I know this is the FI forum, so be warned ahead of time this is not directly related to FI. I chose FI because there are some smart people here. I'm currently working on a little accelerometer project. However, after analyzing my VSS's output at different speeds, I came to the conclusion that its signal is not linearly proportional to the vehicle speed. I'm quite confident that the method of testing I am using is accurate because my results are very repeatable. I do have aftermarket gauge faces (for that ricer in me that never went away ). Can someone please tell me that it's just my speedometer that is off? I do, however, find this hard to believe. Either way, your input is valued. Below is some data I have collected:

 

Where is your data?

But the VSS is a pulsed output that increases in frequency with an increase in speed. Using a counter (which the ECU has), things should be linear. When I was using my OBDII scantool to compare VSS to the speedometer, it was off with the speedometer reading higher than what the ECU was reading.

 

Your speedo is probably not accurate. It is effected by the tire roll out more/less in different gears. Is there a chance that 35 and 55 are the start of a new gear?

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by EE_Chris &raquo;</TD></TR><TR><TD CLASS="quote">Where is your data?</TD></TR></TABLE>

WTDR must have been down...I see it right now.

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by DIRep972 &raquo;</TD></TR><TR><TD CLASS="quote">Your speedo is probably not accurate. It is effected by the tire roll out more/less in different gears. Is there a chance that 35 and 55 are the start of a new gear?</TD></TR></TABLE>

Even if it was the start of another gear, it shouldn't matter since the VSS reads the speed of the axles, not the motor.

Thanks for the replies...I'll have to find one of those radar signs and check my speedometer's accuracy.

 

I'm curious as to how your collecting the speed data. Are you measuring the pulse at particulare speeds and then trying to calculate the correlation between the pulse and speed reading? I'd be interesting if you could datalog via hondata or something what speed the ECU is putting out vs what the speedo is saying vs the pulse from the VSS. From that you should be able to determine the correlation between the two and decide how much and when your off. Assuming the VSS is working properly it should be a linear correlation.

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Lucky_Smurf &raquo;</TD></TR><TR><TD CLASS="quote">I'm curious as to how your collecting the speed data. Are you measuring the pulse at particulare speeds and then trying to calculate the correlation between the pulse and speed reading? </TD></TR></TABLE>

Yes, I maintain a constant speed and am figuring out a formula I can use to translate the VSS signal into a speed. I currently run Uberdata and I think I'm going to take your suggestion and use the datalogging in there to compare with my results. For some reason I forgot about it