[MNMK]

Hi,

I am trying to recognize my flow type in an open rectangular channel. I would like to calculate the Reynold number in a length( y direction) of my channel on a line. How can I write an expression which shows me at every length the Reynold number? In fact, I would like through this expression and define a variable see the change of reynold number in a length of channel.

tanx

[Opaque]

what kind of Reynolds number are you trying to compute?

For pipes, Re number is based on the mean velocity which for incompressible flows is constant through a pipe. The kinematic viscosity does not change much with pressure, and a bit with temperature; therefore, you are looking for tiny variations.

For a constant cross-section pipe, the variations in pressure are not large enough unless you are near choking conditions for gases.

If you mean an open channel, free surface flow? Are you experimenting with large changes of depth along the channel?

[MNMK]

Originally Posted by Opaque

what kind of Reynolds number are you trying to compute?

For pipes, Re number is based on the mean velocity which for incompressible flows is constant through a pipe. The kinematic viscosity does not change much with pressure, and a bit with temperature; therefore, you are looking for tiny variations.

For a constant cross-section pipe, the variations in pressure are not large enough unless you are near choking conditions for gases.

If you mean an open channel, free surface flow? Are you experimenting with large changes of depth along the channel?

In my channel, I have ribs which I would like to see their effects on my heat transfer coefficient. My result shows that my heat transfer coefficient decreases up to a specific length and after that I have an increase.The Re number is interesting to interpret the trend of Heat transfer coefficient according to the change of Re. My question was how could I write a Re expression to see the change of Re in the length of channel?

[ghorrocks]

It sounds like you want a definition of Reynolds number which measures boundary layer growth along the duct. Is this correct?

In that case you can easily set up a variable which uses the length along the duct and flow velocity and fluid properties to give you Reynolds number versus position.

[MNMK]

Originally Posted by ghorrocks

It sounds like you want a definition of Reynolds number which measures boundary layer growth along the duct. Is this correct?

In that case you can easily set up a variable which uses the length along the duct and flow velocity and fluid properties to give you Reynolds number versus position.

Hi,

Yes exactly. I tried the same but I did not work. Can you please write an example which I see where my problem is? (exp. In y direction)

[ghorrocks]

? Puzzled.

Then can't you just write the Reynolds Number equation, like Rho*Vel*y/Visc, where Rho is your density, Vel is Velocity and Visc is viscosity? This assumes the duct starts at length =0 at y=0 and increases in the positive y direction.

[MNMK]

Originally Posted by ghorrocks

? Puzzled.

Then can't you just write the Reynolds Number equation, like Rho*Vel*y/Visc, where Rho is your density, Vel is Velocity and Visc is viscosity? This assumes the duct starts at length =0 at y=0 and increases in the positive y direction.

I think it should be fully clear from my messages that my aim is to evaluate the Reynold changes during the changing of all parameters( Roh, Dyn Visco, Length and Velocity)in one direction. !! In your function what about changing velocity or Density?

Come back to the topic. I have tried to write all of them as a variable which belongs to the direction but it did not work. Can you write an expression to help me to find my mistakes?
I think I have a problem writing an expression in which a variable depends on length.



Please do not say the density or other parameters can be constant.

regards,

[ghorrocks]

I think it should be fully clear from my messages that my aim is to evaluate the Reynold changes during the changing of all parameters( Roh, Dyn Visco, Length and Velocity)in one direction. !! In your function what about changing velocity or Density?

All the variables I used can be either constant values, or values changing with time or space.

Come back to the topic. I have tried to write all of them as a variable which belongs to the direction but it did not work. Can you write an expression to help me to find my mistakes?
I think I have a problem writing an expression in which a variable depends on length.

The function I described will give you Reynolds Number as a function of y position as I stated. So that function is an expression which depends on length.

It might be better if you posted what you tried to do and the error message it gave (or the incorrect result if it gave no error message).

[MNMK]

Originally Posted by ghorrocks

All the variables I used can be either constant values, or values changing with time or space.


The function I described will give you Reynolds Number as a function of y position as I stated. So that function is an expression which depends on length.

It might be better if you posted what you tried to do and the error message it gave (or the incorrect result if it gave no error message).

ave(Density)@Line 1 *ave(Velocity)@Line 1*Y)/ave(Dynamic Viscosity)@Line 1)

As it can be seen, all of my parameters are the average but I would like to calculate them in a direction.
For example how does the velocity change in a line 1 at every Y.

I attached a photo. V(y) is the velocity as a function of length not Vy.

channel.png

[ghorrocks]

OK, that partly explains what you are trying to do. But what is V(y)? Is it the V value at that xyz point? Or an average of V across the cross section at that y value? Or the peak V value across the cross section at that y value? Or something else?

And the same question applies to density and viscosity.

As a side point - the "ave" function calculates the average value of the nodal values on the object. This can be misleading as if the node spacing is not even the average value can be skewed. You probably want the lengthAve function as that does a length weighted average.

[MNMK]

Originally Posted by ghorrocks

OK, that partly explains what you are trying to do. But what is V(y)? Is it the V value at that xyz point? Or an average of V across the cross section at that y value? Or the peak V value across the cross section at that y value? Or something else?

And the same question applies to density and viscosity.

As a side point - the "ave" function calculates the average value of the nodal values on the object. This can be misleading as if the node spacing is not even the average value can be skewed. You probably want the lengthAve function as that does a length weighted average.

Dear Ghorrocks,

I appreciate your reply,

I would like to calculate all variables on a line (x,y,z) in the direction of y. I think the expression: lengthave (Variable)@line gives me the average value of a variable on the whole line, but I would like the value of the variable at every y.

wishes

[ghorrocks]

If you want the average pressure along a line you use "lengthAve(p)@line". If you want the pressure at every point along a line you use "p". As simple as that.

So to calculate the Reynolds number along a line using the local density, velocity and viscosity is just "Density*v*y/Dynamic Viscosity", assuming the V velocity is the velocity of interest and your length is just the distance in the y direction.