Trying to figure what a OBDII system does to VSS....

[Juglenaut]

I am looking for data on what OBDII systems do to the input wave form from the VSS..

Is it digitized, triangulated, serialized, or all. Is the VSS wave form shaped in the PCM, or smoothened to appear as a sin wave. Is is width based or frequency based or a combination of both.

I am trying to find out why a recalibrated speed input to the PCM for larger tires would fail for P1870 Transmission Component Slipping and upward. Is the PCM expecting a certain trigger wave form??

[sharkey]

RE: Trying to figure what a OBDII system does to VSS....

the pcm looks at engine rpm and output rpm (in most cases, your vss) and if it is out of range for the certain gear (usualy in lockup) it sets the code, as it thinks the tranny is slipping. if you recalibrated the speed sensor itself, its sending the wrong signal to the pcm and and the pcm thinks the tranny is slipping.

did the problem start after you recalled the speed sensor??? if so, throw a stock speed sensor in and see if it fixes the problem. if not, then you have transmission issues.

the correct way to recalibrate for bigger tires is to have gm flash the pcm for it.

[Juglenaut]

RE: Trying to figure what a OBDII system does to VSS....

We have a signal interface module. SGI series.
Not really intended to be used as a calibrator for a PCM or ECM. But has worked before on a ton of vehicles without any problems. More or less it is purly a calibration tool for dakota digital gauges and works with alot of different gauges as well. Not intended to interface serial bus systems that use ABS in reference to actual speed. Most ABS systems use wheel sensors to detect lockup, but many are now in reference to speed.

I know that you can recalibrate by using a programmer that alot of companies sell. I would suspect the PCM on 1995 or newer OBD systems are looking for a spike or triangle wave form or a clipped triangle. The reason I would think it is triangular as most VSS's detects rise and fall of a relutor gear either dedicated or inline gear. The SGI series interfaces doesn't copy the form wave to wave it converts it to a standard square wave.

I am wondering if a PLL circuit with wave shape matching would work. The phase delay on the SGI series varies by model but the one in question the wave form is augmented and formed inline and found only a 1-2 percent phase shift on start up with standard tolerance components.

I don't think that the calibration and computation of the speed is purely fequency based. The last time I used PWM it ment pulse width modulation. Now in order to use that as a base the wave has to be converted so the PCM must be triggered by a specific wave form. Example...A given wave form has peaks, valleys, rise, and fall times, zero cross points positive and negative, pulse width, wave length, positive edge, negative edge, clipped duration, and so forth...There is alot of things that the PCM can use as it data.

Possibilities:
Clipped duration with valley fall and rise to average, RMS, and/or edge trigger; data would appear like a dashed line with duration amounts, this would give a calculated pulse width at AVE, RMS and or edge. Clipped duration can be a reference to fequency or period.

If the the wave form going into the pcm is converted to a square wave, some or most of the aspect of the calculation can be thrown off, as AVE and RMS values change.

I was up last night thinking about this untill 4 AM.