You should always allow culvert analysis, which is more conservative...or at least paints a more complete picture, than pipe analysis.  Increased length is accounted for in the equations, I am not aware that it ever affects accuracy.

If done correctly, cuvert analysis accounts for headwater, tailwater and friction.  It also will show when inlet, outlet and friction controls.  If done during iterative design, it will show when one switches to the other.

Pipe analysis applies a friction loss, and can (but doesn't always) apply an assumed loss due to turbulence at the inlet and outlet of each pipe.

Engineering is the practice of the art of science - Steve

A culvert is typically a single run of pipe or box section going under an embankment.  Maximum lengths and diameters should be specified for maintenance purposes.  If you need a manhole for maintenance or for changing alignment, than I would call it a storm drain.  


if done correctly, pipe analysis should also account for the same variables as culvert and might possibly be more accurate.  For instance, all storm drains end in some sort of outfall which should be analyzed for the appropriate tailwater condition.  Appropriate freeboard should also be provided at the inlets to account for the inlet losses. Storm drains should also be analyzed for other losses such as bends, junctions, manholes etc. which are not normally accounted for in the typical culvert analysis.