High Idle Issue

Hi everyone, I have a different question for you all than you are probably used to hearing. I am part of a college mechanical engineering senior design team that is building a formula car for the Society of Automotive Engineers competition. It is a small car (typically 500 - 700 lbs) that in our case, uses a 2003 Yamaha R6 motor and breathes through a restrictor plate with only a 20-mm hole . The car built by last year's team of seniors idles at about 4000 RPM when its warm. Every once in a while, the car will idle at around 1000 RPM, but as soon as the throttle is touched, it returns to idling at 4000. My idea as to why this may be is the intake design. The intake is basically a cylinder lying on its side on top of the engine acting as a manifold with each intake runner going into the round bottom part of the cylinder. The air is filtered, throttled, and restricted through an opening in the end of the cylinder on the flat part. I feel like since the engine is restricted so much, we are probably starving the last cylinder or two which would cause the computer to bump up the RPMs to compensate.

Do any of you see this as a likely scenario? If not, any ideas on why else we might be seeing such a high RPM at idle? Thanks!

makes sense...
4000 is REALLY a high idle.
although, you cant really complain.
our engine isnt even running at the moment.

damn restrictor plates

Are you running the engine from the stock computer? If so, build yourself a MegaSquirt. The stock computers are wonders of engineering... translated, they only work in a stock setup. I had that exact bike and it took a while for the computer to "relearn" when I put an aftermarket air filter on it. The original configuration had individual throttle bodies. By putting a manifold on it they changed the vacuum and airflow characteristics rather drastically. I think the computer is probably saying something like "what the f** do I do with this information?"

You might be able to find someone on the 'net who knows how to hack the stock computer, but they are much more complicated than most automotive ECMs from what I understand.

we are using a homemade computer.
not sure how they make it, but it works pretty well.

We are using the stock R6 computer (for now anyways). The engine has a good 10 hours of racing on it with this setup, so the computer should have learned by now.