Clculating antanna directivity

[juppydu]

the discussion seems to me quite confusing.

My understand is thet the avaialble data are the two main cuts (0° and 90°) and each them have 360 points (that measn 1° step).

To achiece the beam directibity you MUST integrate the beam over the wholes sphere, but for sure two cuts are not sifficent for your purpose.

Bye.

 

[john54]

In most cases are two cuts enough.. I am sure that this is enough in my case because one who is professional in antenna measurement told me so..

 

[juppydu]

It depend on the pattern, In general two cut are very poor pattern sampling...
just in some very smooth pattern you can have "significant" directivity value by pattern integration.

what kind of pattern do you have ?

bye

 

[john54]

It is very smooth.. 1 main lobe, 3 sidelobes and almost nothin else..

 

[juppydu]

it is not so smooth in my opinion...
a smooth pater is e.g. patch pattern, if you have secondary lobes (array ?) I don't think you can calculate the right directivity by means of only 2 cuts...

 

[john54]

I need to have this number.. if it is not 100% correct it is not so important. I am sure it will be close enough...

 

[mauloftin]

mauloftin: do you still have this m file, or do you remember how this think worked?

I think, that no

but think, that this https://www.ece.rutgers.edu/~orfanidi/ewa/ can you help you, may be no.


but my source for formula of directivity was Balanis - Anenna Theory (the picture in attachment)

 

[mauloftin]

I need to have this number.. if it is not 100% correct it is not so important. I am sure it will be close enough...

Yes, I think so. But I think, there is problem only if is 3D radiation pattern very asymmetric, but usualy is approximately symmetrical = there is one main lobe and small sidelobe....

 

[john54]

Is this program OK?

t=(0:180)';
th=t*pi/180;
x=10.^(y/10);
N = size(th( : ), 1) - 1;
Om = 2 * pi * sum(x .* sin(th)) * pi / N;
D = 4 * pi / Om

y is my data.. in normalized units (max=0dB)..


I think that I am still doing something wrong, because I still can't get results that would match.. Program misses for about factor of 10..

 

[antonio_s]

i know the mistake. if someone else does, please email me: ps3anto@hotmail.es

best regards to all

 

[juppydu]

dear all,

to compute the directivity use the directivity definition.

In your code "seems" to me that (I don't know your code "language"):

- the pattern is rotational symmetric (only Th function, right ?)
- x should be "Th" dependent, I don't see that in your code
- you have missed in the last step to multiply by the pattern at boresight (or pointing direction of interest).
- finally you to come back to "dBi".

I hope this can help you.