Anyone know anything about these?

Except that the O2 sensor would be the final determiner of how much fuel
gets added.

 

IMHO, the main effect of air temp sensor modifications like this is
not on air/fuel ratio but rather on ignition timing. When the intake
air is cold, the engine is less likely to "knock" (or "ping",
"detonate", whatever you like to call it). Therefore you can use more
ignition advance, which would normally provide more power. (As a
general rule, on most engines, the ignition timing is set somewhat
retarded from optimum because of the danger of knocking).

So, if you fit a device like this that fools the engine into thinking
the air is very cold, it will apply more ignition advance which will
probably produce more power. But, it gives a very much greater risk of
catastrophic knock, which can effectively destroy the engine! Normally
there is a certain amount of safety reserve built in and so generally
you would get away with it, but things like this give me the willies,
they really do.

When the engine is cold, a false low AIT reading will probably delay
closed-loop AFR control and the engine will run slightly richer. But
the engine will enter closed-loop sooner or later and return to lambda
= 1 at part load, regardless of the outside air temp. At full load the
engine will generally be set to run at optimum lambda anyway
(typically 0.9) and adding any extra fuel is unlikely to be of
significant benefit.

As far as the air swirler thing is concerned, you may like to visit
http://homepage.ntlworld.com/cains1/Fuel_saving_turbulence.htm for
advice on why not to bother with them!

Tony

(replace "nospam" with "fuel" to reply)

 

Your theory sounds good enough, re wanting to change the timing at cooler engine
temperatures, but from the manuals the actual engine control design does not
appear to work quite this way.

For circa 1990s Hondas, the manuals say engine load, engine RPM, vehicle speed
and coolant temperature are fed into the ECU and then determine "basic ignition
timing." For 1995-1997 Civics, the manuals say ignition timing depends on engine
speed, manifold pressure, and coolant temperature. See for example, "Ignition
Timing" at http://www.honda.co.uk/owner/CivicManual/pdf/11-20.pdf .

Now these aren't VTEC engines, so maybe VTEC takes into account AIT in adjusting
timing.

I have the same feeling here, though it's based only on an impression and
textbook engineering and not much actual experience fooling with actual engine
controls.

 

Agreed, it almost certainly depends on the engine. I actually set up
engine management systems for a living, and all the ones I work on
take AIT as an input to ignition timing. But these are on very new
vehicles and certainly some older vehicles with less sophisticated
management systems may well not take AIT into account.

Of course, the manual doesn't always give the full details of how the
system works either. The complete description for the management
systems I work with runs to 3500 pages - which is obviously highly
condensed for the vehicle workshop manual!

Tony

 

You better believe I would have bet beaucoup dollars on this before reading your
post.

I would have guessed more than 3500 pages... ;-)

I've also been figuring the last few months (since I started studying the
details of auto emissions controls) that most of the information about each
model's design is most certainly proprietary.

(I did some control systems design work in grad school, with a touch of bona
fide consulting but as importantly, with occasional chatter with some Ford
emissions control system design-specialized engineers. They're the ones who gave
me some exposure to the practical side of emissions control design.)