Skin-Friction Formulas

[yipster]

maybe a dumb question but how much difference in resistance you get moving a boat from say 5 to 20 celcius waters
thought boat was planning slower in warm water wich struck me as counter intuative
density, displacement etc differs notably?
started asking but never really got a satisfying answer

[mcollins07]

According to Alberty & Daniels, Phyiscal Chemistry text book, 5th edition p. 572. Table 18.1

Viscosity of water in (Pa s) or ( Kg / m sec).

temp./deg. C. -------- Viscosity
0.0 ---------- 0.001793
25 ---------- 0.000895
50 ---------- 0.000549
75 ---------- 0.000380

also

1/n = A exp( -Ea / RT )
Where n is viscosity, Ea is an activation energy related to chemical properties, R is a constant and T is temperature.

So, it looks like viscosity decreases significantly in the temperature regions of interest. Viscocity is decreasing with an increase in temperature.

[PI Design]

Hi,
Very late to this discussion...
As I understand it the 1957 ITTC line skin friction is based upon a flat plate. However, of course, the plate tested had finite (thin) thickness, so the friction measured was not just skin friction.
A colleague of mine did his PhD in this area and performed some CFD on a truly flat plate (i.e. no thickness at all). I believe the outcome was that the slope of the ITTC line was okay, but there was a constant offset between his results and the ITTC.

[Leo Lazauskas]

Originally Posted by yipster

maybe a dumb question but how much difference in resistance you get moving a boat from say 5 to 20 celcius waters
thought boat was planning slower in warm water wich struck me as counter intuative
density, displacement etc differs notably?
started asking but never really got a satisfying answer

I would guess that the drag should be less in warm water.
But maybe something else is going on. I don't know how the efficiency of props changes, and how cavitation might be different at different temperatures.

Good luck!
Leo.

[Leo Lazauskas]

Originally Posted by PI Design

Hi,

A colleague of mine did his PhD in this area and performed some CFD on a truly flat plate (i.e. no thickness at all). I believe the outcome was that the slope of the ITTC line was okay, but there was a constant offset between his results and the ITTC.

It always amuses me to see an opposite view

The Gothenburg 2000 Workshop on Numerical ship hydrodynamics suggested that the ITTC line has an incorrect slope, e.g. in their discussion of scaling effects:
"The predicted form factor was consistently much higher at full scale compared to model scale. Part of the explanation may be an erroneous slope of the ITTC57 correlation line...".

Leo.

[DCockey]

Originally Posted by PI Design

Hi,
Very late to this discussion...
As I understand it the 1957 ITTC line skin friction is based upon a flat plate. However, of course, the plate tested had finite (thin) thickness, so the friction measured was not just skin friction.
A colleague of mine did his PhD in this area and performed some CFD on a truly flat plate (i.e. no thickness at all). I believe the outcome was that the slope of the ITTC line was okay, but there was a constant offset between his results and the ITTC.

Every CFD calculation method I've heard of for turbulent flows uses a turbulence model which depends on some arbitrary coefficients which someone had to pick. Frequently the coefficients were picked, perhaps in the distant past, based on what provides the "best" agreement with some data. So depending on how the coefficients in the CFD method used were picked agreement with the ITTC may not be surprising. So without knowing where the turbulence model came from and how it's various constants were determined

David Cockey

[Leo Lazauskas]

Originally Posted by DCockey

Every CFD calculation method I've heard of for turbulent flows uses a turbulence model which depends on some arbitrary coefficients which someone had to pick. Frequently the coefficients were picked, perhaps in the distant past, based on what provides the "best" agreement with some data. So depending on how the coefficients in the CFD method used were picked agreement with the ITTC may not be surprising. So without knowing where the turbulence model came from and how it's various constants were determined

David Cockey

Quite right, David.

A particularly dubious instance of this sort of thing is the estimation of the two constants in boundary layer analysis, i.e. the von Karman constant "kappa" and the intercept "B0".

In 1968, Coles analysed 500+ boundary layer profiles and found that the best values were kappa=0.41 and B0=5.0. Those values are accepted by many as the "true" values because of Coles's very comprehensive analysis.

Unfortunately, Coles assumed kappa=0.41 and B0=5.0 in his fits to the boundary layer profiles, so his analysis was not independent confirmation of their validity. Other choices for the pair of constants can give fits that are just as good as those of Coles.

Unfortunately, that doesn't stop some people from dismissing out of hand values that differ from those found by Coles. Funny world eh!

Leo.