[Bob Lincoln]

When I bought my 1993 Daytona in June 2009, it came with crusty/rusty 15-inch rims and worn-out tires. The factory door jamb decal said it was built with 14-inch rims. You can't trust that, because my 92 Daytona came new with 14-inch, and the decal said 15-inch.

I swapped in my 14-inch rims and good tires from my 1992 Daytona, which I was preparing to scrap.

The speed sensor gear that came with the car was the 19-tooth yellow gear. I was getting an error, so I swapped in the 20-tooth blue gear from the 92 Daytona, and this gave me a -0.5% odometer error, which was good.

But then the 14-inch rims developed pinhole air leaks from rust, so I bought all new 15-inch aluminum alloy wheels (N/A in 14-inch) and new 15-inch tires. Since then, I've had an odometer error of -2.2%, with the speedometer error being zero (as measured with GPS and by timing mile markers while driving at indicated 60 mph).

Hoping to reduce the odometer error, I put the 19-tooth yellow gear back in yesterday. Today I drove the car with GPS and got the following: +2.8% odometer error and +7.7% speedometer error.

OK, I can understand that the odometer will err the other way. But what I don't get is how the speedometer error does not proportionately track the change in odometer error. The speedo was dead on with the 20-tooth, and with the 19-tooth, it's reading 40 mph at actual 37, and 70 mph at actual 65. This is confirmed by GPS and by comparing known tachometer readings for a given speed.

It looks like I should just put the 20-tooth gear back in and cut my losses. But does anyone have any thoughts as to how the speedometer error does not track the change in odometer error? Both get their input from the Hall effect sensor triggered by the same gear turning in the transaxle. The speedometer uses the digital pulse converted to analog for the analog gauge. The D/A error should be constant and very small.

I guess the best I can do is the -2.2% odometer error and the accurate speedometer that I have with the 20-tooth gear.

 

[John Wood]

As far as % error between the odometer and the speedometer, they should be similar, but there is probably an offset in favor of the speedometer reading higher. Most speedometer makers design the analog needle type speedometer to read a little on the high side (maybe 1 to 3 MPH at 60 MPH). I guess that is to cover their butts for any one who gets a ticket and blames the instrument .

With the newer electronic transmissions and programmable pinion factors, you can get a little more precise on the odometer. Swapping the pinion gears on the older trannies is still done on a best effort basis with the bias (from most speedometer shops) to set the speedometer to read high, knowing that a lot of people are going to exceed the speed limit.

 

[Bob Lincoln]

Well, that won't help here. No programmability, all I can do is swap pinion gears. As far as the speedometer offset, its error should track the error of the odometer. If a gear change induces a 5% change in the odometer, it should also induce a 5% change in the speedometer, not 8%. I have no explanation as to why or how the offset can increase between the two.

 

[bamman]

It could be that the speedometer is just reading wrong. The best way to confirm this is the use a scan tool to see the vehicle speed reported by the speed sensor versus the actual speed displayed on the analog cluster. I know in my case there has been around a one to two MPH difference.

 

[Bob Lincoln]

I checked it against GPS with both gear sets. It was dead-on accurate with the 20-tooth blue sensor, and is now reading high by 7.7 to 8.1% (depending on what speed I measure it at, call it 8%) with the 19-tooth yellow gear. Regardless, it doesn't explain the *change* in offset. Distance was off by -2.2% and speed by zero; now distance is off by +2.8% and speed is off by 8%.

 

[John Wood]

An analog speedometer uses a DC motor that applies torque against a return spring. I found a pretty good writeup from an experimenter who calibrated a speedometer to the odometer (to get them in line) in his unit by using a shunt circuit. He was actually putting a higher reading speedometer in his car. The article will take a long time to read but may give you some ideas. It is the return spring that determines how far the needle will move versus the torque on the meter movement motor. http://www.sccoa.com/faq/SpeedometerConversions.pdf

Chrysler did have a couple of different meter movements on the EEK models of cars that I am familiar with (and there were probably 3). My 87 LeBaron had a 125 MPH speedometer (electronic) and the 91 model cars had 120 MPH heads. In know that the speedometer motors from those models readily interchange, even if the housings are different (I've swapped a 125 MPH movement into a 120 MPH unit). I believe I have seen 145 MPH speedometers in some models in the bone yard.

 

[Bob Lincoln]

Thanks, but since the odometer will still read over or under, I'm going to put the 20-tooth gear back in and at least get the speedo right on target.

 

[bamman]

I checked it against GPS with both gear sets. It was dead-on accurate with the 20-tooth blue sensor, and is now reading high by 7.7 to 8.1% (depending on what speed I measure it at, call it 8%) with the 19-tooth yellow gear. Regardless, it doesn't explain the *change* in offset. Distance was off by -2.2% and speed by zero; now distance is off by +2.8% and speed is off by 8%.

What I was trying to say is you may want to use a scan tool to verify the speed indicated by the VSS. For instance the VSS may be telling the speedometer to read 65 MPH. The speedometer may reading 70 MPH. GPS may confirm 70 MPH. Odometer still records as if it is going 65 MPH.

Just a thought.

 

[Bob Lincoln]

Yeah, but there's nothing I can do about that. If I put the 20-tooth gear back in, then the speedo is dead-on accurate. The odometer will just be -2.2%. With the yellow gear, I get both the speedo and the odometer off.