Re: More on the lambda

[Orin Eman <orin@wolfenet.com>]

> Orin writes:
> >Here is my take on the issue:
> >At WOT, they are dumping lots of extra fuel in to do two things, produce
> >more power and to keep temperatures down.  They calibrate how much at
> >the factory then the MUST ADD EXTRA to ensure they are safe ie. not too
> >hot.  This may take them past the optimimum mixture for power...

> I'm not sure I'm totally with you on this thinking.  Adding extra (beyond
> L=.86) has a huge risk associated with it.  Flame throwing...  (hey let's make
> a EPA inspector wet his pants:).  Specifically, beyond L=.86 you have raw fuel
> in the EM.  This washes catalytic converters, O2 sensors with raw fuel.  This
> actually lowers CAT temps, which increases HC, CO and NOx emissions.  Then,
> after the car goes sub-WOT, the heat builds up again, and actually melts the
> CAT honeycombs (I remember seeing the charred innerds popping out of Ned's
> beast a couple years ago) which then can either clog the cat or move the charr
> downstream and clog the mufflers.  Meanwhile, all that heated raw fuel makes
> for neat flames at night, which combined with red hot turbo temps, fries the
> O2 sensor, as well as the muffler filling.  Sound expensive?  For both the
> customer and the 100k warrantee boys at audi.  Beyond .86 lambda, the line up
> in HC is linear.  Doubt audi goes there ever, in the design of EGT or Lambda
> values.

OK, I did say may... seems that it happens on occasion!

Anyway, I pulled the article I assume was the source of the EGT info, 'The
Audi Magazine', Spring/Summer 1998.  The EGT info is actually one of the
later things mentioned.  Here is a brief summary for anyone that is
interested:

More metal between the bores by reducing bore from 82.5 to 81mm.
Double bolts on main bearing caps.
Sodium cooled exhaust valves.
Revised inlet port shape to generate tumbling motion.
Dual KKK turbos - dual means closer to the engine where
pulse energy is greater.  Smaller turbos - less inertia.
180,000rpm at full load!
1.85bar max boost.  Dual intercoolers reducing temp by 70-80C
(two figures given depending on load).
Now the EGT... quote:
"One new feature used on the S4 for the first time by Audi is an exhaust
temperature feedback control for the engine management system.  To protect
the turbine housing and manifold from thermal overloads, conventional
turbo engines enrich the mixture whenever the operating conditions might
put it at risk; without a temperature sensor, this requires a generous
safety margin.  'Even at full load we can now run at stoichiometric
air/fuel mixtures which reduces emissions and imnproves fuel economy.
It takes time for the temperature to build up and we only add fuel when
it actually starts to exceed the allowed limit.  Then we only use the
minimum amount, monitoring it adaptively to match the conditions...'"
[And if that doesn't work, probably back off on the throttle too
with that fly-by-wire system.]

More summarised stuff:
New Motronic ME7 with drive-by-wire.
NO variable intake geometry, but variable cam timing is retained.
Dual exhaust each with dual cats.

> As you know Orin, Lambda values correspond to Stoich.  1.00 = stoich or 14.7:1
> air to fuel ratio.  A range at WOT between stoich and 12.7:1 (.86 lambda)
> would give better performance than a fixed DC.  Once we can get .86 lambda at
> WOT as easily as the .5 volts we measure lambda of 1.00, the tuner game on
> fuel curves is over.  

Looks like they are going for stoich (or 1.0 if you prefer) and the
EGT is to keep temps and emissions in control.  There was also mention
of meeting 'the latest EU III-D' standard which entitles German
owners to significant tax concessions.  No wonder the emphasis
on emissions.  It does seem that the fuel tuning game ain't over yet.

> It will happen, it's only a matter of time.  Until then, hey EGT works
> "better" than fixed values, let's use the better bandaid.  Let's just not lose
> sight of the target with all audis self promotion.  The concept is simple, the
> execution, it appears, is not so.

I think we got an initial over emphasis on the EGT aspect of this article.
No need to worry about .86 lambda when they are shooting for 1.00 and
low emissions.  A regular O2 sensor can handle 1.00.

BTW, there is development on the diy_efi list on a wide range O2 sensor...
it will be interesting to see how well it turns out, but it _should_
handle .86 easily.  Only problem I've heard of is it may be pressure
sensitive which would be a problem on a turbo engine.

Orin.