The Stovebolt Site Forums The Parking Lot Big Bolts calculating speedometer offset

Adding the 700R4 to the 48 2 ton was bound to produce a speedometer error since that trans was probably set up for a 3.something rear.
But the difference is so much that it is presenting some problems in measuring it.
Tried it today on the town roads with a GPS speedometer app on the cell phone. It looks like 25-30 mph high. To get it really accurate so I can order the right part I need to get a long flat road to use mile markers and a stop watch. This is New England, that only exists on interstates, limited access highways etc.
However the speedo only goes to 80. So I don't want to peg it lets say I hold it on 75. The real speed will be 50 or less. That's going to be really hard to go between 2 exits without being run over!
Looks like the the other 2 measured speeds will be even slower so I still have to find a flat mile on a back road somewhere,

Theoretically, shouldn't your odometer and speedometer be off by the same factor? Why not just drive a fixed distance regardless of the speed, monitor your odometer, and do the math from there? Which would you prefer to be more accurate, the speedometer or the odometer? Do you have reason to believe that they're out of whack with one another?

Even so, I trust my Garmin(a modern one) enough to give me an accurate instantaneous speed measurement. There is about a second of hysteresis on my GPS, but if you're maintaining remotely constant speed, I think the lag would be negligible. I don't think a stopwatch would be the way to go.

On second thought, I do have a watch I would use. It's also a Garmin GPS, however.

You are on the right track with trying to hit "75" on the speedo to get a reference for calibration. Getting the full-scale reading always reduces error, assuming the linearity of the transducer is reasonable. Otherwise, you'd want to get a reference where the speed would matter most to you, like "55" on the speedo.

That's a good point. Using the odometer makes it unnecessary to hold a constant speed so no need for a flat road.
The only thing it wouldn't correct for would be any error between the speed and mileage due to friction,wear or needle misalignment.

True. The amount of torque generated within the speedometer can't really fall out of calibration. It's same principle as an AC induction motor, but kind of in reverse. More speed means more slip between the magnetic field and the rotor, which means more torque. I doubt the magnets or the rotor have changed significantly over time.

The response to the torque, and the subsequent needle deflection, would be dependent on the spring and friction. Corroded springs or pivot points are probably the biggest cause of speedometers getting out of sync with odometers. You may not be suffering from any of that, though.

If your odometer and speedometer don't agree, then the speedometer is already degraded. If you were to calibrate based on speedometer readings, I would fear that the speedometer would degrade even further and your accuracy would be fleeting. I would calibrate based on the odometer, then carefully clean/lube/refurbish the speedometer parts if it still seems inaccurate.

The speedo was stripped, cleaned and inspected.
That said, magnet degradation and loss of spring tension could not be measured easily. There is also needle positioning.
The last would be a constant offset while the first two will show up as a percentage variation.
I found a local road with a 2 mile post marking. I will run that against the odometer to get an over ground error. There is no way that can be used for a speed reference.
Then I will try shorter runs at slow and fast speeds against the GPS and see if I can get enough consistency in the readings to determine if the speedometer tracks the odometer.
If I have to choose, the correct speed is more important than the correct mileage.

Another way to do it, if you have the ability to turn the outputs on each transmission and observe the output of the speedo gear and assuming the old transmission was accurate with the speedometer, is as follows:

Make note of the output driveshaft position (example turn the yoke vertical) and use a pen to mark the speedometer output. Carefully turn the driveshaft through 10 complete rotations while counting the rotations of the speedo output of your new transmission. Estimate as closely as possible any partial turns of the speedometer output during the 10 rotations on the driveshaft. (example 10 driveshaft turns produced 5-7/8 (5.875) turns of the speedo output.) The more turns you do the more accurate you can estimate the ratio, just remember to use the same number of turns every time you count.

repeat this several times to determine an accurate ratio between the driveshaft and speedo output. Sometimes it helps to have a helper slowly turn the driveshaft through the 10 turns while you count the rotations of the speedometer output to simplify concentration on the task at hand.
Repeat this for your old transmission. (example 10 driveshaft turns produced 3-1/4 (3.25) turns of the speedo output.)

Have a ratio adapter made that has the ratio of the numbers you counted from the old trans divided by new trans. (example from above 3.25/5.875=0.553). This works if the only thing changed on the truck was the transmission. Your miles per driveline rotations will be the same, the only difference is the gearing in the new speedometer drive isn't calibrated to the same driveshaft rotations per mile. The ratio adapter effectively cancels out the error and makes your speedometer see the same driveshaft rotations per mile that your old transmission indicated.

Another way to check (roughly)
With the trans in drive (OD locked out) I should be driving 1:1 +/- converter slip (converter lockup not wired yet).
That should be the same as the original trans in 4th.
If I could find the stock speed at 3500 rpms I could calculate the ratio difference. present reading 3500=67mph.
8:20 tires (new)

First Odometer test gave me a an error of 1.5 almost even.
Will repeat again both directions to confirm the fine points.
That would make 75 mph indicated = 50 mph over ground.
pretty much my first estimate.

4 more 2 way runs on the measured stretch all gave me exactly 1.50 to 1.
8.25/20 tires.
But now the speed spread sheet on another thread doesn't agree with that.
were stock '48 2 tons running 60 mph at only 3500 rpms in the day?

Well many of the catalogs I've seen have a LOT of variance in the revs/mile for a given tire size. It could account for a few mph when it's all said and done. I was pulling my numbers from the Cordovan catalog:
http://www.cordovantires.com/catalogs/truck.pdf

Or adapt a GPS speedometer to your dash.
Sorry, had to throw that out there as it IS a viable, affordable option. I bought one from Speedhut a while back. Except for long drives through tunnels, it works perfectly.

A gear/speed/RPM calculator will help answer that question.
I like this one
http://www.onlineconversion.com/bigger_tires.htm
6.13 gears and 35" tires give 3500 rpm at 60 mph, if that is the original tire diameter then yes.