[Diy_efi] Re: [Efi332] Fw: wide-band O2 sensor comparison

[Andrew Brownsword]

FYI, I'm only on the DIY list and I'm receiving these messages.  Ignoring
the heated political content of this discussion:

If the sensors are individually consistent in how they are "off", then it
should be possible to use a commercial meter to do the calibration once.  It
might not even be necessary to do that if you can use an engine with a
properly working HEGO closed loop -- the flip-flopping across stoich ought
to be visible on the DIY-WB, giving a reference point for interpreting the
device's output.  Is this not the case?



on 8/11/02 9:25 AM, Garfield Willis at garwillis at msn.com wrote:
> Notice BTW that none of my stuff is making it out to DIY. "Too personal"
> for you, Rivethead?
> 
> On Fri, 11 Aug 1995 07:59:55 -0400, "Bruce" <nacelp at bright.net> wrote:
> 
>> And for some of us, that's all we need.   Just in reading your posts, one
>> can see the possible sources for errors, so it all does boil down to what
>> someone does consider as close enough.   And that requires a judgement call
>> (on their part), of what, one considers good enough.
>> Thank you, for your endorsement.
>> Bruce
> 
> Let me remind you of something else I've also posted about previously,
> which won't show up in your tests until you have the results logged in
> real-time, and those are the whopping big transient errors we've seen in
> testing your horseshoesOmeter. Have someone with decent equipment like
> that friend of yours, but on some vehicle that can produce load and afr
> swings a good bit faster than a big truck engine, cause a large rich
> swing all the way from dfco to WOT, like you punched it after coasting,
> and see where you land on the rich side *indicated* relative to some
> decent equipment. The errors are a good deal *larger* than even the ones
> you've just reported, because of your sloppy b-grade push-pull driver
> arrangement on the Ip pump. You don't actually have a real servo there,
> and you either under or overshoot the pump current given a good size
> transient, which throws off the transient results even worse than the
> static errors. If you HOLD the rich or lean transient (IOW, you make it
> not a transient but hold the condition long enough), it will settle to
> the level of static errors you're getting now, but that gives you an
> even falser reading for the first few tenths of a second, and puts the
> measurement of any *key* lean transients ALSO in question as to their
> magnitude/severity.
> 
> Yeah, if all you're looking for is "do I have a blip", instead of "do I
> have a blip, and how bad is it?", then once again, it's "good enough"
> for you. Geezus. As I've said before, NOONE in the world of EE *or* real
> proformance would consider any of this "acceptable".


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