Jump to content
Ultimate Subaru Message Board

Recommended Posts

The first thing I'd be checking for is an exhaust leak at a point somewhere after the first O2 sensor but before the second O2 sensor. This would cause the rear sensor to send a leaner-than-normal signal back to the ECU which will then adjust long-term fuel trim upwards (richer) to compensate.

 

If the leak has slowly grown to the point that the ECU can't effectively compensate for the discrepancy anymore, it will be "out of range". Mind you, that any leak before the rear sensor would be causing the car to run slightly richer than normal, so you'd probably be taking a hit on fuel economy if this were actually your problem.

 

My second guess would be the ever-present connector-integrity problem on modern cars, especially salt-driven ones. In other words you may just have a connector between the rear sensor and ECU that isn't making good or consistent connection anymore.

Link to comment
Share on other sites

This is a very common code to set on the Forester and Impreza's that had a MAF sensor recall. I haven't see it on a '95, but you could have a failing MAF that is giving out of spec readings.

 

An air leak before the front cat usually causes the computer to set a P0420 cat efficancy below threshold code. Don't know how much affect and air leak between the front and rear cat would have on the fuel trim. I think the rear O2 sensor is there just to moniter the cat efficancy rather than control the fuel trim. Not positive of the logic used there, but wiht '95 being the first year of OBDII, I'd say that is probably the case.

Link to comment
Share on other sites

WAWalker, it does both.

 

Cat efficiency is determined by the ECU's counting and comparing the crossing rate of the front and rear sensors, then calculating a "crossings ratio". If the ratio drops below a given threshold, the cat efficiency code is displayed.

 

The long-term fuel trim is adjusted by the ECU's shifting of the Lambda target in response to the averaged voltage taken from the rear sensor and logged at the end of each driving event (engine start to key off). The Lambda target is the reference number which the front sensor shoots for as it actively controls real-time fuel mixture.

Link to comment
Share on other sites

WAWalker, it does both.

 

Cat efficiency is determined by the ECU's counting and comparing the crossing rate of the front and rear sensors, then calculating a "crossings ratio". If the ratio drops below a given threshold, the cat efficiency code is displayed.

The rear O2 sensor is switching at a much lower frequency and with much less amplitude in the signal than the front because the cats are burning the hydrocarbons so the rich lean condition that existed before the cats is not as drastic after. If the frequency and amplitude of the rear sensor reach a certain level then the computer knows that the hydrocarbons that the front O2 sensor saw are not being sufficently burnt off by the time they reach the rear sensor. So yes the coumputer is using both sensors to moniter cat effiency. Cat efficency testing is why the rear sensor was added.

 

The long-term fuel trim is adjusted by the ECU's shifting of the Lambda target in response to the averaged voltage taken from the rear sensor and logged at the end of each driving event (engine start to key off). The Lambda target is the reference number which the front sensor shoots for as it actively controls real-time fuel mixture.

I'm sure they are using the rear sensor info to adjust long term fuel trim now. I was questioning weather the first OBDII systems did.

 

The front O2 sensor is not shooting for any thing. The computer is using the info from the front O2 sensor to adjust short term fuel trim. The computer is shooting for an air/fuel mixture of 14.7:1 that switches rich to lean at .5-5Hz.

 

 

So yes a leak between the converters could change the LTFT.

 

But I have seen a cracked header pipe on an '00 OB that set a P0420 cat efficency code. I remember the freeze frame fuel trim readings being abnormal, but no fuel trim codes.

 

Lambda............That is an European car term, not an Aisan car term;)

Link to comment
Share on other sites

The front O2 sensor is not shooting for any thing. The computer is using the info from the front O2 sensor to adjust short term fuel trim. The computer is shooting for an air/fuel mixture of 14.7:1 that switches rich to lean at .5-5Hz.

 

We're basically saying the same thing. So we've basically just got a semantics argument going at this point.

Link to comment
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
 Share

×
×
  • Create New...