Not enough received power for antenna pattern.

[waveboy]

Dear Engineers,

I am trying to get an antenna pattern for a horn antenna which is designed for 38 to 40 GHz. (its aperture is some 2cm by 4cm) The source antenna was at first a rectangular horn antenna with aperture of about 3cm by 6cm. (The antenna under test is acting as receiver). The antennas were placed about 1 metre apart. -However, this test set-up resulted in too little received power. –Therefore, the source antenna was changed to a parabolic dish of diameter 34cm (in order to provide more gain and therefore more received power). –However, the edge of the near field for this dish is 30 metres (according to the formala 2*D^2/Lambda)
Where:-
D = diameter of dish.

It is not practical for me to have the dish 30 metres away from the antenna_under_test. Also, at 30 metres, the (free space) path loss is 30dB more than at 1 metre. –Therefore, the increase in gain will be cancelled out by the increase in path loss at the necessary increased distance.

Any help greatly appreciated. I am wondering if the “D” in the above equation can really be the diameter of the dish(?). Surely the dish is just a reflector and the “antenna” is really the feed probe (a monopole) down in the feed horn of this dish?

 

[VEBill]

Random thoughts:

How much cable loss are you expecting between the test equipment and the antennas? In other words, are you using really expensive cables for your 38GHz? Even at X-band, we used 1-m long cables that costs about $1500. 38GHz is one of those numbers that should attract attention at the outset.

Have you been able to establish any link? If you can't get +20 or +30 dB signal along the boresight, then you are going to have a hard time probing the sidelobes.

 

[BrianR]

D is the largest dimension of the antenna aperture and I would think the coax loss could exceed the benefit of the gain unless you have very good coax. Is amlow gain amplifier out of the question?

 

[Higgler]

waveboy,
make your spacing between the dish and your antenna very short. The D for distance is only for your antenna if you have a good steady field at your test antenna. Usually you have both antennas at 2D^2/Lambda but that's not true in a compact range.

If you rotate your antenna about it's front face (not the back connector) and measure your pattern at two or three different distances from the dish, if your patterns are nearly identical at all 3 places, you know that you have good results.
I did similar recently, we used the distances 2D^2/Lambda of the unit under test, then 2x and 3x that distance. That puts you at 1,2 and 3 meter distance for your testing.