using a universal lambda sensor possible

[cosworthpower]

Ive been looking at replacing the factory sensor to a bosch 4 or 5 wire universal. As I dont have aftermarket management is it possible to use with the factory ECU? (AE101 4AGZE)

If not, what lambda sensor would you recommend to replace the factory one?

[oldcorollas]

5 wire is usually wideband,
1-2-3 or 4 wire are narrow band

yes you can use "universal" narrow band sensor.

the only difference will be in the heater circuit (and possibly the heater current, since different brands use different heater resistances), or the signal ground circuit

[thechuckster]

... the factory ECU only supports a narrowband sensor. Best option to pinch a 4-wire bosch sensor from the wreckers - they're not worth paying for given the limited use the ECU makes of the sensor.

[Z2TT]

You can use a universal one, just to be on the safe side make sure the resistance of the heater circuit is the same as the original one, am not sure what problems can arise if you use sensors with different resistance but rather be on the safe side.

thechuckster, Limited use? wouldn't it need sensor feedback on part throttle to maintain good fuel economy, unless your at the track beating it all the time

[TERRA Operative]

Depends on how much you value fuel economy....

[thechuckster]

narrowband O2 sensor is only going to be of use to the ECU at cruise and warm idle. In any case, most ECU's will have a maximum amount of O2-sensor driven correction (e.g. 5-10% or less).

Having the sensor data produce massive changes in fuelling would make the engine prone to damage (e.g, accidental lean-out due to a misfire).

During acceleration, and moderate to high load, you'll be driving off the fuel map (with associated corrections for coolant and air temp). Likewise an decel activity or warm-up enrichment will ignore O2 sensor data.

If you want closed-loop fuel control (and resulting economy), get an ECU that support/uses wideband sensor output.


Z2TT: heater element just gets the sensor warm instead of waiting for exhaust gasses to do that. As the sensor warms up, the heating element rises in resistance and generates less heat. It's not calibrated element like the bits found in a wideband sensor. A heating element also lets you mount the sensor further away from the head. The heating element does not affect the accuracy of the sensor at all - it just gets the thing usable quicker and gives you flexibility in packaging/engineering the exhaust system.

[CrUZida]

narrowband O2 sensor is only going to be of use to the ECU at cruise and warm idle. In any case, most ECU's will have a maximum amount of O2-sensor driven correction (e.g. 5-10% or less).

I used to think that too, until I had this problem.
http://www.toymods.org.au/forums/showthread.php?t=44212

Turned out a stuffed drivers side o2 sensor was causing the thing to run STUPIDLY lean within 20 seconds of a cold start (and ecu reset).

[thechuckster]

which does suggest the code handling O2 sensor input has no range control to prevent out-of-band/limit movement of fuelling from faulty sensor input.

[CrUZida]

Which was the most surprising part of the whole experience for me.

[oldcorollas]

could be pretty easy to check which ECU's are affected by making up a variable 0-1V power circuit to put in place of the O2 sensor.. ?

[Z2TT]

So Cruzida was it because one of the two sensors was constantly reading 0.8 (rich), that the ecu was constantly trying to lean out big time causing the lean condition?

Most "bad" o2 sensors I've tested had read 0.01 Constantly or 0, I guess something must have caused your o2 sensor to constantly read 0.8 right and the ECU didn't have any failsafe to detect this as an Illegal or bogus reading?

[CrUZida]

I'm pretty sure I tested both sensors (at the diag box) any neither were WAY out of spec.

One may have been 0.5V (the good one) and the other between 0.6 and 0.7, but honestly can't remember.

[AdrianS]

I think (not certain) that some Toyota ECU's adjust the open loop fuelling based on learning in closed loop; ie if the closed loop (part throttle) requires leaner or richer than expected, the ECU will run leaner or richer in open loop as well.

[thechuckster]

toyota ECU's don't 'learn' as such. The point of AFR targeting is for real-time correction not auto-tuning the fuel and spark maps.

And by definition, open loop would not use AFR corrections as that would defeat the purpose of having an open loop mode of operation.

[wilbo666]

toyota ECU's don't 'learn' as such.

Depends what you mean by learn?

They have a Short Term FT and Long Term FT (Fuel Trim)...

Page 6 in the below... from autoshop101.com
44 Fuel System#3 - Closed Loop /Fuel Trim w/qu.pdf file size 112KB

Cheers
Wilbo