UEGO bias (Heater requirements)

[Ezra Hall]

Bob,
 Thanks for the info. I am interested in the schematics you have, any chance
you could scan them? Or fax them? I am particularly interested in any
information on protection of the pumping cell until proper temp has been
obtained, or any other conditions that might be using to determine when it
is safe to enable the feedback loop. The patents I have looked at do not
provide enough details, and I don't want to "blacken" my sensor...

 I made some measurements tonight on my UEGO sensor to determine the Thermal
Coef of resistivity for the heating element. I did this by measuring its
resistance at room temp, then at ~200DegC (heated the unit for 30minutes in
the oven). I fashioned a 10mA current source (calibrated to +-2%), and used
this to apply current to the heating element to more accurately measure
resistance. To measure temp, I placed a thermocouple in one of the holes on
the sensor tip, and wrapped the assembly tightly in Al foil. Here are the
measurements at various temps as it heated up. The last measurement was
after approx. 30 minutes in the oven.
Temp (C)     R (ohms)     Coef (Ohm/C)
25                 3.02             NA
165               4.39             0.0098
209               4.89             0.0102
220               4.95             0.0099
221               5.06             0.0104

I will use 0.01 Ohm/Deg C for simplicity, any idea what metal is used?

I then sampled heater current vs. time (2Hz sample rate) while applying 5
volts then 10 volts to the heater. Here are a few data points for reference:

Time        Amps    Volts   Ohms  Deg C
    0.00     1.23    4.90    3.98  121.37
   10.00    1.06    4.90    4.62  185.26
   20.00    1.01    4.90    4.85  208.15
   30.00    0.97    4.90    5.05  228.15
   40.00    0.94    4.90    5.21  244.28
   50.00    0.94    4.90    5.21  244.28
   60.00    0.93    4.90    5.27  249.88
   70.00    1.64    9.70    5.91  314.46
   80.00    1.40    9.70    6.93  415.86
   90.00    1.33    9.70    7.29  452.32
  100.00   1.29    9.70    7.52  474.94
  110.00   1.27    9.70    7.64  486.78

 Sitting in free air, this thing obviously is not going to reach 650 Deg C
with 9.7 volts to the heater in any reasonable amount of time. That would
explain all the patents on controlling the temp of these sensors by using
resistance as the feedback parameter of a simple control system...

I will CC the diy-efi list in case anyone else is interested in this info.
Steve Gorkowski sent me some pictures of current vs. time for his sensor, so
I know he will be interested. Let me know if you can send me any additional
info from the NTK documents.

Thanks,
 Ezra

-----Original Message-----
From: Bob Hale x594 <bob_hale at or.credence.com>
To: ehall at together.net <ehall at together.net>
Date: Tuesday, March 02, 1999 4:20 PM
Subject: UEGO bias


>I saw your posting in the diy_efi archives regarding the heater voltage
>for the wide range Honda sensor.  I was looking for data on such sensors
>for my own purposes, and thought that I would forward what I have learned
>to you.
>
>I contacted NGK/NTK regarding wide range sensors and they eventually sent
>me some data.  They didn't want to acknowledge the Honda part number, but
>they did send me data on a sensor which appears to be identical to the
>Honda sensor, based on descriptions that I have read in the diy_efi
>archives.
>
>The NTK UEGO part number is TL-7111-W1, and NTK wants $600 plus an
>agreement to not sell it, to use it only on an engine, etc.  The
>companion interface box is another $400, and the cable is $60.
>
>The NTK data says that the heater supply voltage is 10.5 +- 0.5 volts,
>that the heater resistance is 2.9 to 3.4 ohms at room temperature,
>that the pumping current is 3.86 to 5.34 ma with 16% oxygen atmosphere.
>It also cautions that the user shouldn't have the sensor present in
>an exhaust system if the heater isn't powered up, and shouldn't
>install the heater in a low spot where liquid water could condense.
>Also, prohibitions on fuel that contains lead, phosphorous, sulfur,
>and silicon are in effect.  There is a conceptual schematic diagram
>of how to implement instrumentation around it (no component values).
>
>The sensor is a two-stage unit: the first stage is supplied with
>a current and that pumps oxygen into a chamber where the second
>stage reacts that oxygen with exhaust gas with a 450 mv bias present.
>The exhaust oxygen content is inferred by measuring the current in the
>second stage.
>
>I hope that this is useful to you.  Please feel free to make use
>of it as you see fit.
>
>Best regards,
>
>Bob Hale   bob_hale at credence.com