Re: Re :Intercooler/o2 volts/Injector constant

[Barry Tisdale <btisdale@cybersol.com>]

Good stuff, Dig; I just *love* this basic research......

Could you clarify the electrical hookup of the two units (datalogger &
Diacom)?  I'm interested in the voltage discrepancy; different ground
paths, maybe?  Same data line?  Math on the O2 is supposed to be: N*4.44 =
mV (N = ADC output, 0-255).  On my Sy, I found a poor ground from the
engine to the sheet metal.  Added several grounds from engine to body, to
ECU, to batt, etc.  Also sheilded the data line from the ECU ALDL to my
logger tap.  Smoothed out lots of the data & stopped the occasional lost
frame.  Maybe something like that here?

Are you saying that the same 980mV on the Oxy sensor (or Diacom's report of
it) equaled 11.5 *and* 12.4 on the WB (on 2 runs)?  Good to know the
relative numbers are this close, even though the absolute AFR values aren't
gettable.

Thanks for the info - Barry
At 10:11 PM 05/05/2000 -0700, you wrote:
>I did a little research into this with a friend's
>datalogger and a Diacom...
>
>What I found was that, on the Sy we were testing,
>the actual o2 volts measured by the Datalogger were
>.192 lower than the Diacom o2. Not sure what the 
>reasons for this are, but I suspect it has to do 
>with the reference voltage of the A/D and the 
>formula used to calculate o2 volts from the scan
>data.
>
>What I also found (from actual data compared to an 
>NTK wide-band o2) was that the o2 voltage was not
>repeatable on this vehicle. 980 mV during one
>engine run session was 11.5:1 AFR, the next time
>we ran it (on the same day, same temp condtions)
>it was 12.4:1 AFR. There was no way you could 
>guarantee what AFR a particular Diacom voltage
>indicated, even on the same vehicle during the same
>day.
>
>What I *did* find interesting was that the trend 
>of the Diacom o2 voltage at WOT followed the NTK
>wide-band data exactly. The indicated range, however, 
>is very small- you need to export the Diacom data to
>Excel and graph the o2 data from 700mV to 1v or so.
>The peaks and valleys show up quite readily, and
>can be used to show lean/rich spots in a particular
>cal. That-is, lean or rich relative to other points
>in the run, since you don't really know for sure
>where your max power AFR is.
>Later,
>
>Dig

----------------------------------------------------------------------------
To unsubscribe from gmecm, send "unsubscribe gmecm" (without the quotes)
in the body of a message (not the subject) to majordomo@lists.diy-efi.org