o2 Operation/theory

Nistune topics specific to the 6802 cpu

Moderator: Matt

Post Reply
bsdkllr
 

Posts: 56
Joined: Wed Mar 09, 2016 5:32 am
Location: Boston, MA

o2 Operation/theory

Post by bsdkllr »

Quick question about the factory o2 sensor. from my understanding zeroing out cells on the fuel map with no other enrichment's should result in the o2 sensor aiming for ~14.7:1 AFR's. what happens when you have it set to say 8 on the fuel map with the o2 enabled? is it still trying to drive that to ~14.7:1.

at idle i want to get my car around 13.5-13.9 i disable all o2 function and tweak the maps to achieve that on the wideband. turn on o2 sensor and goes right to ~14.7 and causes a lean miss. id like to try and keep the function of the factory o2 if i can for the cruising efficiencies.
Shaker
 

Posts: 38
Joined: Wed Sep 23, 2015 8:00 pm
Location: Germany
Contact:

Re: o2 Operation/theory

Post by Shaker »

From what I experienced the map and all other values and multipliers are used to calculate a "base" pulsewidth of the injectors. Fuel is injected using this time.

If the o2 sensor is reporting a rich or lean exhaust value, the ECU has a certain range to correct this, if the load index currently is in a cell marked as "o2 enabled". It can set -25 to 25 (Short Term Trim), which I think is percent.

My opinion is, you should tweak each cell to a value, so that AFR 14,7 is achieved without o2 loop, as good as can be, and then enable the o2 loop on each cell, where you want it. E.g.: 0-50 load, 500-4500 rpm, take the stock image to get the idea.

In this case, the ECU doesn't need to correct the value too heavy, using Short Term Trim, and the time until the exhaust gas reaches AFR 14,7 is reduced.

With o2 loop completely disabled everywhere, because of driving dynamics you won't be able to get AFR 14,7 reliably, especially when talking about exhaust gas analysis / MOT / TÜV, AU or whatever it is called in your country :)
Matt
Site Admin
 

Posts: 8961
Joined: Sun Jan 29, 2006 1:45 am
Location: Adelaide, Australia
Contact:

Re: o2 Operation/theory

Post by Matt »

The ECU calculates a base fuel injection pulsewidth based from TP (load calculated from MAF / RPM). As TP increases (load goes right) then injection time will increase. The fuel map numbers simply trim these values. This is why I removed 'AFR' display from the fuel map (except in 'expert mode')

So there is an invisible trim map in temporary memory which keeps track of the long term trims. In closed loop conditions the STFT and LTFT will adjust and also update the trim map

Above ^ is correct with regards to O2 flag settings.
bsdkllr
 

Posts: 56
Joined: Wed Mar 09, 2016 5:32 am
Location: Boston, MA

Re: o2 Operation/theory

Post by bsdkllr »

So looking at the idle areas on the stock map have a filtered value of 8. And o2 enablement. So it requires an addition trim in the positive to get the 14.7.

If I enable the o2 on my car I see afr's up to 16-17 when accelerating. But o2 disabled it's in the mid to low 13s. It just seems like the o2 is hampering the operation.
Matt
Site Admin
 

Posts: 8961
Joined: Sun Jan 29, 2006 1:45 am
Location: Adelaide, Australia
Contact:

Re: o2 Operation/theory

Post by Matt »

What are your STFT and LTFT gauges showing at cruise?
bsdkllr
 

Posts: 56
Joined: Wed Mar 09, 2016 5:32 am
Location: Boston, MA

Re: o2 Operation/theory

Post by bsdkllr »

I will try and get a log of those today.
bsdkllr
 

Posts: 56
Joined: Wed Mar 09, 2016 5:32 am
Location: Boston, MA

Re: o2 Operation/theory

Post by bsdkllr »

I'm pretty sure the long term fuel trim is not displaying the correct data. Seems to have no correlation with the short term trim. And almost seems like a engine load. If you put it under heavy load it goes up to a positive number. And when coasting it will drop to -9 or greater. It's also changing quite rapidly. The short term looks to be working correctly. And will bounce +- 3. I'm also using the feature pack roms
Matt
Site Admin
 

Posts: 8961
Joined: Sun Jan 29, 2006 1:45 am
Location: Adelaide, Australia
Contact:

Re: o2 Operation/theory

Post by Matt »

What ECU ID showing in Nistune? I'll check it out

All of the ECU registers (channels) for the Type 1 boards I had to reverse engineer out of the Nissan disassembly by hand. Maybe one of those is incorrect for that ECU?
Matt
Site Admin
 

Posts: 8961
Joined: Sun Jan 29, 2006 1:45 am
Location: Adelaide, Australia
Contact:

Re: o2 Operation/theory

Post by Matt »

Ah... you are 26P0F just read in the other thread. Okay LTFT might be due to the 1986 ECU with 1988 ROM base image loaded
bsdkllr
 

Posts: 56
Joined: Wed Mar 09, 2016 5:32 am
Location: Boston, MA

Re: o2 Operation/theory

Post by bsdkllr »

thats what im thinking. also i cant touch the k value at all or it just kills the engine.

i have been using all the maps from the 87 turbo mt rom and copying them in. i just need to bump the TIM up to 565 do get it to idle with the least amount of misfires. and also timing once warmed up bounces around a bit. but im not sure if thats from the cam and the ecu trying to set the idle too low.
Matt
Site Admin
 

Posts: 8961
Joined: Sun Jan 29, 2006 1:45 am
Location: Adelaide, Australia
Contact:

Re: o2 Operation/theory

Post by Matt »

also i cant touch the k value at all or it just kills the engine
That doesn't sound right. K adjusts TP and injection time

You probably shouldn't copy the 1987 one. However watch injection time as you make small changes to K and see if it increases/decreases. Take a log whilst doing it

So does the car do 14.7:1 AFRs at cruise, and idle at 14.7:1 when you have no o2 enabled/connected?
bsdkllr
 

Posts: 56
Joined: Wed Mar 09, 2016 5:32 am
Location: Boston, MA

Re: o2 Operation/theory

Post by bsdkllr »

i reloaded it with the default 88 fp rom. and again touching the k value up or down just seems to kill the engine. i want to say it stops injecting fuel. i will check again but i think i saw my wideband go to 22.
Post Reply