5
$\begingroup$

You buy a "20dBi omni" Wifi antenna, operating in the 2.4-2.5GHz band. You note that the expected range is not achieved from this 300mm long antenna. What could be wrong?

Can someone please tell me whether may approach is correct

λ(min) = 300/2500 = 0.12m = 120mm

Dipole length is therefore = 300mm/120mm = 2.5λ => 0.5λ (neglect '2' since its just a full revolution on smith chart??)

driven element have length of = 0.5λ/2 = 0.25λ

Actual gain of the antenna = 300mm/(0.25λ) = 10dB, therefore antenna in question is not 20dB and therefore thats why doesnt meet expected range. - is this the right way to calculate gain?

-Thanks.

$\endgroup$

2 Answers 2

1
$\begingroup$

Nice try but that's definitely not how you calculate gain!

Gain depends on many things, but first is the number of dipole elements in the stack. You could estimate the gain by guessing the number of half-wave dipoles that fit in the antenna. I suppose in your case a 300 mm antenna would fit four dipoles, to leave space for the top cap, connector and matching at the bottom.

Then the maximum theoretical gain will be:

2 dBi for the first dipole, + 3 dB for each further doubling, so 8 dBi.

In practice, each additional antenna also brings some further losses. So it is probably closer to: 1 dBi for one dipole, 3 dBi for two, 5 dBi for 4, 7 dBi for 8.

I've never heard of 20 dBi in an omni before. That would require a minimum length of 32 wavelengths, 3 m, and more in practice. It would have an elevation beamwidth of around 1 degree. Not much use having such a flat beam anyway. I think you have a 5 or 6 dBi omni, which should give you about double the range of the little dipole that comes with the router.

$\endgroup$
1
$\begingroup$

Given the size (30cm) and assuming it is omnidirectional, it is probably a collinear array.

For this type of antenna, claiming 20dBi gain is quite exaggerated. You would expect something around 6 to 8 dBi.

An example of 2.4GHz collinears:

A 20dBi gain is expected from a directional antenna ("cantenna", helical, patch array), not omni.

$\endgroup$
1
  • $\begingroup$ Thanks, I suspected that the given gain was too high, is the way I calculated the gain correct (given that that's the only information available?) $\endgroup$
    – HappyFeet
    Jun 10, 2015 at 6:31

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .