Re: CIS Cheap Power

[Dave Weiss <dzweiss@worldnet.att.net>]

Angela Dupin <dupina@crisny.org> wrote:

> All this talk is making me dizzy.  is this talk of rich mixture and 
> fooled flow meter truth based on actual measurement and experience or 
> just conjecture?  my take on this is that whether the air is cold, 
> warm, the air filter is plugged or the air is shoved in the flap will 
> respond to the mass of air and adjust the mixture accordingly.  Show 
> me!

The later FI's do use a hot wire mass flow meter.  The older CIS and
etc. use a variable area flow meter which is sensitive to density.  If
you're dizzy now, skip to last two paragraphs to avoid a massive
headache!

<Non-nerds begin skip here.>

The formula is a bit ugly, but with several simplifying assumptions,
including assuming the discharge coefficient (meter "constant") is
actually a constant in the range of interest (and no, I haven't
calculated the Reynolds number in the annulus to make sure!), mass flow
rate through the V-area meter can be expressed as

w = K * (Annular Area) * sqrt(Air Density)

where w is mass flow rate, K is a constant that can be calculated from
the meter geometry and float material,  Annular Area varies with the
height of the float in the body, and Air Density (at inlet) can be
calculated from the Ideal Gas Law.

So at sea level, and assuming negligible pressure loss in the ducting
ahead of the meter:

<Ok, it's safe now.>

Normally the meter sees 14.7 psia, and the fuel curve assumes this.  If
you add just 1 psi from the ram, the meter will indicate sqrt(14.7/15.7)
or 96.8% of the actual value.  When you're open loop, this makes a
difference.  

Temperatures other than the calibration temperature would cause similar
effects (32 F vs. 60 F would make the meter off by 2.8%). Pulsed
injection systems typically correct for air temp; I haven't seen this
correction on CIS, and don't know why not!  (Well, except that it's not
there...)  Altitude is typically compensated for in CIS.

HTH,
Dave Weiss
'91 V8 5-spd
'93 90 CS UnQ