[Diy_efi] wide band O2's

[gary]

Message> I guess I was leaning towards the controllers/sensors ability to maintain 
> sufficient accuracy for a complete "life cycle" (typically 160,000km for 
> standard zirconium sensors) Sufficient I guess for complying with FTP, 
> but not necessarily with the high accuracy demands of a competition tune.

Maybe you misunderstood me.  In an OEM stoichiometric environment,
there is no need for the periodic recal.  As noted earlier, the accelerated
drifting (actually aging) of the WB sensor is a result of (predominately) 
operating rich of stoich.  This is not a condition experienced by OEM 
(for the most part).  How often does the average OEM compliant vehicle 
operate at WOT, or IOW in open loop and rich of stoich?  It is not a 
question of accuracy per sa, but that of operating the sensor under 
conditions that accelerate the sensor drifting/aging.  This in turn, sets
the stage, for inaccurate sensor controller data output. (Or maybe I 
misunderstood you. :) )
  

> However a "free-air" calibration process only confirms the accuracy of 

> the sensor under lean conditions only (@20% concentration-partial 

> pressure) 



Agreed, but should be accurate at stoich as well.  



> If these must be performed routinely, under the severe use of a 

> performance application, wouldn't it then make sense that the 

> sensor also be tested for "rich-end" or even stoich accuracy 

> using various BAR certified blends? 

>

> So my question is (in a nut shell) if one check and recalibration 

> is so important (lean/free-air), why not the other side of the AFR 

> equation??



First off, various WB controllers use the same or similar sensors.

The variable is the controller itself.  If in working order, the sensor

is just a messenger.  To continue...

You could I 'spose.  However, if it is recalibrated periodically (with

frequency dependant of severity of use), one could then utilize any

documentation from mfr. to help determine the ultimate accuracy 

of their device.  Such as the following from one WB controller mfr.

Not ABSOLUTELY infallible, but a far cry from blind faith, as some

mfrs. expect out of you.  :)   OK, now from one particular mfr.:



"I can send you the logs we did with calibrated gases at 0.8, 0.85 

and 0.90 lambda as well as a plot from testing response from free 

air to lambda 0.8. For the test, the sensor was mounted in a small 

measurement chamber (< 1cc) connected through a solenoid to 

cal-gas at 1.5 bar.  With the cal gases, we found a max error of 

less than 0.07 AFR tested over about 30 sensors."


The above would work for me.  :)  But how 'bout the sensor?  Well,
are you able to (re)calibrate it?  I would then consider the sensor a
useable one.


GAS


> Incidentally, no one responded to my query from the other day.   



Hmmm, gee whiz.  How 'bout that,aye?  First time, shame on you.

Second time, shame on me.  BTW, I play softball, not hardball.




----- Original Message ----- 
  From: WopOnTour 
  To: diy_efi at diy-efi.org 
  Sent: Wednesday, July 06, 2005 9:45 AM
  Subject: Re: [Diy_efi] wide band O2's


  Thanks Gary

  That makes sense, certainly the oems need to consider the negative consumer effects of requiring a routine calibration.  In this day-and-age of maximizing service intervals that wouldn't make sense. I guess I was leaning towards the controllers/sensors ability to maintain sufficient accuracy for a complete "life cycle" (typically 160,000km for standard zirconium sensors) Sufficient I guess for complying with FTP, but not necessarily with the high accuracy demands of a competition tune.

    

  However a "free-air" calibration process only confirms the accuracy of the sensor under lean conditions only (@20% concentration-partial pressure) If these must be performed routinely, under the severe use of a performance application, wouldn't it then make sense that the sensor also be tested for "rich-end" or even stoich accuracy using various BAR certified blends? This to calibrate for the sensor skew you describe that should affect the sensor ability to react to HC, H2 and CO to establish an accurate rich AFR. So my question is (in a nut shell) if one check and recalibration is so important (lean/free-air), why not the other side of the AFR equation??



  Regards

  WopOnTour

    ----- Original Message ----- 
    From: gary 
    To: diy_efi at diy-efi.org 
    Sent: Wednesday, July 06, 2005 7:57 AM
    Subject: Re: [Diy_efi] wide band O2's


    > One thing I noticed is that the oe solutions from Bosch NEVER 
    > describe any sort of routine calibration, as most of the current 
    > aftermarket solutions claim was necessary. So it is my desire to 
    > determine (if I can) HOW the oem WB02 hardware and/or control 
    > strategies are so different that they do not require any sort of "zero 
    > oxygen" calibration apparently for the life of the vehicle (or the life 
    > of the sensor- which ever comes first! LOL

    > It's been known that these ASICS do not require routine calibration 
    > as they derive a trim solution based on the sensor calibration resistance. 

    > I'm curious what the source/root cause of any significant drift might 
    > be - normal chemical breakdown? contamination?

    The ALL analog circuitry was designed for OEM applications. The 
    OEM WB sensor is not subjected to the abuse it endures in a 
    'performance' or 'tuning' tupe application, be it lead, or something 
    less severe like AFRs on the rich side of stoich, (IOW 12.0 AFR 
    or even richer) something not experienced by a sensor in an OEM 
    application.   This unusual wear 'n' tear causes the WB sensors to 
    drift.  Without the abuse that causes this drift, there is/was no need 
    for manual calibrating/compensating circuitry in the OEM analog 
    circuitry.  HTH.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.diy-efi.org/pipermail/diy_efi/attachments/20050706/31da6ddf/attachment.html