Air/ Fuel adjustment, idea/ MOD

12th April 2010, 11:50

Ok, so here is the thing- I think my Saxo runs a bit rich. Even if it didnt, I would like to know how it is running.

Then knowing this, it would be handy to have some sort of adjustment.

Also, it would be good once that is set up to be able to 'cancel' it.

----
So from what I know or dont know, here is what I have ben thinking, but I need a bit of info here and there to fill in the Gaps.

Ok, so on Ebay there are the horrible tuning mods to 'fool the ecu' which we all know and hate (the resistor).
There is also another one which you can tweak either way (bear with me here) to give less and more resistance.

They use the air temp to fiddle with the ECU- the different resistance working into the ECU delivers a diffent air fuel mix.

The main problems are in reality:
Even before tampering with the fuel mix (via the air intake temp) you dont know what your burn ratio is anyway as you could have:
Slight misread from Lambda Sensor (being caked up), dodgy air intake temp, spark plug(s) with a slight adjustment due to age, injectors with deposits on them etc. Lets not get bogged down with that so much as you could argue with the above:
but lets agree that certainly a car with some proper 'age' to it OR/AND possibly free flowing exhaust, breather filter, decat pipe, induction kit of some description etc. that we dont actually know our O2 reading and secondly, without a way of adjusting it even in a basic fashion there is little we can do without going to a standalone system, having it mapped etc.

Before I go on, I assume that there is nowhere on the system to just turn a screw to adjust mixture throughout the range? Right?

Also, I know that the Lambda sensor reads all the time giving a voltage between 0 and 1 volts.

So-
Do we know what the 'correct' voltage would read from the Lambda sensor is back to the standard ecu- for say 'around 14' (i know the ecu takes this and reads it as 'rich' or 'lean' but the actual sensor is changing all the time)

If not, can people read their sensors to see what their voltage is reading?

Then, if we had a tuning method we could see what the results are by reading our standard Lambda sensors with a multimeter- maybe try and get a graph or table of comparisons between 0 and 1 volts to see what is going on.

Secondly, with the correct variable resistor plus another resistor 'wired correctly' (not just added in series so that adding resistance is the only thing you can do- so like series, parallel, I havnt done the maths or electronics here yet)_ then we could adjust the overall map.

What this does 'instead of' is to make a system to monitor the burn rate for the price of a multimeter (instead of a wideband system (150 pounds plus))
and have a simple modification to make a best fit adjustment if the fuelling is out.

12th April 2010, 11:59

Oh yes and the 'cancel' idea I have is to wire in a switch so everything would disconnect.

Reading the Lambda 0-1 volts should be fine in parallel and not warp the reading back to the ECU. Think thats it!

12th April 2010, 12:01

This is obviously not the ideal mapping situation but one where a bad mix could be detected/ known and adjusted in a blanket fashion using the air intake temp levels to tweak the system.

12th April 2010, 15:26

Looked into how to read the signal from Lambda- it should read rich lean rich lean- all the time but will actually average .45 volts if correct as an average.

Multimeter wont read this as to log it and read it you would need to look at an oscilloscope- but interestingly enough you could use a line in on a sound card and record the graph onto some audio software- easier and more user friendly to go down the 'wideband' route though!

12th April 2010, 16:09

so after all that what is the exact question you are posing.
if it is can you alter at will the signal seen by the std ecu ,by modifying the lambda signal ,and be more accurate than the std ecu -then answer is no .
can you read value given by lambda sensor 10 times a second with a norrow band lambda --NO
is it acurate at anything other than from 13.5 to 15 --no
I can see you have done alot of thinking --but i would suggest you put that brain power to better use and fit a stand alone with a w/b and tune it correctly to what ever value you wish

12th April 2010, 23:39

Oh dear lol.

13th April 2010, 00:27

I like your thinking. But, if I've read your OP correctly, you can't really alter the output since as you say, a narrowband lambda is either saying 'It's LEAN!!!' (say 0v) or 'It's RICH!!!' (say 1v). It's behaviour is almost digital....

see blue line off this diagram I nabbed...

So therefore you couldn't add a resistor to modify the signal to the ecu since the ecu is 'sniffing' for either a 1v or a 0v. Add a resistor and the voltage will then switch between say 0.7v and 0v, so essentially you're telling the ECU over a period of time, that you're more lean than rich... meaning the fueling will increase only, increasing the burn rate dude.

This will not the increase the power or stabilise the idle or since WideOpenThrottle and Idle are open loop configurations respectively i.e. the signal from lambda is ignored.

So ultimately, your cruise conditions will be modified and that's when you want the AFR to be at stoichiometric (for the sake of the Catalyst working optimally) or slightly lean for better mpg (albeit worse catalyst performance).

But top stuff anyway!

Ads

13th April 2010, 04:20

I didnt ask about modifying the Lambda output but about reading it. The modification of the fuelling would be from the Air Temp intake (as in the post).

No, you cant read the narrowband Lambda in terms of a voltage (doh), you could in terms of bandwidth or rather pulsewidth, but there is little point going into that as working it out accurately you might as well fit a wideband and read from that and get a true result. So narrowband, it wouldnt be practical to read from.

Although you write this from the point of view that you think i want to fiddle with the lambda output- it actually contains good info which I also stumbled upon last night when writing this:
"This will not the increase the power or stabilise the idle or since WideOpenThrottle and Idle are open loop configurations respectively i.e. the signal from lambda is ignored."

Here is a question back- in open loop for full throttle, is it taking into account the air intake temp on the map? I am guessing it should as a super cold or very hot day would make a difference?
------
If it does, then this answers my original post which was, know how its running (fit wideband to monitor)- and adjust accordingly if it is 'always' rich or always 'lean', not hard to do by fitting resistance to adjust up or down in addition to the air temp sensor. Then I just want a switch so I can return to bog standard, which might also be good for comparison.