Focal Point of Dish

[q12345]

This message is for wejones.

I found a link of yours in another post. You used small mirrors taped to your FC90 to find the focal point. I have a 90cm dish fitted with a multi-lnbf bracket. I tried your experiment and found that the point at which the reflections from the mirrors converged was an inch closer to the dish than the lnbf holder was set to.

So I adjusted the bracket to put the face of the feedhorn on the focal point to see if I could get a better signal (i.e. better performance from my dish). It turns out that instead of an inch, half an inch gave the best result. I used G25 for my test.

I wonder if the focal point has to be at the face of the lnbf or at the "cross-hairs" near the back of it. I can post some shots if there is interest.

[wejones]

Originally Posted by q12345

This message is for wejones.

I found a link of yours in another post. You used small mirrors taped to your FC90 to find the focal point. I have a 90cm dish fitted with a multi-lnbf bracket. I tried your experiment and found that the point at which the reflections from the mirrors converged was an inch closer to the dish than the lnbf holder was set to.

So I adjusted the bracket to put the face of the feedhorn on the focal point to see if I could get a better signal (i.e. better performance from my dish). It turns out that instead of an inch, half an inch gave the best result. I used G25 for my test.

I wonder if the focal point has to be at the face of the lnbf or at the "cross-hairs" near the back of it. I can post some shots if there is interest.

Nice to see that there is someone else crazy enough to put mirrors on their dish. It was really educational for me.

Re where on the lnbf the focus should be... I'm not sure about these little lnbfs, but on the big C/Ku feedhorns, I think that they say the focal point should be something like 1/4" inside the front of the tubular throat, not at the back where the probes are.
When I did my mirror experiments, I had something like 5 different mirrors, and except for the one main focal points, when you got off center, the reflections from the different mirrors were zipping around in different directions as you moved the dish around, and moved the feed (in my case I used a fake feed) closer and further from the dish. It seemed like there were multiple places where a couple of the mirrors would hit the same place, but no other place where they all hit. I got the impression that if I had 10 mirrors, I'd see even more variability. Ie, if you only had 2 or 3 mirrirs, you well may have found a spot 1" in that looked good for a couple of the mirrors, but was worse for other points on the dish. Basically there is only ONE focal point, so when you go off center you're just trying to find a spot that works OK, but not perfectly. In my case with 5 mirrors, I found that my LNBF was off position by about 2 inches, but the actual reception at the place where the lnbf was, vs the real focal point, wasn't enough to see a difference in the quality levels on the Fortec receiver I had connected at the time. Ie even though the 2" off focus seemed a lot worse than the real focal point with 5 mirrors, apparently the reflections from all the places that I didn't have mirrors made the reception not all that worse when measured with a receiver. Anyway, I didn't even bother bending my feed to the proper focal point.

[q12345]

Well, I thought that the mirror thing was ingenious and knew right away you were an innovative guy (person of choice if ever stranded on a deserted island)!

I put up nine little mirrors, in a cross configuration like yours. At first, I had the outer mirrors right at the edge of the dish, but later decided to make the cross a little smaller because I kept blocking the sun on a few of them while trying to tilt the dish and hold a piece of stiff cardboard near the lnbf.

I couldn't get them all to come to a point, but the closest they came was about an area the size of the face of the lnbf. I got these mirrors from a small disco ball I bought from the local dollar store (I just love what you can find in these bargain stores ). They weren't the completely flat ones that you used. I think they had a glass thickness of about 2mm. Might have thrown off the focal point a bit.

On another note, I tried something that was interesting. I had my 75cm dish pointing to AMC4 (101W), TP freq 12120V. I covered up parts of the dish with my two hands together simulating an obstruction. I found that the top of the dish was the most sensitive area to obstruction. The right and left sides were not particularly affected by my two hands.

On still another note, I recall from my astronomy course way back in university, that you could combine the signal from two or more dishes in an array which would give a resolution equal to a dish that was the diameter of the distance from the furthest dishes of the array. For example, if you had two 100cm dishes side by side touching each other (and obviously pointing in the same direction) the resolution of the signal would be equal to having a 200cm dish. If there was a receiver that could combine signals from two or more dishes of an array then you could conceivably receive C-band transmission from two smaller dishes. Wouldn't that be cool!

[wejones]

Originally Posted by q12345

I put up nine little mirrors, in a cross configuration like yours.

I wish I had used that many mirrors, and spread them all over the dish.

Originally Posted by q12345

I couldn't get them all to come to a point, but the closest they came was about an area the size of the face of the lnbf. I got these mirrors from a small disco ball I bought from the local dollar store (I just love what you can find in these bargain stores ). They weren't the completely flat ones that you used. I think they had a glass thickness of about 2mm. Might have thrown off the focal point a bit.

I used two different thicknesses of mirrors, because I was concerned about throwing off the shape of the parabola, but in both cases, it agreed with what you said, ie the result was a focal area about the size of the throat of the lnbf, so perhaps that is the limit of the shape of these little dishes.

Originally Posted by q12345

On another note, I tried something that was interesting. I had my 75cm dish pointing to AMC4 (101W), TP freq 12120V. I covered up parts of the dish with my two hands together simulating an obstruction. I found that the top of the dish was the most sensitive area to obstruction. The right and left sides were not particularly affected by my two hands.

That's interesting. I read an article on the internet several months ago (I read it months ago, I'm not sure how long it has been there) that did a semi scientific study of just this. Ie how important different parts of the dish are. Unfortunately, I don't think I saved the URL, but I do seem to remember it coming to a similar conclusion, at least for the small offset dishes (which was what was being studied). I don't think it would be the case for prime focus dishes, and it may be greatly affected by the aim and shape of the lnbf itself, compared to the shape of the dish. These little offset dishes are theoretically just a portion of a prime focus dish, however they differ with respect to where the feedhorn is aimed. Ie a prime focus dish has the feed aimed at the center of the parabolic surface, but this would be just off the bottom edge of these offset dishes, which wouldn't be very effective, so they aim the things higher toward where there is actually dish surface.
A couple years ago, I didn't realize that a tree had blocked the bottom half of my big dish on my true south satellite, and I went up on a front end loader, and was moving the aim of the lnbf around while listening to the beep of a signal meter. I found that the maximum signal was when the lnbf was aimed at the top half of the dish (obviously because that's the half that the signal was coming from, but also because big 10' dishes don't really see the outside of the dish very well on Ku, ie they are generally using the interior of the dish for Ku, except in this case that was blocked by the tree), and being confused, I got down and looked at where the dish was aimed, and eventually discovered that the tree was blocking the bottom half of the dish. But basically, what I'm getting to, is that I think the part of the dish that is most sensitive to being blocked is the part of the dish that the lnbf is aimed at, but that may not be true for all dishe/lnbf combinations.

Originally Posted by q12345

On still another note, I recall from my astronomy course way back in university, that you could combine the signal from two or more dishes in an array which would give a resolution equal to a dish that was the diameter of the distance from the furthest dishes of the array. For example, if you had two 100cm dishes side by side touching each other (and obviously pointing in the same direction) the resolution of the signal would be equal to having a 200cm dish. If there was a receiver that could combine signals from two or more dishes of an array then you could conceivably receive C-band transmission from two smaller dishes. Wouldn't that be cool!

This has been brought up very often over the years in different forums (even a couple days ago in this forum). Getting this to even give the same performance as a single dish would take some careful phasing of the signals, and wouldn't be easy.

However assuming that you've done this, I am very confused about how this would affect resolution. I found equations for resolution or beamwidth from several different fields, like ham radio, TVRO, and radio astronomy, and the different equations come up with slightly different results. In my "tvcalc" program that I have on my web page, I put all three calculations there, because I didn't fully understand the implication of the differences. However all three equations pretty much suggest that the resolution should be related to the gain of the dish, and at least intuitively, the gain should be related to the total surface area, not just the widest separation obtained by the dish or dishes. However this doesn't seem to jive with obvious comparison to optical telescopes and cameras, where resolution seems to be more related to focal length and F/D (ie a long focal length small {high F/D} lens seem to give better resolution than a short focal length big {low F/D} lens). But this is why I think high F/D sat dishes work better than low F/D dishes, even though the resolution equations don't include anything related to F/D.
Anyway, I'd be interested in any references (hopefully internet URLs) you might have, related to the theory behind those big radio telescopes made up of multiple small reflectors, because the astronomy book I got my equation from only discussed resolution of a single reflector.
Anyway, I'm far from understanding what is going on here, because the various things I've read don't seem completely intuitive. From the equations I've seen, the separation of 2 or more small dishes shouldn't be as important as the total area (gain), but intuitively it seems like it should. I guess my question would be, which direction would it be best to separate two small dishes, ie along the plane of the Clarke belt or along a line perpendicular to this? This also relates to whether a short wide eliptical dish would give better resolution than a tall skinny dish. Intuitively, I'd think it would, but from the equations I've seen, there wouldn't be any difference. So I'm confused about several things here.

[q12345]

Now you're getting into some of the more technical aspects of multi-dish setups and I don't recall all the details from '85 when I took that course. The coolest thing I remember is that there was an international collaboration of scientists that would share data from their respective radio telescopes such that the effective dish size was the Earth's diameter. Data taken from countries latitudinally (is that a word?) from one end of the Earth to the other would be compiled and crunched to give a picture of a given section of sky with tremendous resolution. This would seem to support your observation of the elliptical dish - wide is better as opposed to tall and skinny. But maybe this is true because all geostationary satellites are along the equator.

If you look up VLA (very large array), San Agustin, New Mexico, you'll see the best example of an array of dishes working as one. I think this was even shown in the movie "Contact" with Jodie Foster, near the end of it.

By the way, have you seen photos of the Arecibo dish? Awesome eh!!!