triumphs
[Top] [All Lists]

Lamda sensor - update

To: jag-lovers@sn.no
Subject: Lamda sensor - update
From: Egil Kvaleberg <egilk@sn.no>
Date: Sun, 25 Aug 1996 16:53:20 GMT
Newsgroups: mail.triumphs
Organization: Siving Egil Kvaleberg AS
Xref: egil mail.triumphs:7582
Having received quite a number of mail messages since my little article 
on the Lambda sensora while ago, I would like to add the following notes:

   1.   There has been reports of unstable voltmeter readings. I 
        personally find that the readings are stable to within 5 or
        at least 10 mV, so this surprised me. BUT, what I didn't
        mention in my previous article was that the quality of the
        ground connection is just as important as the centre connection.
        
        If one relies on the existing ground connection, usually via
        various exhaust system joints in various state of decay, through
        the engine block, the engine ground strap and finally the chassis,
        there would surely be a lot of noise induced. 
        
        I used a shielded coaxial cable for the connection. The centre 
        cable goes to the sensor lead, of course. The shield is connected
        to the sensor body via a 10 cm long piece of cable. I used a lab
        type cable with many strands and heat-resistant isolation. I
        connected the ground connection to the sensor body using a
        stainless hose clip. In the other end of the coaxial cable, I
        measure the voltage between the centre conductor and the shield,
        and do not have any ground connection.
          
   2.   It has been mentioned that a three wire sensor might provide 
        better accuracy. I thought the only purpose of the internal heat
        element of the three wire sensor was to reduce the heat-up time. 
        I have no experience with this, so I really can't say much more.
        But my impression is that once heated up after a cold start (a 
        minute or two), the one-wire sensor performs very good and
        consistently.
        
   3.   There has been warnings that lead and silicon may quickly 
        destroy a sensor. Since I have only been running on unleaded
        gasoline, again I have no experience, and can only forward the
        warning. (The typical source of silicone would be remains of
        silicone gasket material that pass through the exhaust.)

   4.   As a further warning, one should of course use common sense if
        attempting to lean out the mixture too far beyond stoichiometric.
        A lean mixture will rise engine temperatures. A too lean mixture
        on a high performance engine in a high load situation may cause
        permanent damage. 

        It has always been good practice to go for a slightly rich
        mixture under wide-open-throttle situations, and it still is.
        It gives more horsepower to play with, too. 

Egil
-- 
Email: egilk@sn.no  Voice: +47 22523641, 92022780 Fax: +47 22525899
Snail: Egil Kvaleberg, Husebybakken 14A, 0379 Oslo, Norway
URL:   http://www.sn.no/home/egilk/
URL:   http://www.sn.no/home/egilk/t2000.html


<Prev in Thread] Current Thread [Next in Thread>
  • Lamda sensor - update, Egil Kvaleberg <=