open TELEMAC-MASCARET The mathematically superior suite of solvers

Dear all,

I know the friction coefficient will significantly affect the velocity accuracy.

I also notice the most use is the Strickler coefficeint St. and the Nikuradse bed roughness Ks,

I know the St =1/n; and n we can get from a table list.


Bur for the Nikuradse bed roughness Ks

Cd=2[K/log(12h/Ks)]^2

I know K is karmen cosntant and equal to 0.4, but I am not sure how to define the number Ks. Ks=2.5*D50 ?? is it right or not? or there are some table list for the Ks defination ?.any paper recomment about this value ?

I also confuse about the relationship between Nikurase bed roughness and the figure below

For example, in some cases, we need define both

'FRICTION COEFFICIENT' and 'ROUGHNESS COEFFICIENT OF BOUNDARIES' based on the Nikuradse formula. we should have a suitable range for the coefficient for diffirent materials.

Hello,

it is a interesting question and defining the roughness values implicates a lot of uncertainties.
The picture shows the Moody diagram for laminar and turbulent flow with the coefficient of friction lambda as function of Re, Nikuradse roughness height epsilon and pipe diameter d. It says for example that at high Reynolds number the coefficient of friction isn't dependent anymore on the Reynolds number. Acutally we could use this diagram also in free surface flow modelling for the calculation of the friciton losses but we don't do that. In most of the free surface flow cases we are in the fully turbulent rough regime where the formula for the quadratic friciton coefficient which you mentioned, is valid.

It is hard to recommend values for ks. You will find a lot of relationships between ks and d50 or d90 or relationships which accounts also for bed form heights.
A formula for converting between ks and Strickler is for example:
kst = 26/ks^(1/6) accotrding to Meyer-Peter and Müller but other researcher propose also other values for the numerator in the range of 21 - 26.

Hope I could shed some light on this issue.
Clemens

Dear Clemens,

Thanks so much for your quick reply, I think the parameter kst is the stricker coefficient which is range from 20 (rough stone and rough surface) to 80 m^(1/3)/s (smooth concrete and cast iron).

I also notice St = 26/ks^(1/6) in somewhere.

Hi,

yes, kst is the strickler coefficient.
For the assumption of the ks value, I usually use 3 times d50, with was suggested by van Rijn for the flat bed.

Best regards,
Gabi