little deuce overfueling issue resolved.... i think???

[Paul Woods]

Just seeing this thread now, if the fuel trims come down to around zero when the engine is revved while still in closed loop (under 3500rpm from memory) then it sounds like the AFR sensors are working fine, check to see if both sensors start switching when the revs are held at say 2500rpm and that fuel trims return to normal, my bet is they do. It is highly unlikely both sensors have gone down at the same time unless their power feed was interrupted.

If the sensors do start switching when revs are held to 2500 and the fuel trims hover around zero that would mean you just have something upsetting the mixture at idle, you can further test the system is working properly by letting the engine idle and spray a small dose of carb cleaner/wd40 into your air filter, with your current fuel trims maxed out at a positive value the ECU is trying to add fuel at idle to correct an overly lean reading coming from the AFRs, so by squirting WD40/carb cleaner into the intake you should see the short term fuel trims shoot down to a negative value (depending on how much you squirt in)

Your MAF values at idle seem a little high to me, that will have a bad effect on idle fuel trims, I would replace it, if it helps a Mk3 Mr2 one is identical to the Rx300 one, from memory a reading closer to 2.5 g/s is more normal at idle.

Conduct the test above, that will confirm your AFRs are working and that the control system is operating as it should be.

If the AFRs are fine your pegged out positive fuel trims mean the mixture is too lean at idle, the ECU is trying to add more injector duty to switch the AFR sensors back to rich, people think the AFR sensors drive the fuel trims when in fact it's the trims that drive the AFR sensors which are merely reporting back what is happening. The ECUs mission is to keep the AFR on target by first of all altering the short term trims which directly drive the AFRs, if they peg out to maximum/minimum then the long term trim steps in and pulls the short term back into range by adding/subtracting a larger amount of fuel duty. In combination they allow the AFRs to drive in the "sweet spot"/stoich.

Your system is basically telling you that even after adding the maximum amount of duty in both long and short term trims the AFR sensors are still seeing a weak mixture.

Do the tests above and report back, I can guarantee you the trims will have been perfect 4 years ago as I monitor stuff like that religiously upon completion of a swap. The fact that it is affecting both banks in the same way means you either have an air leak somewhere or the MAF is bad.

Hope that helps.

PS I should add that even something as slight as a vac hose not seated correctly or split, will slowly drive the fuel trims to max out at idle. Once you've confirmed the AFRs are working try spraying carb cleaner at suspect points on the engine while monitoring the short term trims (be careful not to get it near the air filter) , if the short term trims shoot to negative or go down you have found an air leak.

[rainen]

Glad you have come online on this one Paul. Always enjoy reading about fuel trims, getting closer to understanding these a bit more myself. When I first got my Lexus if you can recall my MAF was reporting 4.95 g/s but the fuel trims were high negative. When I replaced the MAF with a new one (as cleaning did not help!) with idle at 700 and trims close to zero the reading was 3.42 g/s. Anyway i'm not the expert just a piece of data from my running Lexus engine before conversion. Really interested to see how this one pans out in terms of fault and cure.

[Paul Woods]

Yes that's right I remember, 3.5 g/s may well be within tolerance at idle, I just seem to recall a value lower than that, to be honest it's not a figure I spend a lot of time looking at, the fuel trims and AFR readings are all that really matter.

I remember back in the day when I attended a Lucas electronics/engine management course, one of the exam questions was....

"On a four cylinder combustion engine, If your short term fuel trim was pegged at -19.5% and your long term fuel trim was pegged at +30.5%, explain the engine fault?"

I'll let you chaps ponder that one for a while :)

[rainen]

Well Paul that's a tough question. Been pondering it for the afternoon (too much time on my hands!). Conclusion is i'm not knowledgeable enough to explain any faults under those conditions. All I can say is that some articles talk about the combination of both trims being within +/- 10% is OK. So taking these two together at +11% is only just outside of that. Over the long term the engine is lean and adding fuel but the short term is recording rich and taking it away. The only fault I could find close to having opposing trims is a blocked cat causing back pressure but in that example the trims were listed as the exact opposite of yours. I'm sure you will offer another lesson with the actual answer which will again grow the knowledge base. Sorry for the slight deviation on your thread Ashley!

[D2W]

Hi Paul. Enjoying the problem you posed. Is the mention of "four cylinder engine" significant?


Sent from my iPhone using Tapatalk

[Paul Woods]

Nice try and reasoning Nigel, unfortunately not correct but good effort sir.

No the 4 cylinder part is not relevant, only serves to show it's not a V engine with two separate fuel control systems, just has the one.

Fuel trim diagnostics are a fantastic tool for funding all sorts of hard to trace faults, as long as a good understanding of how the system works is learned then you really can use them to narrow any problem down.

If nobody gets the teaser question I will post up the answer tomorrow, but more importantly we need to fix the OPs issue.

[D2W]

Hi All

I'm on my way to Australia. First leg is Paris - Abu Dhabi through the night. I planned to sleep but Paul's poser is keeping me awake. I have an answer but it's probably horseshit.

(Honestly, I don't really have a clue what I'm talking about. I just have a restless imagination.)

Am assuming (assumption 1!) that there is only ONE fault that is supposed to explain both wayward fuel trims.

So.... my thought process is to imagine my brain is the ECU and I'm setting the long and short trims in response to signals from sensors which are WORKING PROPERLY (assumption 2 - a bust sensor is a boring answer). Am guessing here because I really have never worked this through before. But I'm having fun and thus may offer a little entertainment to those of you who do really know what you're talking about.

So... I think that the ECU thinks:......

<O2/Lamda sensor input>
FUCK. Way too rich. BIG trim to lean right now.

<Some signal input>
Ooh. Trim richer now.

<Some signal input>
Ooh. Trim more richer.

<Some signal input>
Trim more richer still.

<Some signal input>
Keeeeep going. More richer.

<Some signal input>
More richer. That's it keep going. Nearly there.

<Signal input reaches desired value>
OK we're fine now.

<O2/Lamda sensor input>
Oh FUCK. Now way too rich. BIG trim to lean right now.

And round we go again.


So what signal would drive the ECU to a richer mixture and then send a happy signal. What signal would ECU consider more important than the opposing signals from Lamda/O2 sensors? What signal could be persistent and long term?

My best guess:.. Some kinda anti-knock system. Knock sensor detects knock. ECU enriches mixture to compensate. Knock sensor says thank you very much I'm all good now. ECU: no problem sir; I've always got time for First Class customers like you. Lambda / O2 pipe up from Economy in the back: OI Chief! 'Ave you seen the state of this mixture!? For fuck sake your way too rich. Sort it out you towrag!

And with a 'happy' knock signal, the ECU happily trims big time back to lean again.

...Then... ... there's a little knock.... ...


See how my mind works. Even if I am right (and let's face facts - I have two hopes here and one of them Bob)... Even if I'm right, what kind of psychopath lays awake working this all out on a wing and a prayer!?

Of course, none of this is limited to a 4-cylinder engine... Hmmm...

Then you have to ask... What is causing knock?... Advanced timing?... Overheating?... Failing EGR... bad knock sensor?... Lean mixture!!??...

OK. Time to put this down before I disappear up my one arse.


Hope you're having fun
D2W






Sent from my iPhone using Tapatalk

[D2W]

PS:... What time is it where you are?


Sent from my iPhone using Tapatalk

[Paul Woods]

I'm guessing you're not using flight mode on your phone, naughty boy!



Good try but no, any signal from the AFR/lambda will get the same positive or negative response from the ECU and the fuel trims will go in the same direction, the long term can take a little longer to adjust so sometimes you might see a small positive short term value and a small negative long term, but that only lasts a few minutes, if the short term stays in one direction for too long the long term will reign it back to closer to zero. I've noticed on Toyota/Lexus ECUs that the long term doesn't like the short term going much beyond + or - 7% before reacting to pull it back to zero.

Ok, answer time, the only condition that explains two pegged out fuel trims in completely opposite directions is....

The engine HAD a fault, for example a bad intake manifold air leak which would peg both short and long terms to +19.5% and +30.5% respectively. The technician then fixes the air leak but forgets to reset the ECU, and immediately the short term will fling to -19.5% to try and get the fuelling back on target. Bit of a trick question in that there is no actual sensor failure or current fault, just an ECU trying to deal with a previous fault and no ECU reset.

Ashley have you done those tests yet?

[ziggy]

Hi Paul

Slight problem, Ashley(my son & the brains) is in Bournemouth, the car and code reader are with me in Derby.

The only thing I can get out of this code reader is Coronation Street ! Shows my mechanical knowledge.

Will hope to get some meaningful readings back to you at the weekend when he has a bit of time.

Please stick with it, all help very much appreciated.

Thank you