IAC logic "Topic of the week 2/19/12"

adds to, but definitely not a significant portion. how large is the smallest cross-section in the IAC path? that's the peak flow, assuming the IAC is fully opened.

Rough measurment of IAC opening on this TB is .40 inch, so not much...

assuming no other flow restrictions (and we know there are), a .4" hole has a total area of ~.125 square inches and can flow roughly ~11.6cfm at most.

Quote:

Originally Posted by RobertISaar

assuming no other flow restrictions (and we know there are), a .4" hole has a total area of ~.125 square inches and can flow roughly ~11.6cfm at most.

And if you fully retract the IAC at WOT that 11.6 cfm helps a bit :) Not so much on a bored TBI as I and some others are running , but it helps nevertheless .

TOM

i guess it really depends on the application.

in the 60V6 world, all small port (160HP) 3100s come with a 52mm TB that flows 308cfm.

assuming the passage is the same size, that's 3.77% of total airflow. not really significant in our application.

It probably isn't significant in the BBC world either ,when you are talking a 900CFM+ TBI unit it is about 1% or so , not alot , but when you are on the edge already , i'll take what I can get :)

TOM

Quote:

Originally Posted by EagleMark

One point made by daleulan that I thought interesting is mention of Idle, VSS control of IAC to prevent them from wearing out. I haven't done enough data logging with masks with Idle and Off Idle tables to notice but he's saying when switched to OFF idle IAC would stop moving. This would not only help wear of IAC but also force more air over throttle blades to further mix/atomise fuel and air in TBI units instead of air coming through IAC which would do less, IMO. My 1227747 $42 has one table and IAC moves constantly even at highway speeds. My LT1 $EE has 2 tables like the $OD, $OE stuff but I never paid attention to IAC counts other then idle. On MPFI injected air through IAC or throttle plates would be of no gain for fuel atomization...

The IAC still will move around during driving - a throttle follower sort of mode to ensure that it is ready for the dashpot function. The wearout issue is more for a steady-state idle. If using the IAC for a true PID control, the derivative term (and proportional, too) will cause the IAC to move up and down rapidly - a couple steps either way, but ten or twenty times per second - trying to maintain the idle speed spot-on. By putting a deadband in the proportional and derivative airpath terms, the IAC can move much less - only by the integral term and then only if you apply a load to the engine from the D-term and P-term. The spark idle control takes care of the small torque adjustments - less than two or three steps of equivalent torque. A deadband makes for an unstable loop so you take the deadband in airpath and apply a correction to sparkpath instead. Plus, sparkpath has a fast response time - the worst case is one calculation loop time plus one cylinder event time. For airpath you have one calculation loop time, an intake manifold filling time constant (not a delay but a filter) followed by a one engine revolution delay. That time lag can make a PID controller less than thrilled. You can lower the gains like the older ECMs did, but people seem to demand better idle speed control these days.

The sparkpath proportional+derivative logic plus the airpath deadband should keep the IAC position from toggling very rapidly... 56,55,54,57,56,55,56,55,57,54,55,55,54,56... all in one second. If it's calibrated right, the IAC should be stable and you'd see somewhere between maybe 3 and 10 crank degrees of spark timing 'jitter' that goes away when you hit the pedal. I was stunned when I put a timing light on my 2001 VW - at how much timing dither there is at idle. Some engines I've worked with needed only a few crank degrees of jitter to make the IAC dithering go away (for most of the engines I work on it's a drive-by-wire throttle but it amounts to the same thing). Some engines seem to take 15 degrees of jitter. I have a 'idle airpath I-gain vs. error', 'idle airpath P-gain vs. error', 'idle sparkpath P-gain vs. error', 'idle airpath D-gain vs. error', and 'idle sparkpath D-gain vs. error' gain tables and they all end up being wrenched around differently for different engines and intake systems. A big variable is the ratio of the intake manifold volume and engine volume.

I would guess that the flow of the IAC is probably pretty insignificant since the Cd looks to be pretty bad with all of the twists and turns in the IAC passage. It was designed to flow under sonic conditions where essentially only the orifice size matters (manifold pressure below about 48 kPa absolute at the current barometric pressure of 89 kPa). I guess someone could flowbench it, but I'd be surprised if there was much of a difference in engine airflow with it open or closed when the throttle is wide open. I don't have a flowbench and all of the engines I have on dynos here are drive-by-wire so I can't test the IAC airflow on any of the engines I can run.