[Diy_efi] Maybe a little O2 sense

[Bruce]

Mess around trying to get a car to pass an emissions test at idle, and one
of the first things they mention is raising the idle speed to the max
possible.   And the excuse most often given is that it mins the effects of
EGR and reversion, since the *flows are so low*.   OK, at a glance that is
resonable, and maybe even true, but.
   Now we get to McDs.   I've been noticing the under just the right
lighting the GN smokes a lil at idle.   Most notably while in line at McDs
for am coffee.   For just about forever, I'd thought the Turbo had a seal
the was just so ever slowly seeping, and it was just some oil but the color
was more gray then blue.
  Hmm.
   Using the WB, BL numbers, O2 voltages, commanded AFRs, and everything
looked fine.   So I'd just left it at that.
   Enter the Open Loop chip, and the *what ifs*.
   In closed loop you'll always see the O2 toggling (open loop or closed
loop)(just a matter of when the sensor is sampled and what's going past it
when it's being sampled), and the BL numbers take a set after a while.  Hmm,
for that to happen, means that the ecm is doing alot of filtering.  Hmm, O2s
toggling, and lots of filtering.  At idle with the exhaust pulses so spaced
out, then the O2 has lots of none reading time that it filters into the
equation.  Hmm, just for grins, I tried some experiments with the MAF tables
and Scalers to really lean down the idle.   And there was no difference in
idle quality, even thou I'd dropped the PW from 2.2 to 1.95msec..
   If ya'll remember way back in Tuning tips are in about that time frame I
said I'd noted some situations where the engine ran better at the extremes
of the BL at times, I think this explains it alot better then the self EGR,
reversion deal.
  A low engine speeds there is just such a large volume of voids in the
exhaust stream, it's hit ot miss when the O2 is sampling a rich slug or lean
one.   Since the *leans* out displace the rich ones, then the O2 overcomps
by going too rich to get the right average.
  Or so it looks to me.
  Bruce




_______________________________________________
Diy_efi mailing list
Diy_efi at diy-efi.org
http://www.diy-efi.org/mailman/listinfo/diy_efi