oxygen sensor location?

Walter Sherwin wsherwin at home.com
Tue Jul 31 03:29:01 GMT 2001




> On the H pipe it seems that you would mainly get the sequential cylinders
on
> a bank.  For example, on a Chevy 8 and 4 fire before an odd cylinder
fires,
> and 5 and 7 fire before an even cylinder.  I believe that you would have 8
> and 5 cylinders sampled inappropriately.
>
> dh
>
>
> >    Is there a consensus on a good location for the oxygen sensor in a
> > true dual exhaust system?  I've been doing some research, and some
> > people say to just stick the sensor on one bank, others say put it
> > in the H pipe to read both banks, others say there isn't enough flow
> > in the H pipe to get a true reading...  Its a bit confusing.  :-)
> >
> >    I suppose it would be possible to use 2 sensors, one on each
> > bank, just past the collectors, and average the readings, either
> > via a digital circuit or analog, but I don't know if that is
> > really nescessary or not.  In my case, the fuel injection
> > system is just a simple TBI setup, so the computer isn't
> > looking to equalize the banks or anything, it'll just adjust the
> > mixture for the entire engine in closed loop mode based on one oxygen
> > sensor signal, so I want to be sure I give it the most accurate
> > signal possible.
> >
> >   Any info or insight would be greatly appreciated.  Thanks!
> >
> >                                               -Jon-
> >





Irregardless of whether or not you have headers, you should really try to
keep your O2 sensor in roughly the same location (and sampling the same
cylinders) as what your stock software setup accommodated, especially with
GM TBI, after your conversion.  This way, all of the C/L INT and PROP and
TRANSPORT delays (and functions) of your OE software can be somewhat
persevered.  If your OE stuff  sampled one bank originally, then do the same
in your conversion.  If it was in the Y-pipe originally, then do the same in
your conversion.  If you are much further downstream of the OE sample point
after conversion (or if you have thin tubes), then consider a heated O2
sensor if not already equipped so, and switch it to an ignition source.

Chances are probable that an O2 sensor in a conventional "H" pipe will not
sample sufficient exhaust gas flow, and will not work successfully, as an
"H" pipe is somewhat of a dynamic pulse device as opposed to a flow device.
If you really want to average your O2 signals for the whole engine, then
consider the mechanical version of averaging  via either a "Dr. Gas"
cross-over or a setup kind of like the one shown here that works quite well
ftp://ftp.diy-efi.org/incoming/DualHeadersO2.bmp  .  You may have to twiddle
the C/L software parameters to suit your application,  if your OE
application was not already in the "H" pipe and located well downstream of
the cylinders.

The referenced photo is of a system once used in a single wire zirc O2 90's
TBI GM truck dual exhaust application, with stock oriented software, with
great success.


Walt.


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




More information about the Diy_efi mailing list