RX8rider

I've been keeping track of my MPG lately and I have this neat program on my Palm for that. It asks to input the starting and ending odometer reading and then figures out the total miles travelled. Well, I thought that maybe it was a mistake but now it's happened three times in a row....the starting and ending odometer readings are not consistent with the total trip miles travelled. What's going on?! I guess to my benefit though, it seems like it's about 50 miles off (less than actual trip miles) on a tank (about 210 miles). So it seems to be a little more than 25% off. I knew something was up because Mazda sent me one of those letters letting me know that I should be about ready for my 5000 mile oil change and I only had about 3000 mile on the Odometer. Any thoughts??

jaehyun127

keep on early oil changes... but i guess this is GREAT for your RESALE VALUE!!!

Go48

That certainly is an interesting observation and one that I will check out myself. That is, if this weather ever improves and I get a chance to drive the car.

I'm curious about your statement that suggests Mazda knows how many miles should be/are on you car and how that relates to your "discovery". How would Mazda know your driving practices and how that translates to your mileage over time? Maybe I'm missing something here. If so, sorry.

RX8 XTC

Damn i thought i WAS THE ONLY ONE having that Problem !!! Man i am going Down to MAZDA the 1st thing !!

hotpot

Damn, the mpg is even worse than we thought, supposing the trip counter is overstating the miles travelled.

RX8rider

I'm not sure whether it's the trip counter that's overstating the miles travelled or if the odometer is understating

Go48

The more I think about this the more I realize the huge potential implications of significant odometer errors like you state.

I'm gonna try some GPS runs in the next day or so to see if I can determine whether the total miles recorded by the odometer are reasonably accurate. I'm a little rusty with my GPS unit so I don't guarantee anything. It would be nice if someone with the integrated nav unit would try to do some testing also.

It would also be nice if one of our knowledgable members would enlighten us as to how the odometer and/or speedometer work in the RX-8. Looking at the wiring diagrams it appears that there is a "microcomputer" integrated with the instrument cluster that includes the odo/speedo LCDs, but it's not clear how and from where the data is coming from for the odo/speedo.

ÜberJumper

I'll check as well on my next fill up. Probably easiest to note the starting overall KM, reset the trip odometer, and on the following fill up, take the trip odometer, and subtract it's mileage from the overal odometer mileage, should come up with your starting mileage.

Actually, I'm going to try testing this on drive home today using the second trip odometer.

czr

iTrader: (3)

If you have access to another car you can test out to see if your 8's odometer/trip counter is accurate. Either way let us know. I don't think I have that problem and I I haven't heard of it before.

zoom44

first off the odo's get their info from the speedo. we already know the speedo is off- in fact the faster you go the more it is off. also the odo's dont show 1/10 of a mile. when you reset the trip odo the main odo could actually be say at 9/10 so its already close to a mile off. the farthur you go the more they will be different. another forum member talked about this in the past and it was confirmed. do you have the navi?

zoom44

oh yeah and it has nothing to do with that letter- they have no idea how far you drive everyday. remeber that the interval is distance or time whichever comes first so they probably send it out a little earlier than the time and miles so you have time to make an appointment and also if you go early everytime because of the letter than they get more service dollars out of you in the long run.

hotpot

Maybe this could explain the sharp variations in mpg that different owners have been recording. Faulty odometers/trip recorders?

RX8rider

No, i don't have the navi. I might just confirm the mileage with my CRV and then see what the RX8 gives me on the same trip

Go48

Did a brief GPS run today in an effort to check out the speedometer and odometer accuracy. I was able to check the accuracy of the speedometer at various speeds and it turns out the speedometer is amazingly accurate (IMO). With the cruise control set at a steady 60MPH the GPS ground speed fluctuated between 59.3 and 59.5MPH as the GPS unit updated it's location. At 75MPH, the GPS readings were between 73.9 and 74.1MPH. That's much better than I expected.

As far as the odometer accuracy goes, I discovered only that I need to be retrained on my GPS unit that has lain idle for 4-5 years. I thought I had set it up correctly, but after awhile it became obvious that was not the case. I will run it again, weather permitting, once I get my stuff together. Stay tuned.

I do have one additional data point I will mention that bears on the reported disparity between the trip mileage and the actual running mileage readings that was the subject of the original post. While I was out on my GPS run, I gassed up. My trip odometer mileage reading since the last fill up was 250.5 miles and the actual miles for the same period, i.e., subtracting the previous total running mileage from the total running mileage at fill up, was 250. So, at least in this one case there was no significant difference between the 2 odometer readings. Whether they are accurate in an absolute sense I hope to find out in the near future.

EDIT: Removed my faulty math regarding possible worst case/best case error percentages between trip mileage and running mileage. Figure it out if you must. I'm going to bed.

Go48

DISCLAIMER: I must point out that these results are for MY RX-8 only, and that these data represent only 2 data points based on fairly short travel distances. Additional runs by others may provide different, perhaps even more encouraging, results.

To add to my previous post on this topic, I made a couple of GPS runs today on the nearby Interstate to test the odometer accuracy on my RX-8. Details on the runs are provided below, but the results indicate about a 5% error in the odometer readings on the high side (i.e., the odometer is recording about 5% more miles than the car is actually traveling).

First let me point out that my previous post provided data that indicated the trip meter and odometer were recording close to, if not exactly, the same mileage readings. So in the test runs I used the trip meter readings for comparison with the GPS distance readings. In addition, I did verify, as much as possible, that the running odometer mileages were consistent with the trip meter mileages on both runs. (Remember, the odometer does not display tenths of miles, so it's not possible to determine precise consistency between it and the trip meter.)

On the first run, I logged 28.45 miles distance on the GPS receiver which compared to a trip meter distance of 30.0 miles. That equates to about 5.2% additional miles on the cars trip meter/odometer. The second run, which was also done on the same Interstate section in the reverse direction, but of a somewhat longer distance, resulted in a GPS distance reading of 35.10 miles and a trip meter reading of 36.9 miles. That's a 4.9% high reading on the trip meter/odometer.

So, about 5% high seems to be a reasonable number--for starters anyway--for the accuracy in the RX-8 odometer. What does that mean? To put it into practical terms, if you travel 260 miles on 13 gallons of gas, instead of getting 20 MPG, you actually will be achieving about 1 MPG less than that. Not good news, especially for those who are already reporting very poor fuel mileages. Also, you will reach your warranty cutoff a little sooner than what the true mileage would dictate.

On the positive side, as I reported in the previous posting, the accuracy of the speedometer is quite good, much better than the odometer. And although the odometer reportedly works off the speedometer, it appears that there is something else going on here. Perhaps within the microcomputer integrated with the instrument cluster.

Having said all that, I am not particularly surprised or disturbed by these results and will not let it affect, one iota, the enjoyment I get from driving this unique and exciting machine! :D

Go48

After thinking about the odometer test runs I made using GPS positioning, I thought I would try a more mundane test using mile markers on the same section of the local Interstate. The prelimilary results below obviously assume mile marker accuracy. Additional longer-distance runs will be needed to refine the data.

In a 20 mile run, the odometer was reading high by about .05 miles. (In other words, about half way between the .9 tenths marker and the whole number mile marker.) That extrapolates to .25% (.0025) high reading in 100 miles. If my test is valid and my calculations are correct, that certainly is significantly better than the GPS-derived error of about 5%. In which case the odo error is not particularly worrisome. I tend to believe these results over the GPS results. In summary, it looks like the odometer is recording about 1/4 mile high in 100 miles.

On a related "issue", I have not noted any significant difference between the trip meter readings and the odometer readings as suggested in the original post in this thread.

Paul_in_DC

Also if you have your car registered with mazdausa.com, they'll have an idea of your current mileage and mileage over time.

Paul_in_DC

A question on the GPS experiments... is there a possibility that your car and the mile markers report actual driving distance, while the GPS reports straight-line (as-the-crow-flies) distance? That could easily make for different readings if the roads you're testing on aren't arrow-straight. Last time I was in Pennsyltucky ( ) the roads undulated somewhat. Terrain and altitude changes could make a big difference.

Go48

I thought about that possibility, but decided that was not the case. Here's why. The GPS receiver updates every second or so and it's my understanding that the segment distances between the update positions are calculated by the receiver and added to previous segments to arrive at the distance traveled. My guess is that the apparent "large" GPS-derived error is due primarily to an accumulation of minor errors inherent in the GPS-generated positions, particularly those that are acquired from a moving vehicle.

That said, I would entertain an alternative analysis from someone more knowledgeable than me about all things GPS. I am certainly not an expert in that field.

MX6_2_RX8

Going back to the original problem, is it possible that the trip odometer was reset at some point because of a low battery and cold temperatures? The main odometer is unaffected by battery disconnection but my trip was reset every time I had a battery issue before it was replaced. Of course this doesn’t make sense if the trip is reading higher than the total. I couldn’t tell from the original post which was higher. Just a thought.

I am happy to report that after 1400 miles I have gone through an entire tank of gas with no problems. I’d just like to say, it’s about time. Anyway, for the first time I got to calculate my MPG: 18, better than I was expecting.

Go48

This is what is said in the original post in part: "I guess to my benefit though, it seems like it's about 50 miles off (less than actual trip miles) on a tank (about 210 miles)."

I interpret that as saying the odometer mileage is 50 miles less than the trip meter miles. I've been monitoring my odometer and trip meter miles for a couple of fill-ups and don't see any disparity between the two. I can't imagine there would be a 25% difference between the two, since it's my understanding that both the odo and the trip meter get their data from the same source. So I have no idea why he is seeing this disparity.


Yeah, I have nearly 3K miles on mine and the overall average for that distance is 19.75 MPG. Segment averages range from just over 18 to nearly 22. See the attachment.

Slims8

As I've always been told : "You can measure it with a micrometer, mark it with a soap stone, and cut it with a blow torch."

It all depends on how accurate you want to get vs. how far you are going, vs the accuracy of what you are measuring it with. But, the bottom line is if your speedo is correct but your odometer isn't... you have a weird problem. They are supposed to be linked.

MyRxBad

iTrader: (1)

You believe some software for the Palm that some person wrote of years of engineering? I wouldn't. You program probably doesn't have the resolution of the odometer, or it is rounding up!

Either that or it's human error in the readings. Double check your readings and get back. Also you should be checking the accuracy of your Palm program w/another car as well.

Go48

Yeah, that's my understanding, too. In my initial runs using GPS, the ground speeds indicated on the receiver were remarkably close to what was being displayed on the speedometer. The data from the latest odometer experiment using mileage markers is certainly more in agreement with the speedometer accuracy indicated in the initial runs.

Paul_in_DC

Filling up the tank reminded me to check my own numbers, so here goes. I've recorded every fill up since I bought the car; trip odometer miles for that tank and gallons of gas to fill it up.

-- Total of all trip odometer readings = 3,376 miles.
-- Miles on the car when I bought it = not certain, but around 10 miles.
-- Current odometer reading for my 8 = 3,386 miles.

So for my car, it appears that the trip odometer and the "total" odometer are in agreement, i.e., the sum of the individual trip odometer readings = the total odometer reading.

Now... whether the odometer is accurately measuring 1 mile as 1 mile, dunno yet, but if I remember next time I hit I-95 I'll check it against the mile markers.