Water velocity Calculation

[macvalsa]

Gentleman and Others,

A bit of assistance if you dont mind.

I have ordered Crane 410 but expect it will be several days befor i receive it. In the interim i would like to calculate some water velocities caused by installing a valve.

Each valve type and manufacturer quantify the flow through the valve by a Cv and obviusly different valves will cause different increases in velocity based on their effeiencyso using the simple orifice calculation will not work

Most valves are also given a non-dimensional K figure (resistance co-efficent)

Could somone please advise what is the formula for calculating the velocity at a given flow. I have seen

h = KV^2/2g but this does not take into account flow

Looking forward to hearing from you

Regards

 

[rbcoulter]

velocity = volumetric flow rate / cross-sectional area

 

[DonyWane]

Just a few tips:

You can relate a Cv to a K factor through the following equation:

Cv=29.9*d^2*(K)^(-1/2) (where d is diameter in inches)

Using this Equation, you can convert Cv's to K's when K factors are not given ... which is very helpful since the most common form of the head loss equation takes the form of:

h = K(v^2/(2gc))

The velocity here is the velocity in the pipe that the valve is installed on (not the velocity through the valve). I think that it is technically the pipe leading up to the valve (if there is a change in line size). However, it would be acceptable to use whichever is conservative for your purposes (i.e., using the velocity in the smaller line size if you are worried about not getting enough flow through your line).

You can simply take the length of your pipe (L) and its diameter in feet (D) and write a full expression for your piping as

h = (fL/D + K)*v^2/(2gc)

f here is the moody friction factor. The fanning friction factor can be used be re-writing as:

h = (4fL/D + K)*v^2/(2gc)

If you know velocity and you know pipe diameter, you know flow...so I think that you should be set.

 

[macvalsa]

Thanks DonyWane,

This has allowed me to calculate the "K" factors successfully.

in the formula h = kV^2/2g h is the frictional head loss due to the resistance (k) so we can calculate what the head loss created by the restriction will be (in our case a valve)

In our case we are trying to calculate what the flow through the valve will be at a velocity of 20fps the h we need is the differntial which will allow us to calculate from q = cv*h^(1/2).

Any further suggestions ??

 

[BRIS]

As noted by Doneywane the velocity is the velocity in the pipe upstream of the valve (the pipe diameter is the same as the valve nominal diameter). Thus if the velocity (v) is 20 fps then the flow is V x A where A is the area of the pipe. It is as simple as that. But, I don't think that is the question you are asking.

I think what you now want to know is what the head loss (H)will be across the valve at this flow rate (v x A).

H can be calculated from either formula i.e from K if you know K or from CV if you know cv for the valve you have. Alternatively if you are trying to select a valve taht will give you a certain flow rate at a certain head loss then calculate K or cv from the velocity and head loss and select a valve that will give these values when fully open. (The maximum K for a butterfly valve will be about 0.5 it will be more for other valves). Thus if for example you are proposing to use a butterfly valve and K is less than 1 you need a bigger valve and you will need to recalculate the velocity and repaet the iteration.

You need to provide more information on what you are trying to do because you seem to already have the answer to the question??

Brian

 

[rbcoulter]

It appears that there are some semantic difficulties in relation to this question. On the surface it appears that the questioner wants to know the flow rate versus velocity relationship for a particular valve. Also, the questioner does not appear to be interested in using the cross-sectional area of the valve but would rather use the Cv coefficient provided by the valve manufacturer.
The Crane handbook does indeed provide an equation for relating K to Cv. By manipulating equations 3-16 & 3-14 one can obtain an equation that relates Q versus V for a particular valve (dropping out thr K, Cv, & head terms). Why is this better than using the actual cross-sectional area of the valve? These Crane equations are emperical afterall.

 

[25362]

The Pipe Friction Manual of the Hydraulic Institute gives the resistance coefficients for valves as a function of diameter and type of valves in graphical form.

An attached table states that ranges of variation are greater than +/- 25%. For flanged check valves this range may be from -80% to +200%. At lower than 15 fps check valves would be only partially open and exhibit k values larger than those shown in the charts.