SatSig topic:

"Massaging" the f/d of a corrugated horn with a hacksaw?

(Read 6350 times)

Aug 4^th, 2014 at 12:58am

.
Say I have an offset VSAT antenna that works fine -except the link margin is a bit too tight.

Then I have a spare non offset dish reflector, with a bigger diameter; and it entails almost a halving of the f/d

If I remove the front first "steps" of the horn, to give it a smaller aperture, to match the decreased f/d, how close would this simple "hack" Cheesy

get me to optimum performance of the offset feed on the deeper dish?

I suppose the overall caracteristics of the crudely shortened horn would not be perfect, but how close?

Anybody with practical experience, and/or theorical facts?

Thanks!

Christian
.

Aug 4^th, 2014 at 2:54pm

Rather than "hack off" the conical section of the feed I suggest you get the correct feed for the short f/D, axially symmetric, circular dish.

The reason why you need different feed types is to obtain the correct dish edge illumination level. (tx power per sq m at the edge of the dish versus the tx power per sq m in the middle)

If you use a tiny feed on a long f/D dish, lots of transmit power will miss the dish and hit the ground. You will have an edge illumination like -5 dB. You will have an abnormally narrow main beam, very high side-lobes and low gain. You will pick up lots of noise from the warm ground.

If you use as big horn feed on a short f/D dish you will only illuminate (light up) the middle central part of the dish and most of the dish will be doing nothing. Edge illumination might be -25 dB. The main beam will be much wider and lower gain than expected, but side lobes will be exceptionally low and ground noise pick up negligible.

Edge illumination should be -10 dB for a receive-only dish with no adjacent satellites on the same frequency. Gives maximum gain but with significant sidelobes.

Edge illumination should be -16 dB for a tx/rx VSAT dish. Best compromise: gives high gain and low sidelobes. Interference to and from adjacent satellites is acceptable.

I would investigate the shape of the present offset dish. If it's surface is distorted this will reduce the gain and increase interference from adjacent satellites. Do the crossed threads test to check rim flatness, and aim for 1mm maximum gap at Ku band. Re-peak. Does that give any improvement ?

See: https://www.satsig.net/pointing/rear-side-struts.htm

If you do try to use the larger circular dish, be aware that the same rim flatness rule as above applies and that the positioning of the feed is particularly sensitive (both sideways, up/down and the in/out focus). If the feed is not in the middle some part of the first side-lobe ring will be high and interference possible. Longer f/D antennas are more forgiving on feed location and focus.

Best regards, Eric

Aug 4^th, 2014 at 10:24pm

Hi Eric,

As I said the present offset dish is working fine; and also I'm not thinking of removing the entire conical part of the horm, just a short length at the front, to make its illumination match the smaller f/D of the other dish. But my understanding is that the cone angle is not supposed to remain constant when the length of the horn is changed.

My concern is how far from optimum have the original dimensions of the corrugated steps, and the original cone angle, become, with the new short length of the cone.

Cheers

Aug 5^th, 2014 at 9:30am

With a large reduction in f/D you probably need to remove the cone completely. See the images above.
If there is a circular waveguide flange just behind the cone, try removing the cone at the flange and just use the open ended waveguide.

It is the diameter of the feed aperture that largely determines the beam-width of the feed pattern and thus the edge illumination.

The radiation from the feed spreads out towards the dish in a spherical fashion, such that it appears to have come from a point source, called the feed phase centre, which needs to coincide with the focal point of the dish, by adjustment of the distance.

In the case of a open ended waveguide feed the phase centre is at the aperture.

In the case of a conical horn the phase centre can be deep down inside, near the throat, in the case of a wide angle flare or near the aperture in the case of a narrow angle flare. If the flare angle is intermediate, the position of the phase centre will change with frequency.

Best regards, Eric.

Aug 11^th, 2014 at 11:56am

In any case, personally I'd try to "play" with other feed horns / LNB's, or make a horn with an adjustable flat scalar disc, rather than butchering a conical horn. I recently found that I got a dish with an LNB where the horn's opening agle would be badly adapted to the dish, but only slightly, and I got a much better signal (3dB more) with a simple Venton EXL "rocket feed" LNB, which uses a dielectric lens instead of a horn. maybe the different opening angle *and* the said-to-be more homogeneous polarization of the dielectric lens made the difference.