WB O2 and EGT compensation

[bcroe at juno.com]

Since the sensor heater increases in resistance as
it heats up, it would be possible to periodically check
the resistance increase beyond normal (caused by
very hot exhaust).  Heater power could then be 
adjusted as required to maintain the desired 
resistance (and temperature).  This would take 
another servo loop, which could be done in the
analog or digital domain.  

Practically speaking, .05 A/F accuracy or even
0.1 is not really of value for the type of projects I
see the list doing.  The WB signal moves 
constantly much more than this.  The heater 
already partially compensates without external
feedback.  If you really need certified accuracy, 
buy the $2000-$4000.

Bruce Roe

On Wed, 24 Oct 2001 11:47:47 -0700 (PDT) David Harris
<dharris2112 at yahoo.com> writes:

> One thing I was curious about was the EGT temperature
> compensation. The reading on UEGO's has a temperature
> dependent variable. I forget which way it goes but I
> believe the UEGO reads richer than reality when EGT's
> increase. This is why many WB O2 meters include an EGT
> probe to get accurate readings.
> 
> I had asked FJO about how they obtained such good
> accuracy without an EGT probe. Here is their response:
> 
> "While it is true that the UEGO sensors are somewhat
> exhaust gas temperature dependent, this effect can
> be minimized by several techniques, some of which use
> the characteristics of the sensor itself. In fact,
> using an EGT probe to compensate would prove to be
> a less effective solution as the probe would not
> accurately reflect the UEGO sensing element
> temperature. In our wideband controller we have
> employed techniques to reduce the effect of
> temperature-induced error to less than 0.05 AFR with
> an overall accuracy of better than 0.1 AFR.
> Unfortunately, I can't go into detail on exactly how
> this is achieved as this is proprietary information."
> 
> Since the current DIY solution is analog only, I
> assume that we will have the same temperature induced
> error. I'm guessing that FJO characterized the sensor
> against a "real" WB O2 setup or came up with a
> compensation curve vs temperature.
> 
> So what do you guys think they did? What about their
> claims of accuracy? Do they sound feasible?
> 
> Dave
----------------------------------------------------------------------------
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