DFI, Batch Fire, and other myths

[Shannen Durphey]

The sometimes less than accurate GM classes I've attended teach that
injection should occur when the intake valve is closed,and be
completed immediately prior to intake valve opening.  This would make
use of the heat in the head of the valve help keep fuel from pooling.
This was mentioned earlier.  This might be GM's approach to sefi. 
According to a service bulletin, the 3.8l engine suffers from carbon
accumulation on the back of the intake valve which significantly
reduces performance.  I've seen the effects. It's very similar to
losing the accelerator pump on a carb'ed engine, and the exhaust is
excessively rich.  My thought was the carbon is insulating the valve,
preventing vaporization.
Shannen

diy_efi at efi332.eng.ohio-state.edu wrote:
> 
> Recent research on fuel droplet sizes has shown that injecting fuel onto a
> closed intake valve, then letting a shot of exhaust hit it when the valve
> opens (reversion) produces better and smaller droplets than the injector
> itself can generate.  See the SAE Journal from a month or two ago.
> 
> With this info, SFI could be used to insure that all the fuel is injected
> while the valve is closed, not open.  How are the SFI systems actually
> timed?
> 
> Gary Derian <gderian at cybergate.net>