Homework Help: Streamlines flows fluid

1. Dec 4, 2013

hvthvt

1. The problem statement, all variables and given/known data

The two-dimensional steady flow of a fluid with density ρ is given by

v=K(-yex + xey) / (x2 + y2) where K is a constant
(a) Can this flow corespond to the flow of anincompressible fluid?
(b) Determine and sketch the streamlines of this flow.
(c) Determine and sketch the aceleration of this field
(d) Determine the pressure difference between two points that are a distance r1 and r2 away from the origin. (r2 > r1) ?

2. Relevant equations

dx/u=dy/w

3. The attempt at a solution

I see that there is x^2+y^2, this makes me think that there is a circle based streamline flow?? I do not understand how I can visualize this in my head, such an equation. How can I know whether its incompresssible? The acceleration should be the derivative of those streamlines I guess.. Can anybody help me out with these questions PLEASE?

File size:
29.2 KB
Views:
119
2. Dec 4, 2013

voko

There is a condition involving the divergence of velocity...

3. Dec 4, 2013

Staff: Mentor

You're correct. The streamlines are circles. Under those circumstances, what is the acceleration of a particle traveling around the circumference of a circle?

4. Dec 5, 2013

hvthvt

The acceleration is a=v^2/r ?

5. Dec 5, 2013

hvthvt

Incompressibilty implies the divergence of velocity is zero...

6. Dec 5, 2013

voko

Good. Can you check that?

7. Dec 5, 2013

hvthvt

(-ex + x)/ (x^2 + y^2) + (-ex + x)/(x^2 + y^2) does NOT equal zero
So its not an incompressible fluid? What to do with the derivative to x and to y of (x^2 + y^2)

8. Dec 5, 2013

voko

I am not sure what you did, but that does not look like you computed divergence. Do you understand what it is?

9. Dec 5, 2013

hvthvt

Not really.. I am pretty stuck at this moment to be honest. Can you help me out?

10. Dec 5, 2013

hvthvt

I need to check whether ∇.v = 0
So whether d/dx(formula) + d/dy(formula) = 0 right? But I'm having difficulty with deriving the x^2+y^2 part

11. Dec 5, 2013

voko

No, it is not d/dx(formula) + d/dy(formula).

Divergence applies to a vector, which can be written as two compoents $$f(x, y) \hat \imath + g(x, y) \hat \jmath ,$$ where $f(x, y)$ is the x-component of the vector, and $g(x, y)$ is the y-component of the vector. The divergence then is $$\frac {\partial f} {\partial x} + \frac {\partial g} {\partial y}.$$

This is what you need to compute and check whether it is zero.

12. Dec 5, 2013

hvthvt

Oooh really, thanks!! I understand that it can be written as two components: but what is the x component, f(x,y) in this formula?
Is that
-yex / (x2 + y2) ??

13. Dec 5, 2013

Staff: Mentor

Yes. Do you know how to get v^2 when v is a vector expressed in component form?

14. Dec 5, 2013

voko

Yes it is. But without the $e_x$ part, it is there to denote that it is the x component.

15. Dec 5, 2013

hvthvt

I found:
∂f/∂x= (2xye - y^3 - x^2y)/(x^2+y^2)^2
∂g/∂x=(-2xye+x^3 + xy^2)/(x^2 + y^2)^2

Computing ∂f∂x+∂g∂y = (x^3 - y^3 + xy^2 - x^2y)/(x^2+y^2)^2

This is not zero. So the fluid is imcompressible. Is this correct?

16. Dec 5, 2013

hvthvt

Ooooh, without the e_x part, I get it! Then it gets much easier!!
df/dx = 2xy/(x^2 + y^2)^2
dg/dx=-2xy(x^2+y^2)^2
df/dx-dg/dx = 0 so the flow corresponds to the flow of an incompressible fluid! right?

17. Dec 5, 2013

voko

Your final result is correct, but you there is a typo: it is not df/dx - dg/dy, but df/dx + dg/dy. If it were df/dx - dg/dy, you would not have obtained zero.

So, the flow is incompressible.

18. Dec 5, 2013

hvthvt

Yes, sorry, that was my mistake. I typed it incorrectly, but I understand it right now. Thank you.
So now I am asked in (b) to determine and sketch the streamlines of the flow.
How should I determine them?
First of all, the flow is incompressible, so what kind of information does this supply?? Does that influence the way in which the streamlines flow?

How to determine the streamlines?
dx/vx = dy/vy --> dx/dy=vx/vy ??

19. Dec 5, 2013

voko

I think you have already determined - or, at least, guessed - that the streamlines are circular.

So what you can do is this. Choose some radius. Select a few points on a circle of that radius. Compute the velocity vector at those points. Draw the resultant vector as arrows on paper. Repeat.

20. Dec 5, 2013

hvthvt

I figured out that 1/2(x^2+y^2) = constant. Does this have any significance for answering (b)?