affected by the “jump”. The internal calculation has you at the same place it did before the jump – it’s just adjusted your placement on the visual map to something that is more likely correct. Once again, it’s figuring distance traveled, but the picture of the road may be slightly off – particularly on non-major roads.

Now my disclaimer: I *know* that the GPS is not 100% accurate. However… I also *know* that the odo is also not 100% accurate (even the speedometers are inaccurate - but calibrated to be most accurate at highway speed). And… I *know* that there are published reports of gas tank registers not being 100% accurate. And… there seems to be evidence that gas shrinks and expands in volume in connection to temperature, which can also affect accurate calculations.

Basically, it comes down to this: we can’t know *exactly* what we’re getting. Regardless of what you choose, you will be close – even, I would say, within a statistically acceptable claim. But what mechanisms do you trust? Considering the above paragraph’s arguments, *I* just trust the Navi.

As far as what is the legal standard, you have me on the odo. However, this post might get a bit more heated were I to discuss other “legal standards” that are suspect….

Disclaimer #2: Navi is a *bit* more optimistic and that could have some influence on my opinion. I would like to think I would argue the point if I was simply a walking pedestrian, but there’s no way to know.

At any rate, you make a good argument and I’ve enjoyed the debate. Thanks.

 

Am I correct in thinking that the odometer is based on wheel revolutions ?

If so, how much difference does underinflation and overinflation of tires make in the calculation ?

 

Hi Tink:

___I use (2) Sirf IIe/LP chipset based GPS receivers (CF and a Serial based version for 2 different PDA’s) with Mapopolis as my SW of choice. You need just 3 birds for an initial track but 4 will give you a lock. 9 can be seen most of the time. The 2 GPS receivers I use can see 9 birds in view at any given moment with locks as low as 27 - 29% most of the time? I have not watched the SAT screen much since I initially began GPS NAVI ownership so my numbers might be off by a small amount? Bad weather, tall buildings (traveling in the inner city), and tree canopies can drop both the number of birds seen and their signal strengths in my experience.

___Back to the Honda’s NAVI. It is far off before a lock just as all GPS receivers are. My home is on a digitized road but when it first boots, it places me all over the subdivision, in the lake, in the fields, etc. Another way you can generate an error is by not following a pre-planned NAVI route. Plan any route on a highway near you. Watch your NAVI display and instead of taking the path it wants you too, do the opposite and watch the NAVI take you down the ramp you did not take for ~ 10 seconds before it begins a recalc and it snaps back. Another way to see an error generated. Drive down any non-digitized road for over ½ a mile w/ breadcrumbs enabled and drive back. Do this a few times. My place of employment is un-digitized for a particular reason thus I see it every time I arrive at work. The track as seen by the bread crumbs is scattering and yours will as well.

___My thoughts anyway …

___EricGo, your speedometer/odometer is based off wheel rotation. With < 20,000 on the MDX and at differing pressures from 32 - 40 #’s, the Ranger at 50 #’s, and the Accord at 50 #’s, my Casio and Compaq based NAVI solutions speed, lat/long, and elevation display screen matched basically dead on with the speedometers of all 3. 60 mph at 60 mph. The Insight’s speedometer was ~ 1 mph higher then that of the NAVI unit when running 55 #’s at 60 mph. As the tires wear from their original 2 + ft. of diameter, (9 - 11/32 of tread towards 2/32’s), you will see the odometer track slightly higher miles traveled and higher speeds from the same original non-worn tires distance and speed as a comparison. I think the manufacturers cal the odometer/speedometer to ½ the wear of the tires but this is a guess?

___Good Luck

___Wayne R. Gerdes
___Waynegerdes@earthlink.net

 

Thanks everybody for your input and Jason for fixing my signature. Since my original post, I have seen a correction in my mileage. When I posted, I had about a 10 mpg difference between my calculated and screen mileages. On my next tank, I saw the reverse. I talked to the Prius technician at my local dealeship. He said that the Prius has a bladder inside the gas tank, which may be the cause of this difference. The bladder might have a fold in it or something else that does not allow me to use its entire volume, which can cause the pump to think that the tank is full when it is not. This can show up from tank to tank, but over time will cancel itself out.