ScoobyNet.com - Subaru Enthusiast Forum

ScoobyNet.com - Subaru Enthusiast Forum (https://www.scoobynet.com/)
-   Drivetrain (https://www.scoobynet.com/drivetrain-11/)
-   -   Techie Lambda Question (https://www.scoobynet.com/drivetrain-11/300857-techie-lambda-question.html)

Scott.T 10 February 2004 08:37 AM

Techie Lambda Question
 
After changing my Lambda sesnor I thought I would have a read up on what is in a Lambda sensor, and try to understand how it works.

An old Bosch fuel injection techie guide helped, but it didn't explain everything......

Basically the Lambda Sesnor is like a battery, it has a hollow section for ambient air and a section for Exhaust gases. These are all enclosed with electrodes and ceramics. As the exhaust gases flow over the tip of the sensor a voltage is created due to the differences in oxygen between the ambient air and the exhaust gases (~1.0V) and read by the ECU.

Now in Closed loop mode the ECU continually adjusts the Injector duty cycle to try an obtain the correct mixture, and at idle the swing from lean to rich and back to lean can take 1 to 2 seconds.

As more exhaust gases flows over the sensor the swing rate from lean to rich and back to lean increases in frequency to several times per second.

O.K, so this is all very well in closed loop mode and using an AFR you can see this swinging AFR reading being output from the Lambda Sensor.

BUT, when you go 'positive' boost or above say 3,000rpm the ECU flicks into Open-loop mode. This then causes the Lambda to give a constant display........WHY......whats caused the lambda to stop outputing a pulse in relation to exhaust gas flow rate.

I know the ECU has changed modes, but why has the lambda signal changed.....it has no other connections to tell it to change ?

TopBanana 10 February 2004 08:52 AM

The lambda sensor is always reading, it doesn't output pulses. The changes you're seeing are a reflection of the ECU overshooting and undershooting in closed loop. In open loop it just looks up the correct amount of fuel and applies it. It should still change, just not swing around. If your sensor is a narrowband one (the standard type), it probably won't read much difference between 10:1 and 12:1 as narrowbands are only accurate around 14.7:1.

... unless I've misunderstood your question!

SiHethers 10 February 2004 09:03 AM

Hi Scott

As you've said, the lambda probe merely reports back to the ecu the oxygen content of the exhaust gases. In closed loop mode, the actual fuelling is determined by the base map value +/- a correction value which the ecu determines from the lambda input. The ecu then swings across stoich repeatedly to attain best catalyst efficiency.

In open loop, the ecu ignores the input from the lambda making it redundant, and fuels only according to the fuel map (without any correction). As the ecu is no longer paying any attention to the lambda probe and trying to achieve as close to stoich ratio as possible, the probe just reads rich (as most ej20 ecu fuel maps run in the 11:1 or greater AFR when in open loop mode).

Make sense?

Simon

Scott.T 10 February 2004 09:04 AM

Bingo...
 
I'm with you there... I mis-interpreted what the book said about the pulsing It read as if the Lambda was pulsing, when actually, as you say it is the ECU swinging it about to achieve Stoich.

The ol' Lambda Sensor is a clever little thing..being able to generate Volt from Oxygen and all that.......


All times are GMT +1. The time now is 08:10 PM.


© 2024 MH Sub I, LLC dba Internet Brands