I purchased a mobile GPS navigation unit (Pioneer AVIC-S1) for my dad for Christmas, and played around with it for a few days myself. I noticed that my speedo reads high by about 5% when monitoring the indicated speed on the GPS screen. I'm sure it is much more accurate than the OEM speedo, so I'm curious if anyone else has done the same.

I'm also about to install new rear tires that are about 3.5% shorter than OEM (255/35-18), so that will add additional error. I don't want to have to mentally adjust the speedo reading by 8.5% to determine my actual speed, so I want to add a speedo correction device. Has anyone done this successfully on the 2IS?

I used this on my S2000 when I installed 4.77 gears, and it worked perfectly.

 

I forgot to mention that I also want to correct the reading so my odometer doesn't read 8.5% higher mileage than I've really driven.

 

I also forgot to mention that the correction might even raise the speed limiter. Not that it really matters ...

 

I would believe the speedometer before the gps unit. As far as a speedo correction device goes I doubt that you could get one to work on the IS, I don't know for sure but I would bet that the Is doesn't have a dedicated speed sensor, it probably takes an average reading from all of the wheel speed sensors and the trans speed sensor and then the corresponding module sends the info to the speedometer..this is just a guess but is probably pretty close.

 

My speedo reads fast also (comparing the speedo to the local school radar 'YOUR SPEED' readings.) This is also with tires that are slightly larger in diameter (265/40/18's.)

I read an article in one of the car rags (I think it was the Technical Correspondence in C&D) where they say many speedos are off slightly but the odometer is usually very accurate. Didn't make much sense to me but that's what they claimed.

However, by going with much smaller tires, you're gonna have to figure something out.

 

Why would you say? Factory speedos read high by design to prevent people from speeding unknowingly. It's the same reason that drug test kit manufacturers would rather give false negatives than false positives.

Regarding the sensors, I agree that it could be possible that the speedo averages all 4 wheels, but that would be fine too. I would just need to get a yellow box for each rear wheel (hopefully). However, my S2000 had speed sensors on each wheel, but it still used the tranny sensor for the speedo. I'm pretty sure the speedo in the IS doesn't shoot to 30+ MPH when you do a burnout with VDIM turned off.

 

I would consider those road side radar devices as pretty unreliable too. First off, they measure closing speed (toward the device), not the actual speed. That is equal to the actual speed times cosign of approach angle. Secondly, they are also going to err on the high side to encourage slower speed.

Regarding the odometer, I wouldn't be surprised if the OEMs intentionally calibrated the odometer acurately than the speedo, since that would benefit them on warranty. If so, screw them! I want my speedo to be accruate! If making my speedo accurate gives me a few extra miles of warranty, that's their friggin' fault!

 

I thought the same but there's a set of them on the way to the range and I get caught between them, approaching 1 and leaving the other. They both read the same (as I can see the one behind my in my rearview.)

Can't argue with you here.

 

Not really sure what you mean, but they operate off of Doppler shift, which is totally vector dependant. It was actually one of my senior year final exam math problems in an RF class. If they both read the same even though you are shifted over a lane, then you are probably seeing the 2nd one when you are further away (similar vector angle).

 

The speedo being off doesn't surprise me. I did a little test while driving on the freeway today. I had my cruise control set with the speedo reading exactly 80mph, then I reset the average speed indicator and it read 76mph. Did this a few times with the same results each time.

 

My speedo also reads high by about 5%, as measured by the speeds shown on the speed monitoring S.M.A.R.T. trailers that the cops around here use.

I've found that most of the cars I've owned read a few percent high, but none were as far off as the IS, even the '80s vintage mechanical ones.

 

I have an Alpine Blackbird GPS unit for my company vehicle which displays my speed. I used it once in my IS just to see the speed difference and at about 70mph the speedo reads about 1-2mph faster than the GPS readout.

 

The speed is derived from the four wheels. The transmission input is ignored. If you are putting smaller tires on the rear (smaller diameter) you're going to really **** off VDIM. How much I can't say without testing, but VDIM is constantly comparing front/rear and left/right wheel speeds and deciding what (if anything) to do about it. Sounds like a slightly expensive experiment if it doesn't work.

FWIW, I've pulled apart combination meters from cars I've owned and adjusted the tension spring on the speedometer needle to get parity between indicated speed and actual speed. Keep in mind, tire wear will affect this, since you typically lose about 0.5" (12mm) of diameter in a tire from new to worn out. Depending on the original diameter, this may be very significant. I am not sure if this will work on the a 2IS because many of the newer gauges use digital stepping motors, not analog devices so they do not have return springs. It is possible however, that one could remove the needle and make small adjustments by putting it back at an altered zero point.

I've only had one car that came from the factory dead on the money - my Supra. All the rest have indicated high. I measure this using the mile marker and stop watch method. Since the cruise is rated at +/- 1mph, I set the cruise control for a given speed, then use a stop watch and time anywhere from 10 to 25 miles, then do the arithmetic. The longer the distance the less the precision of the start and stop events matters. In any case, I've been on long road trips enough to know the odometers typically indicate 2% less than actual mileage (depending on tire wear), and speedometers tend to be about 3% higher than actual speed.

Gernby - to answer your question directly, there is a signal from the skid control ECU (underneath the HVAC blower motor) that goes to the combination meter on pin 17 of J34 (the 24 pin connector.) It is not packaged and sent over CAN, most likely because this signal is used by the odometer and anything affecting the odometer gets special treatment, and usually a dedicated circuit because falsifying an odometer reading is serious business.

What would be ideal is to interrupt the speed signal to the ECM and limit it to 142 mph to thwart the top speed limiter. But that's a different egg to crack. (Speed signal exits the combination meter on pin 5 of J34...). Also FWIW, here is a list of all the things you are going to irritate by interrupting the speed signal.

 

One other useful detail for those curious minds out there:

 

I've seen plug-and-play fixes via OBD that will do this, not sure about the availability for Toyota/Lexus. The car KNOWS how fast it's going, but will tell the speedo to display a higher figure.

From what lobux says, looks like the process will be a bit more complex than just plugging in a unit underneath the steering wheel. Probably not worth the extra effort.