Speedo Correction

[Blacky]

To get an accurate calibration, "real speed" or "actual speed" MUST be measured independantly of the vehicle speed sensor on the vehicle.
i.e. you MUST use a radar gun, or GPS or some other external speed measuring device.

There are two factors that come into play here:

1. The VSS pickup may be calibrated incorrectly in the PCM and the dealer or whoever last worked on the car may have "corrected" the problem in the cluster - instead of in the PCM. So if you correc tyour PCM, the cluster will be read wrong.
or
2. The VSS is correct and the cluster is incorrect.

You can't really tell which without an independent measurement.

The best you can do without an independent measurement is to measure the distance travelled by one complete tire revolution (or 10 or 100 would be more accurate if you have the space, then divide by 10 or 100) to determine the rolling circumference of the tire.

Then plug the values into EFILive's speedo calculator and let it figure out the correct pulses per mile.
You should be reasonably confident that the scanner will now be reading the correct mph - regardless of what the cluster is showing.

Then you should be able to adjust the speed displayed by the cluster without changing the actual pulses per mile by changing the speedo output pulses calibrations - hopefully your vehicle has that cal. i.e. H0105 in the LS1 cals.

Regards
Paul

[kbracing96]

Ok, so I jump in the middle and do a little thread whoring, lol. My scantool speed reads dead on my GPS, but at 80MPH, the cluster speedo reads 82. Is there a way to change just the cluster reading to match the Scan tool?

Edit: after reading a little more I believe I have come to the conclusion that I should be able to take 80/82=.9756 Then * that by 4000 (H0105) and come up with about 3900. That should fix my speed difference without affecting the scantool speed, right? Also, if H0105 is the primary output rate, then what is H0106, the secondary output used for? (just wondering)

[wait4me]

Tech 2 would be needed. That or just change the speedometer pulse output to fine tune the speedo cluster only. That is H0105. But, it is always better to just change the H0101, And just let the pcm be 2 mph off at 80. That is not very far off in terms of gm slop. .

[hquick]

Hey KB...whore away.
I ran a couple of 1.5 hr logs today during a long trip. Freeway as well as winding hills.
On the way home there was a speed check area...where the display is mounted above the overpass. I don't know how acurate they are but I was the only car on the road and at the point where you check the speed, I was spot on 100kph via my speedo. The display on the speed check sign said 95-6 as did the scanner.
I think I'll leave it. I'm happy for the cluster to be reading 4 k's over. I'm less likely to get speeding tickets that way! Now, if I can just make myself forget it reads 4k's fast!

[gto_in_nc]

Do the speedo and odo use the same pulses? If so, a fairly precise determination of the variance can be determined without radar or GPS if a measured mile is handy.

[Blacky]

Yes, as far as I know they do use the same pulses. Maybe some vehicles exist that use the ABS sensor(s) for odo - but I doubt it.

The trip computer "distance travelled" would be an option as well, since it usually has 2 decimal places of accuracy.

Regards
Paul

[Biggsy]

I know holden do tend to make their speedo's show 1-2% higher reading than actual, even on their earlier electronic speedo's (unless you have an ex-cop, their speedo's got calibrated regulary)
I have seen brand new cars showing 5% higher than actual!

There is a PPK (pulses per km) setting which they adjust the speedo with through the tech 2

Cheers,

[gto_in_nc]

thanks, guys! that's perfect and just what I hoped/expected. (The intentional miscalibration being a litigation avoidance measure, I suspect.)

I think I would be very hard-pressed to measure a tire circumference to within +/- 0.5% and, while +/- 27 feet over a mile is not GPS-spec, it is pretty darn good. Plus, I have a measured three-mile nearby, which makes it like +/- 0.17%. I can live with that...

[hquick]

I just noticed, when I put my tyre dimensions in the calculator it comes up with 245x75x16 = 30.47in. If I do the same in JET's DST it comes up with 29.58in.
Any thoughts?
The JET calc was much closer to my measured diameter.

[Blacky]

The calculation EFILive uses is:

d = ((w/25.4)*(p/100)*2)+r

Where:
d = diameter in inches
w = width in millimeters
p = profile in %
r = rim diameter in inches.

So for 245x75x16, you will get:

d = ((245/25.4)*(75/100)*2)+16
d = (9.64567*0.75) +16
d = 14.4658 + 16
d = 30.47 inches

I'm not sure what calculation Jet uses, or whether Jet adds a "fudge factor" to attempt to account for tire distortion, tread wear or any other factor that may cause a difference.

Regards
Paul