When is too big to big

[Johnny]

At 10:33 PM 11/28/95 -0600, you wrote:
>>A question for all of us interested in vagaries of injection.
>>
>>I put Injection on my race car and lost 40 horsepower.
>>
>>History
>>Engine  ISUZU DOHC 2 valve per cylinder 4 cyl normally aspirated
>>        2 litre
>>        Fuel 100-103 octane leaded AvGAs
>>        Exhaust 1.75 primary 1.875 secondary 4-2-1
>>        Carburettors    2 x Weber 48DCOE with 2 inch long ram tubes
>>                        and 3 inch spacer from port to carby
>>        Ports size 39 mm valve 44mm
>>        Approx horsepower at flywheel 200
>>        Approx horsepower on rear wheels 145
>>
>>I changed nothing in engine internals , exhaust, fuel and then installed
>>following injection
>>
>>        50mm individual throttle bodies, Injectors from Nissan FJ20
>>Turbo rated to 240 bhp flow. Autronic sequential engine management (
>>Very Similiar to Motec designed by the original motec designer).
>>General motors bosch type fuel pump. 2.5 bar regulator mounted at end of
>>fuel rail.
>>        Effective inlet tract length is 7 mm longer then carby setup.
>>        manifold tapered from buterfly linear to port opening. Dowelled
>>manifold to head.
>>
>>After 12 hours on dyno maximum power at rear wheels 110 bhp ( down 35 )
>>bolt carbies back on and horsepower returns.
>>
>>        CO sensor shows good mixtures on the injection 4.5% 4.8% from
>>4000 to max 8000 rpm. and power loss is pretty much all through the rev
>>range.
>>
>>        Can only suggest the following ideas
>>        1.      Inlet increase at butterfly from 42 to 50 mm
>>                and lack of venturi effect is stalling air
>>                speed in inlet tract. 2 other cars I race against use
>>                50mm throttles..an FJ20 16 valve 2 litre nissan ,
>>                and an A15 1500 8 valve overhead valve motor .
>>                both with no problems
>>        2.      Some type of pulsing problem ( not sure how
>>                i would work this out).
>>        3.      Computer design is bad ( have tried a seperate
>>                new one as well with same table in it
>>                so not faulty)
>>
>>        I would be happy to provide as much info as is requested but not
>>sure where to start.
>>
>>I would welcome any comments suggestions as i have spnt a fortune
>>already on the dyno and rebuild an engine because of the time on dyno.
>>
>>Regards
>>Doug Robson
>>Club Car Racing Register of NSW
>>Sydney Australia
>
>I can most likely guess that the problem is in the dynamics of the intake,
>not necessarly the size of the throttle boddies.  You might try connecting
>all of the tubes from the cylinders to a common plenum and placing the
>throttle body on the plenum.  Also try monitoring the exaust with an o2
>sensor and guage instead of monitoring for co2 content.  Find a mixture of
>about 12 to 1 for gas or 8 to 1 for alcie.  I have experienced similar
>problems on a v8 before.  Also, the fuel curve in the computer may not be
>right.  I know that it might sound crazy, but you might also try throttle
>body injection.  Quite often it yields better mixture, especially at high
>revs.  Hope it was some help.

uh... nix on the O2 sensor, dude is runnin' leaded avgas, r'member? I have
tried using O2 sensors with leaded. They seem to get erratic after a short
while. I agree that the problem probably lies in the intake. The plenum
method might be a good one to try, the other would be to experiment with
changing the taper on the "stub" manifolds from linear to some other shape,
like a bell or tapered step. Or you might want to change the length of the
runners. The other obvious problem would be the injector sizing or
controller program itself. It's a bitch not having a box you can just hook
up to your PC while it is on the dyno. Makes it a lot easier.

-j-