Home Forum Technical Discussions     

Re: k factor

Posted by: James Barshinger Profile (PID_688), E-mail: Address, on December 15, 2008 at 23:02 :

In Reply to: k factor posted by : Ed Ginzel Profile   Exhibitors stand , E-mail: Address, on December 15, 2008 at 15:59 :

Ed,

To derive the values for "k" in the beam divergence equation for circular oscillators, 1st start with the equation:

R(g)=2*J1(x)/x where x=pi*D/l*sin(g)

- These equations are in krautramer though I have used g instead of "gamma" and l instead of "lambda"

To find a "k" for a certain dB drop, you start with the J1(x)/x term, equating that to the desired amplitude reduction. For example: J1(x)/x=.5 for a 6dB drop of the "free field", or for the "echo field", (J1(x)/x)^2=.5. You need to then use a numerical solver to obtain the x that satisfies the equation for the particular dB drop you are interested in.

Now, using the second equation, x=pi*D/l*sin(g). You can rewrite this equation as: sin(g)=(x/pi)*(l/D). "k" is simply the quantity (x/pi), thus you get the equation in the Krautkramer book.

For rectangular oscillators, the process is the same with the exception of using Sin(x)/x instead of J1(x)/x

-Jim

----------- Start Original Message -----------
: I have been asked how the values for "k" used in the beam divergence equations have been derived. My favourite references of Krautkramer and Ermolov touch on the subject and allude to the use of the Bessel J1(x) function. However, I cannot find the link to the k values for dB drop using this function described in either text.
: Does anyone have a more complete formulation of the equations to derive k?
------------ End Original Message ------------




Follow Ups:



Post a Followup:

Name:
E-Mail:

Subject:

Comments:

Optional Link URL:
Example: "http://www....html". Alternative write "http://www.ndt.net/wshop/attach/filename.htm" if you transfered a filename.htm via FTP
Page Title: is important!
Optional Image URL:
http://www....gif


Attention! Please click the button only once. This may take a moment to process


Forum Technical Discussions Help