[cre]

Get it on a dyno and REALLY measure the speed; While you're there get a comparison on the tach readings.

Otherwise, you need to just start measuring things; pull the speedometer drive gear and cross reference the part number and find out what you REALLY have in the car. There are too many variables. If things are so far off your car's had something swapped and possibly more than one part.

You can't base any conclusions entirely on the ECU's response in most circumstances in my opinion either... Leaking capacitors, 5% tolerance resistors... skewed tachometers and speedometers...




There are two different types of offset you will see with the stock speedometer: linear and logarithmic.

I'm not going to cover how it all works... the short and simple of it is that if the speedometer is off by a constant amount regardless of speed it is due to one of the following: tire size, end gear ratio, transmission gearing, speedometer drive gear.

If it is off by a percentage it is the speedometer (the clock/tension spring inside can be used to adjust this to a degree, but it will never be accurate over the entire speedometer's scale... choose the rage where you want it accurate and accept that there will be some shift at other speeds).

If both are off it's not fun to calibrate... best is to start with physically identifying components such as the speedo drive gear and counting wheel revolutions to ID the diff.




I'm not guaranteeing the numbers in that chart are correct, but they're the output from a spreadsheet based on very simple math... if they're off I'd be very surprised, but it is possible.