So I have an odd question. On my 84 I installed front knuckles and wheel bearings from an 87 when I did my C5 brake upgrade. I then used an inductive prox switch to see the OEM tone ring instead of the OEM abs sensor. That prox feeds the signal to my speedometer. I did this because I didn't want to tie into the VSS sensor and split that signal going to my Holley ECU. So I realize because it's seeing the pulses from one front wheel, when I turn it will lose some accuracy (outside wheel will turn faster). The question is how much? So far I've used a GPS and driving on the highway the speedometer shows within 1 mph of the GPS. But I haven't tried this on sharp turns. So on a hard left turn the speed would show higher than a right turn. But how much difference would there be?

Here is a pic of how I mounted the prox.


I've had this system on the car for a while now. But the first prox I used had a lot of problems and the speedo only worked intermittently. So I recently upgraded to this brand and it has been 100% so far.

I have three of these on the car. One on both front wheels and one on the driver's side rear. The passenger front is for speedo and the other two sense traction loss.

Now I realize you probably don't need to know your exact speed while hitting the apex of a turn. But it would be nice to at least know how far off it could be. I'm not looking for exact numbers here. I understand there are too many variables for that. Just a ballpark.

 

If your speedo is accurate to within 1mph compared to GPS you're I believe better than most! If it's reliable then I say you don't worry. Do you have an odometer?

 

It has an odometer. It's the Intelliteonix speedometer M9222. It works well with either NPN or PNP pulses but can't see a weak sign wave very well like what comes out of the stock abs sensor. That's why I went with a high speed prox.

As far as accuracy it seems good for highway driving. But I didn't want to try driving hard into some curves while looking at the GPS speed on my phone lol.

 

I sure theres a Brainy answer with X & Ys and PI but maybe to see the amount of distance gained/Lost might be as easy as a line on the tire and ground Travel 1 or 10 rotations in a straight line measure have a base distance. then with both tires marked do the same again turns measure again each track and see how each measure is different is it a Huge Feet or Ainol Minimal inches difference

 

Once it's been validated as being correct I'd think the appropriate position for the phone would be in the 'pocket'?

 

If we consider say a 30 mph corner at 0.8 g, this gives us a curve radius of 75 ft. The front track width is 5 ft. So the curve inside wheel is travelling around a 72.5' radius, while the curve outside wheel is travelling around a 77.5' radius at the same lateral acceleration (0.8 g).

V cg = 30 mph = 44 ft/sec. g = V^2/r.

25.81 ft/sec^2 = V inside ^2/r inside and = V outside ^2/r outside.

V outside = 44.727 ft/sec = 30.495 mph
V inside = 43.257 ft/sec = 29.494 mph

So about +/- 1/2 mph error depending on the curve direction. It will get worse at lower speeds and tighter curve radius and better at higher speeds larger curve radius.

 

Never mind. Tequilla boy's calculation is much more scientific. And accurate.

I came up with a 12% difference between the front wheels in a 50' radius turn for the outside wheel. <1% in a 1,000' radius turn.
The tighter the turn the greater the difference. (I ignored tire slip simply because I have no idea how to calculate it.)

 

Can you not use the Holley's speedo output, or configure your own PWM output table to give you what you need?

For the math, if you're doing a U-turn at full lock (I'm using a 22 foot turning radius and a 5 foot track width as an example, you're looking at 11% slower than the actual vehicle speed for the inner wheel and 11% faster than the actual vehicle speed for the outer wheel.

((Turning Radius Of Wheel / Turning Radius of Car) - 1) x 100 = % variance.
((24.5 / 22) - 1) x 100 = 11.36%

 

I could have. But I ran out of outputs. And a $30 prox switch was a lot less expensive than a Holley add on I/O module.

 

Wow thank you. That's way less drastic than I thought it would be.

 

Unless you are only driving on a race oval (one-way turns), ignore any difference. Any short-term errors will be cancelled out over the combination of right and left-hand turns. It is 'much ado about nothing'.

Getting the speedo to agree with the GPS within 1 mph is quite an accomplishment!

 

Well I should say it was within 1 mph on a reasonably straight highway. No sharp turns. To calibrate that speedo you need to press the calibration button then drive one mile and press it again. While you're driving it shows how many pulses it's seen from the sensor. So I cheated. I figured out how many pulses I would have in a mile based on tire diameter and the 47 tooth tone ring. Then I just jacked the wheel off the ground and spun the wheel by hand in my garage until I met that number on the display. I figured I couldn't get it that accurate by driving a mile.

 

So now you just do a some random GPS speed confirmations for the speed and find some mile markers to check the odo?