The help mentions that the generalized friction formulation is only valid for slopes less than 10%. What does this mean? Will the calculations not be correct for steep slopes?
Manning’s equation is empirical and so it will break down when certain assumptions aren't met. The further you get from typical river or sewer slopes the less confident you should be in the results. 10% is the generally accepted limit, but having a slope greater than this will not prevent the model from calculating. Meaning, the 10% figure is not a black and white rule (it doesn't mean that 9.99 is ok and 10.01 fails). It means that the assumptions made in deriving the equations become less valid as the slope increases and at about 10% you start to notice it. You will need to use your judgment to decide if the results are acceptable.
Note: As described in the article in the "See also" section below, many numerical models assume that the grade of the channel is small, nominally less than 1:10, and therefore the sine and cosine of the channel slope can be assumed equal to zero and unity respectively, allowing the cos θ term to be neglected from the calculation of hydrostatic pressure (P = ρgd cos θ) and elevation (z = z0 + d cos θ).