I would not do what RWF7437 recommends. TR55 and Rational will almost always give vastly different Q values, as they have different applications. Usually, I use Rational for small areas (10 acres or less) with a higher percentage of development. Over 10 acres and Rational underestimates, I think. I use SCS methods for mid-sized areas which are more natural and/or agricultural. Anything over 0.5 square mile and TR55 vastly overestimates in my opinion. Try a flood model (HEC or PSU4) above 0.5 square mile.
If you must convert CN to C, Rossmiller's Conversion can be applied. It is a monster equation, but here it is:
C=(7.2*(10)^-7)*(CN^3)*(RI^0.05)*{[(0.01*CN)^0.6]^-s^0.2}*[(0.001*CN^1.48)^(0.15-0.1*I)]*{[(IMP+1)/2]^0.7}, where RI=return interval(yr), IMP=impervious coverage( decimal)and S=average land slope (whole number percent). I have a spreadsheet which I have used (and has passed reviews) in the past, but it has not been QC'd by another engineer. If you leave your email address here, I will send it to you to play with.