Richer on higher IATs?

[AVP]

I may be completely wrong here, but i need some advice from the more experienced in the field.

I have been told that the stock ECU of the Audi, tends to make the car a bit more rich when the IATs become higher so as to keep the same timing(?) without detonation or any risk of that. I am not sure that this applies for both on and off boost, as i would expect the motronic to have a closed loop narrow lamda system for cruising and lamda would be kept on 1 roughly all throughout.

On my VEMS i have found a nice equilibrium of cruising at about 1.09- 1.03 on the midrange with equivalent timing that runs perfectly ok up to 25C IATs. however i get the impression(and that is just my feeling) that whenever the car is hotter in any way and iats go beyond 30C, the current lamda values i use are making the car appear as it's hesitating a bit. As if, the car would like it being closer to 1 lamda i suppose.

Timing wise, i tend to have it steady on low TPS without taking much more than 1-1.5deg but above 15% TPS i take it to -2deg to 2.5deg timing.

Fuel enrichment obviously and according to the standard is less than 100 on those areas for the EGO to keep the target lamda, and its also nicely calculated, so i dont see much movement around 100 on the display.(ie EGO is not adding or taking away much fuel).

My question is: is it possible that the car is running happily on colder weather up to 1.09 lamda and having nice fuel consumption, while when it gets hotter, it needs a bit higher target lamda to be able to be 'happy' again?

If so, then i see no way of accomodating this, as you only have 1 target lamda and cannot alter it with IATs. I could increase the fuel enrichment table, but EGO would still take the lamda towards the target anyway, OR i could make all my target lamda richer, sacrifice some fuel economy when the weather is cold and make sure that the car runs nicer on hotter weather.

Am i going to far?

[rob@vems.co.uk]

Its a good question - how will the target lambda's effect the erichment, I *think* that the enrichments are applied after the initial PW calculations  - PW = VE/Lambda target.

[gunni]

When you think about fuel economy lambda is not a good enough sign. As it value is only proportional to the fuel and air at any given time.

Fuel economy comes from running the least pulsewidth you can. So here is a test.

What is the LEAST amount of pulsewidth you can use to cruise at 70mph. Is it the same pulsewidth if lambda is different?
i.e the load is higher and so is lambda versus less load and lower lambda (richer).
This test will show you what you should really be aiming at.

Engines have less pumping losses with higher manifold pressures, this suggest run as much load as possible with the least amount of fuel.
I.e higher KPA and higher Lambda. but the BSFC differene may be minute. And therefor running lean with higher load is useless.

Can you post a log of your lambda when the IAT goes above 30C°?

[AVP]

very interesting data Gunnar, i will have to look for such a log.

once i have the car back i will also do the test you recommended in order to see what the PW is on specific lamda.

i do however increase the lamda taking into 1.03-1.00 territory as load gets closer to 100-130kpa. Its something these audi engines like, and work will with, without needing to go richer. The trouble is that this seems to brake down as IATs get higher, and its either a matter of reducing timing to make the engine more viable on that lamda target, or making it richer for that timing..kind of thing

[gunni]

What exactly do you mean by "break down"

You have to give the engine what it wants, not what you want to give it.

If it needs more fuel then give it more fuel.

[AVP]

well i dont know if that is what it wants, or if it is less timing! hence another good reason for a load dyno session

[gunni]

give it some fuel and see what happens .