Date: Sat, 9 Mar 2013 23:10:12 -0500
Reply-To: David Beierl <dbeierl@ATTGLOBAL.NET>
Sender: Vanagon Mailing List <vanagon@gerry.vanagon.com>
From: David Beierl <dbeierl@ATTGLOBAL.NET>
Subject: Re: 84 Vanagon--Won't start at 40 degrees
In-Reply-To: <513BF9B3.1040704@turbovans.com>
Content-Type: text/plain; charset="us-ascii"; format=flowed
At 10:10 PM 3/9/2013, Scott Daniel - Turbovans wrote:
>fwiw,
>if I suspect Temsor II ..
>which I sure would on an 84 waterboxer vanagon
>( or conenctions to it ! )
>I would pull out my ohm meter and measure the reistance and compare that
>to the chart in Bentley.
>
>that is there standard test there too I believe..
>measure resistance of Temp Sensor II ..right there onthe thermostat
>housing , or near it ..
>two wire connector there.
>Be *real sure* it's getting a good connections. That connector gets old
>and can be dodgy.
Hi Scott,
In general I completely agree, and you know I've told people to do
both those things about a thousand times or so. But there's at least
some method in my madness.
If Temp-II is causing a no-start, it has to be either
I) not enough enrichment, because:
a) Temp II line shorted to ground (voltage approaching zero)
b) ECU providing insufficient drive on the line (internal pull-up
resistor or its connections)
c) Sender drastically low value (not very likely)
d) ECU enrichment logic not working right
OR
II) too much enrichment causing flooding, because:
a) Bad sender connection (if open circuit, voltage will be
approaching five volts)
b) Sender much too high value (not very likely unless moisture
has gotten inside)
c) ECU providing too much drive (internal pull-up resistor
partially shorted, very unlikely)
d) ECU enrichment logic not working right
Checking the Temp-II working voltage by back-probing the connector
will instantly show whether a problem is case I, voltage too low; or
case II , voltage too high; and Temp I is a handy reference point
because at equal temperatures the voltages should be very nearly
equal. If they are, then you have largely eliminated Temp-II as the
problem source. Bad ECU internal logic won't be detected, but
everything else will.
If they're not about equal, then it's time to start distinguishing
farther with resistance tests. But the voltage test is very quick,
and will detect an ECU drive problem the the resistance tests won't,
because what the ECU is actually measuring is voltage; and there are
things inside the ECU that can affect that voltage. Electrically,
the ECU internally supplies +5V to one end of a resistor. The other
end is connected to the sender which is further connected to
ground. The pair of resistances form a voltage divider, and the
voltage measured at the junction between the two is proportional to
the ratio of Rsender (including connections outside the ECU and the
ground path back to the ECU; and inside the ECU as far as the
measurement point) : Rtotal (including all connections and ground path).
There are lots of ways to get there....
Yours,
David
| 
