16v has more power when motor is cold?

[treed2@u.washington.edu (T. Reed)]

Hey all,

Just got back from a drive tonight and I noticed that during the first 2
minutes or so the car had kick ass power.. I mean, I was barely laying in
to the throttle .. probably just a touch over half way, and the tires were
chirping in 1st and 2nd without even trying. After the car was warmed
up I went over the same stretch of road and while it was still fast and
there was no "huge" reduction in power, I couldn't get the tires to chirp
again with the same amount of throttle. Both times I was just doing my
normal smooth declutching action.

I know that's not very scientific, but I'm feeling like my motor is
running "torquier" when it hasn't warmed up yet. Ambient air temp is about
50 F, but I can't imagine that cold vs. warm air could make that much of a
difference, besides.. my engine runs cold to begin with (I never get to
the LED on the water temp gauge)

I do have my mixture set to 4 mA +/- 1 mA.. Stock is 5 mA or 10 mA
depending on who you ask. This is a 2.0 16v with 1.8 head, BTW.

I've noticed that my idle bounces a bit until the car warms up. I've heard
that this can indicate a rich mixture, but I can't rememebr why.

I suspect that what might be happening is while the car is in open loop
mode, the mixture is not influenced by the O2 sensor and is going rich
because of my non-stock mechanical setting of the mixture screw. But
once the car warms up, the oxygen sensor bypasses the mechanical setting,
leaning out the mixture and causing me to loose power..

I replaced the sensor about 1.5 years ago with a universal bosch sensor
and spliced the wires in to the original harness.

So I'm wondering:

- what is the difference between euro mechanical CIS for 16v's and our
  CIS-E running in open loop mode? It seems to me like there should be no
  difference.. in steady-state conditions the mixture is based completely
  off of the mechanical adjust screw when in open loop mode.. it should
  be possible to force the car in to open loop (by disconnecting the o2
  sensor) and tune the screw to keep something just a little richer than
  stoich. This would also preserve all the warm-up and acceleration
  enrichment features of CIS-E (assuming they normally operate when in
  open loop)

- has anyone tried placing a resistor in line with the oxygen sensor to
  cause its output voltage to be divided between the resistor and the
  computer, effectively making the computer "see" a lower voltage (which
  corresponds to a leaner mixture) and thus causing it to richen the
  mixture on its own? It seems like this would work.. if it did then you
  could put a potentiometer in line and tune the mixture to your liking,
  electrically.

I think I might try the latter, and borrow my friend's homemade air fuel
meter that hooks to the o2 sensor and see if putting resistance inline
is indeed effective in changing the mixture. First, though.. I'll need to
put an ammeter inline with the o2 sensor and see how much current is going
through the circuit so I can determine what range of resistance is needed.

Anyway, I just thought I'd share my ideas about this since I know CIS-E is
notorious for leaning things out in the name of fuel economy.

I've thought about buying one of those euro CIS setups for $300-ish but in
my opinion there is nothing mechanically wrong with the CIS-E I already
have.. its more than capable of delivering the amount of fuel (and the
proper amount, at that).. so why junk it? As an electrical engineering
student I'm inclined to find an electrical (read: cheap; components cost
just pennies and are readily available) solution that uses the existing
mechanical parts and sensors to accomplish the same goal.

-Toby