Silo

iTrader: (1)

Hello,

I keep hearing that the 350Z ECU has "self adjusting" abilities and therefor equalizes gains from performance bolt-on parts like exhaust, headers, intake etc.

I can not quite believe this as I don't see how the ECU should be capable of sensing the changes. IMHO the only way it could self adjust would be through O2 feedback and AFAIK none 350Z is equipped with wideband O2 sensors. Relying on its rather unprecise narrowband sensors the ECU can't use their input for WOT operation adjustments.

So, what's the deal - myth or true?

Tex Willer

I say myth

stock I was trapping 100mph at the 1/4m

now I trap 104mph (best of 104.8)

with boltons

King Tut

iTrader: (1)

I think you forgot about a couple other important sensors such as the MAF and the knock sensor. It also uses those readings to determine fuel and timing. If the ECU didn't self adjust we would all be running lean as the more air you get in and the quicker you get it out the more fuel you need which is where the MAF comes in. If you happen to be running lean then that is going to set off your knock sensor and tell the ECU maybe it should pull some timing to help out.

ZinMiami

iTrader: (2)

I am sure that some type of adjusting and learning goes on as many other cars' ECU's do this nowadays. I don't believe the common statement around that the ECU negates bolt-on performance. That I do not think would be possible unless as was mentioned above every one of those bolt-ons had an associated sensor to detect and adjust it out of the equation.

That we won't get the full effect of bolt-ons without ECU flash or ECU replacement is also probably accurate and has been shown to happen on cars that have had a successful TS reflash.

Silo

iTrader: (1)

Actual I did not "forget" about them but my question is aimed differently, as you might have noticed. I asked if the ECU has the ability to equalize power gains resulting from bolt on parts.

The MAF is, if anything, helping to gain more power as it senses the higher air flow entering the engine and maintaining correct fueling. The knock sensor would cut power but only if you were to experience knocking with the perfomed mods (unlikely for n/a bolt-ons).

King Tut

iTrader: (1)

There is no way the ECU "equalize"s power gains. That would imply that it knows how much timing it can safely run or how much fuel needs to be injected and it decides to run an incorrect amount therefore equalizing power gains.

Mr B

iTrader: (2)

O2 readings and yes, the ECU does take into fact of mods BECAUSE of what the O2 is sensing.

GQ 350z

iTrader: (13)

So does one need to reset the ecu for gains to be realized?

copba1t

iTrader: (2)

You are referring to only the pre-cat sensors on the manifold, correct?

Silo

iTrader: (1)

But how can this work off of narrowband sensor readings? As we all know narrowband O2 sensors can precisely determine stoich mixture only. And stoich is not what you aim for at WOT operation. Also, what you say would imply that there is no "open loop" operation mode with this ECU.

Please enlighten me. Thanks.

ZinMiami

iTrader: (2)

It's funny. This same discussion rages on and on at every auto forum I've been a part of. It was the perinnial debate on the spyderchat board when I had my MR2 Spyder. There was never a consensus or any concrete proof that the ECU would learn it's way around or out of bolt-on improvements.

I wonder what has to happen to ever have a definative answer to this age old question about modern ECU controlled cars?????

Brandon@Forged

We have to get the engineering team for the VQ35DE on here, that's the only solution.



Anybody want to send nissanusa.com an email suggesting they send their engineering team that worked with the VQ35DE to my350z.com?

wfcall

Just my observations.

The ECU likes stoich if in part throttle-(closed loop)
A/F guage was almost first mod.
Every bolt on we've done which has increased air flow would technically make the engine run lean if the ecu was not adjusting for it by use of the feedback from the pre-cat o2 sensors.
The A/F has always corrected to stoich at part throttle as mods were made.

Now here's my theory

With the ATI DFMU increasing fuel pressure when at WOT(open loop) and in boost then if there was a learn feature wouldn't the ECU eventually start trying to narrow the pulse cycle on the injectors to combat the "too rich" A/F it's seeing due to the DFMU?

Or is the learn feature so slow that the occassional resets the ECU gets from disconnecting the battery while working on the car enough to get it back to default values?

Incidentally I also got this response from AAM while talking too them about a reflash and it is apparently thier belief (they use the same software as technosquare) that thier IS a "learn feature"

"Unfortunately the ECU Flash can only "totally" cover the fuel curve to injectors size 360-380cc. So to allow find tuning and and to run 440cc injectors we use the assistance of the GReddy E-Manage. The Flash will handle everything else, and we set the learning feature of the ecu to target 11.5-11.8 range so the ECU wont be trying to overcorrect.

You could keep the J&S Safeguard because you already have it, but we haven't seen a need for it with our revised timing maps."

Hope this helps,
Thanks,
Mike Mahaffey
2003 350Z 12.8@113mph
1994 Stealth TT 10.7@137mph
MikeM@alteredatmosphere.com
www.alteredatmosphere.com
301.330.8835

Silo

iTrader: (1)

That's nothing new. All emission controlled vehicles using a 02 sensor aim for stoich mixture at idle and light load because that's the mixture that allows the cat converter to work best. And stoich is also the only state that the normaly used narrowband sensors (up to 4 cable sensors) can precisely determine. As you correctly mentioned this is called close loop operation and usualy you don't want to alter this as you don't need any more power at idle or cruise!


If they believe there is such a feature, they should be able to explain how it is supposed to work technicaly! It can not be through the stock O2 sensors as these are not suitable to sense rich mixtures exactly.

This sounds like there is indeed such a thing like a target A/F ratio for WOT but HOW THE HECK (I am getting frustrated) should the ECU determine "rich" A/F ratios depedably when it only has narrowband O2 sensors for input???

hfm

iTrader: (1)

Myth. Confirmed with a Nissan engineer a over a year ago.

Silo

iTrader: (1)

Reliable source? So what is AAM telling people about ECU's learning feature A/F target range etc?

Michael-Dallas

Short version: Nissan (and possibly other auto manufacturers) recognizes that as an engine ages, its performance deteriorates (i.e. sensor wear, engine wear, etc.). To compensate for this, the ECU stores short-term and long-term fuel trim among all things. These trim values are added/subtracted from the base maps. There's a schpiel in the FSM about it.

When you reset the ECU, you're clearing these trim values. At 100k miles, the maps may be too lean for an engine so it "re-learns" and re-builds these trim values over time. To a certain extent, an engine running lean produces more power than an engine running rich. That is why some people claim that their car gained power after an ECU reset.

Michael.

Michael-Dallas

Taken from the FSM, EC-26:

350Zteve

iTrader: (6)

All modern engines are so smog conscience that they do learn. What they learn is how the different fuels and weather conditions effect mixtures. The computer then compensates to provide the ideal air fuel mixture.

If you add parts to make the motor breath better, the computer will try to add the necessary fuel. The computer will never try to counter act the effects of mods, rather it will try to make them work better.

Silo

iTrader: (1)

This I agree with but in regard to the quoted text from FSM I would like to note that it can only relate to stoich mixture for most cars. Very few cars (lean burn engines) are equipped with wideband O2 sensors stock. Only these could make active adjustments for a "target" A/F ratio beyond stoich. And the 350Z does not have wideband O2 sensors - that's a fact.

So I will ask again, without any prejudice, what does AAM mean by saying "we set the learning feature of the ecu to target 11.5-11.8 range so the ECU wont be trying to overcorrect."