There is no maximum value established for velocity magnitude in a turbulent flow. In turbulent flow, the velocity oscillations are, until now, not predictable. But if you are trying to reach a mean velocity magnitude value, the limit could be the backward bulk velocity, usually. Obviously, it can occur spots with higher velocities, but it is a magnitude order.

Its depend on your application like rotation or irrotational flow but its value for fully develop ranged 0.5-0.85 %

find the link :

https://www.cfd-online.com/Wiki/Introduction_to_turbulence/Wall_bounded_turbulent_flows

 @Guilherme B. Lopes Jr Thanks for your comment.

@Ghassan Fadhil Smaisim Thanks for your comment and link

Eventually you can make use of the last part of the chapter found in the link "https://www.intechopen.com/books/hydrodynamics-advanced-topics/one-dimensional-turbulent-transfer-using-random-square-waves-scalar-velocity-and-velocity-velocity-i", part 3, FIgure 9. It represents an attempt in the sense you have mentioned. 

This is part of a study in which the behavior of statistical properties of tubulence is the focus. The use of bi-modal random fluctuations allowed some conclusions. As I said, eventually these conclusions can be interesting for you.

For the scalar part of the formulation some theoretical solutions for mean properties were already obtained (and accepted for puclication). Studies are now being conducted for the vectorial part (velocity - part 3 of the mentioned chapter). 

Turbulence is an interesting theme. I hope you can find the best answer for you.

Cheers.

Harry.

@Harry Edmar Schulz Thank you, Prof. Schulz

Hi Farhad,

For a uniform flow, the shear velocity is equal to sqrt (g R i) where R is the hydraulics radio (~water depth) and i the slope of the channel.

Hope it helps