[mehyan]

hi dears,

wht is the expression of hydraulic diameter for conical tubes?

InletDiameter = sqrt(4*area()@Inlet/pi)

this is for straight tube


so what is expression for conical tube?

because hydraulic diameter changes along the tube length.

thanx

[tomson199]

Maybee try use the arithmetic average values of inlet hydraulic diameter and outlet hydraulic diameter?

[mehyan]

tanx for your rpy,
i think so ,

but i want a way that calculate each point nusselt number at that point , this may be give more precise value of nusselt number.

[tomson199]

I've got idea. Your conial tube is still circular in each 2D section perpedicular to the axis of revolution. For circular tube hydraulic diameter is a normal diameter. Maybee expression for this diameter should look like this:

d=2*sqrt(x^2+y^2)?

[flotus1]

I think you should ask a slightly different question: Is there any point in defining a hydraulic diameter for a conical tube?

[-Maxim-]

Originally Posted by flotus1

I think you should ask a slightly different question: Is there any point in defining a hydraulic diameter for a conical tube?

I guess to calculate the Reynolds/Nusselt number? What about the diameter of the tube? I once did some research with stuff build into a canal and I used the distance between to geometries inside the canal...

[flotus1]

I guess to calculate the Reynolds/Nusselt number?

If you need a hydraulic diameter to compare your results to some analytical or empirical result or use some kind of correlation in your simulations your only choice is to use the same definition as in the work you are referring to.
Otherwise the choice of a "correct" hydraulic diameter is somewhat arbitrary especially for complex (complex as in "not a straight tube") geometries.

[Opaque]

I agree with flotus1 comment.

Perhaps, it will be good to understand why dimensionless numbers are used:

• Parameter reduction
• Parameter correlation

For parameter reduction, we can determine exactly how many independent parameters our problem depends on

For parameter correlation, we may (luckily) obtain a single expression that correlates (it is valid) for a general set of problems. Let say for example, transition from laminar to turbulent flow in pipes occurs approximately at Re = 2000. If you use the hydraulic diameter, the shape of the cross section of the tube can be eliminated, but if you use another characteristic length, you may end up with very different criteria.

The accuracy of your calculations cannot depend on how you define your characteristic length. By computing Nu number, you are just reducing the data from dimensional to dimensionless, and notice the parameter correlation I described above if any.

I would look to see what characteristic length is being used in the literature for this geometry. However, if your geometry is unique, feel free to pick anything that is simple to explain to someone else.

Keep in mind that Nu, or Re numbers are useless if the definition of the characteristic length, properties and hydrodynamic conditions are not clearly described. If such information is missing, their values mean nothing.

Hope the above helps

[-Maxim-]

I agree with you flotus1 and Opaque. I just tried to _guess_ what user meyhan originally would like to achieve with question.
But I suppose he just has to explain himself...