EFI and Two-Strokes

[jb24 at chrysler.com]

>Ed Mellinger wrote:

>Regardless, the point is that the valve timing can be optimized better
than in a port-only design.  I >always figured they probably scavenged
(or blew down) better than a port-only design, too, but >I've no
numbers to back that up.

At the Motorsports Engineering Conference last Dec., a paper was
presented that had experimental evidence pertaining to the Discharge
Coefficient of poppet valves at various pressure ratios.  At high
pressure ratios (like at exhaust opening), flow is masked (restricted)
by  the cylinder wall near the valve if the valve diameter is too big,
which is often the case when trying to get more valve area in a given
bore.  Anyhow, this would lead me to believe that a poppet valve is not
quite as good as a port, flow-wise.  I have done calculations for port
opening areas, but not for poppet valves, so I know you can get a lot
of area very quickly with a port, again probably better than a valve.
I think the primary reason for valves is asymmetric timing, i.e.
exhaust open - transfer open - exhaust close - transfer close.  Anybody
know how fast Detroit Diesel two-strokes spin?  2000, 2500 rpm max?

Reading the info on the diy_efi web page makes me think this hardware
would be good for me, but I guess I am still ignorant about the
intended sensor inputs.  I was thinking I would need rpm, tps, MAP,
MAF, HEGO and coolant temp.  Is this hardware intended to be generic,
wire a few sensors in, and then program around them?  Really, I would
like a little data recording for development purposes.  Is the hardware
intended to do adaptive fueling, and if so, could this memory be
accessed for data logging?  I appreciate any info.

John Bucknell
jb24 at chrysler.com