homebrew parabolic dishes - how to?
qwa...@hotmail.com
I would like to experiment with space communications, moonbounce, amateur
seti, and radio astronomy. a dish seems perfect for multipurpose use, but I
need a big one (or two!). how do I go about making a dish? the only info I
could find was on spin-casting, and it was only in reference to small dishes.
apparently spin casting is very precise and good for beyond 10 GHZ, which is
about as far as I will go, but will it work for a 10 meter+ dish? is it
feasible? I wish some of those eme'rs who made they're own dishes would
explain on their websites.
-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/ Now offering spam-free web-based newsreading
JBotts9786
Dear OM. You didn't mention an exact frequency you plan to operate or listen on
so figure it somewhere from 432 mhz to 10 Ghz. Tis my opinion that your best
deal, dollars and cents-wise,
is to purchase a used satellite TV dish of the meshed variety. You would be
surprised just how cheap those things are now-a-days with the advent of the
"small", 18-24" dishes for 11Ghz. I have picked up several 8 foot mesh jobs for
less than $25 each and a 14 foot mesh job was given to me just to take it out
of the owners yard. I have turned down offers of free 10 foot spun aluminum
dishes simply because I could not safely take them down or haul them to the
junque yard. They're out there-you just need to look around and ask. Believe
me, unless you are really into some "heavy duty" EME, etc. work, a 14 -18 foot
mesh dish or a 6-10 foot solid aluminum/fiberglass dish is about all you will
be able to handle/turn/point and elevate--they are a "hand full". The Gaasfet
pre-amp reduced the size of the required dish for all "space" as well as
terrestiral communications considerably. Good Luck and 73, Bunky, K4EJQ
K4...@juno.com
e.schu...@worldnet.att.net
For Sale Andrews 10 foot diameter spun Aluminum dish. In good shape with
no dents or corrosion. $60 located in Los Angeles, CA
Eric WB6KCN
qwa...@hotmail.com
yes, you are correct, I will be using 432Mhz-10Ghz+. I have access to several
TV dishes, but I am more concerned with the feasiblility of *much* bigger
dishes. Serious EME would accurately describe what I want to find out more
about, but it could be an understatement depending on whether I decide to
shell out the bucks for something incredible. anyway, I just want to know how
the big radio telescopes are constructed with precision so I can decide
whether I want to try something.
In article <199804222015...@ladder01.news.aol.com>,
Gary Coffman
On Wed, 22 Apr 1998 14:17:49 -0600, qwa...@hotmail.com wrote:
>I would like to experiment with space communications, moonbounce, amateur
>seti, and radio astronomy. a dish seems perfect for multipurpose use, but I
>need a big one (or two!). how do I go about making a dish? the only info I
>could find was on spin-casting, and it was only in reference to small dishes.
>apparently spin casting is very precise and good for beyond 10 GHZ, which is
>about as far as I will go, but will it work for a 10 meter+ dish? is it
>feasible? I wish some of those eme'rs who made they're own dishes would
>explain on their websites.
For dishes up through the 3 meter size, the best solution is a used TVRO
dish. These have become a drag on the market since the advent of DSS.
You can often get one simply by offering to take it down and haul it away.
Many of them have surface tolerances suitable for use through K band.
For a bigger dish, there are two approaches. The simplest home approach
is the stressed dish. The ribs (typically EMT) are attached to a central hub.
A long shaft passes through the hub and non-conductive guys extend from
the outer rib ends to a point up this shaft. The guys are tensioned to warp the
ribs into roughly parabolic shape. A little creative persuasion with a mallet
and a plywood template finishes achieving the parabola. The ribs are
then covered with hardware cloth to form the dish surface. The ARRL
books have had a plan for a 24 foot stressed dish like this for many years.
The more complex method is to build up self supporting rib trusses using
a template and then assembling them to form the dish. Again, cover with
wire mesh or solid aluminum sheathing to form the dish surface. This is
the way very large steerable dishes are normally made. This method is
expensive, and the dish produced is heavy. But it can be made very large,
and very good parabolic tolerances can be maintained.
A third method can suffice for EME. Because the Moon will be at a known
point in your sky on a regular basis, you can build a fixed cylindrical reflector
on the ground using poles for support and template bent EMT for ribs. Cover
with wire mesh as usual. A cylindrical reflector has a line focus, so you need
a colinear array along the line of focus for the feed. If you angle this dish
properly, you can benefit from ground reflection gain to the newly risen or
soon to set Moon. That's the cheapest and dirtiest way to do EME on upper
UHF. (On 2m and 70cm, yagi arrays are generally more practical than using
a dish.) The drawback is that you can only operate periodically when the
Moon wanders into the beam of the dish. (If you're clever, you'll site it to
be pointing at the Moon during the next EME contest.)
Gary
Gary Coffman KE4ZV | You make it |mail to ke...@bellsouth.net
534 Shannon Way | We break it |
Lawrenceville, GA | Guaranteed |
qwa...@hotmail.com
as usual gary, you have suplied me with useful info. thanks. I want to know
more about the trussed method you mentioned. Where can I get a template? is
it possible to make or borrow one? who makes the professional astronomers's
telescopes? or are they custom built by the groups that operate them?
about that stressed method, will it work for a 30 meter dish at 10 Ghz, or is
it too inaccurate? how are multiple small dishes connected together for an
array?
In article <353f5fd3....@news.atl.bellsouth.net>#1/1,
Ian White, G3SEK
wrote:
>as usual gary, you have suplied me with useful info. thanks. I want to know
>more about the trussed method you mentioned. Where can I get a template? is
>it possible to make or borrow one? who makes the professional astronomers's
>telescopes? or are they custom built by the groups that operate them?
>
the Central States VHF Convention series (ARRL). Sorry, I don't have a
bibliography here but there is a lot of sound practical experience out
there. You can also check W6/PA0ZN's EME and Radio Astronomy site at
http://www.nitehawk.com/rasmit/
>about that stressed method, will it work for a 30 meter dish at 10 Ghz, or is
>it too inaccurate? how are multiple small dishes connected together for an
>array?
No, the experience of EMEers is that even a 20ft stressed dish is
hardly good enough at 1296. That's mainly because stressed dishes tend
to have large errors in the parabolic rib profile. Paraboloids can
tolerate quite large errors in the surface profile if those errors are
truly random (eg odd dents and bulges in the surface between the ribs)
but systematic errors due to the ribs not being quite the right shape
can cause significant loss in gain, and unwanted sidelobes.
Stressed dishes also have mechanical stability problems - they don't
have much strength against wind or snow loading.
That has been the experience of the EME community on the bands up to
2304. We keep in pretty close touch but right now I know of nobody in
the world who's using a stressed dish... though many have tried.
Sadly, for big microwave dishes that must maintain an accurate profile
at all times, there's no substitute for trussed ribs and a lot of
tedious mechanical work.
73 from Ian G3SEK Editor, 'The VHF/UHF DX Book'
'In Practice' columnist for RadCom (RSGB)
http://www.ifwtech.demon.co.uk/g3sek
Gary Coffman
On Thu, 23 Apr 1998 13:58:48 -0600, qwa...@hotmail.com wrote:
>as usual gary, you have suplied me with useful info. thanks. I want to know
>more about the trussed method you mentioned. Where can I get a template? is
>it possible to make or borrow one? who makes the professional astronomers's
>telescopes? or are they custom built by the groups that operate them?
>
>it too inaccurate? how are multiple small dishes connected together for an
>array?
The stressed dish approach probably can't maintain good enough surface
tolerance for 10 GHz operation in a very large steerable design (though it
can work for a fixed dish). A few ripples and bumps don't matter much, but
the overall shape of the curve in the outer third of the dish (where most of the
dish area is) has to be pretty accurate. You can improve this with a mallet and
template as I noted, but the accuracy needed at 10 GHz is probably too much
to ask of that relatively crude shaping method. The 24 footer in the Handbook
is Ok for L band, and if built carefully, Ok for S band. But it wouldn't do for C
band or X band. The gravity loads on a 30m design would make it almost
impossible to maintain in true as you change pointing angle.
You can easily make a template using the standard formula for a parabola.
Just lay the curve out on plywood (only one half of the curve is needed).
You can then cut out the template and use it to check the dish surface.
Or you can screw forming blocks to the template and use them to aid in
bending your ribs to fit. (This can work for up to a double sheet, IE a dish
about 32 feet in diameter, but larger dish templates need to be made of
steel, and are called jigs.)
If you want a 30m dish capable of operation at 10 GHz, you're looking at truss
construction. A custom truss jig will be created by the truss fabricator and
your trusses will be constructed using that fixture. This is a *big* steel fixture
which will cost several thousand dollars to fabricate for making the ribs of a
30m dish. You'll want to double check that jig for accuracy before allowing
construction of the rib trusses to proceed. The rib trusses will be joined by a
number of concentric ring trusses. Jigs for them will have to be fabricated
and checked too. (These are somewhat easier to do, you only need one segment
of the ring, but you need a separate jig for each concentric ring, so the jig cost
will still run on up there.)
Now a 30m truss dish is *huge*, and weighs many tons. As a benchmark,
the 34m dishes at Goldstone weigh in at about 100 tons. Yours won't be
much lighter. A support back design is required to keep your trusses from
deforming under gravity loads (and wind loads) at various pointing angles.
(This is relatively standard bridge engineering.)
AFAIK dishes in this size range are all custom designs. Any engineering
company experienced in bridge construction should be able to build the
trusses, support back, and mount pedestal from a set of detailed design
plans, but the level of precision required is higher than what they normally
do, so some care should be used in selecting a contractor. Bechtel has the
experience to tackle this sort of project. The hydraulics for positioning may
require specialized engineering support. Bechtel can supply that too. You'd
better be independently wealthy, or have a government grant, if you plan to
erect a dish in this size range.
A more reasonable size dish, up to 10 meters or a tad larger, can be purchased
off the shelf from a number of companies. The one down the street from me is
called Scientific Atlanta. They can supply a turnkey installation if you wish, or you
can employ your own local contractor for site prep and dish assembly. I've installed
a number of their 5m dishes, and one 10m job. They aren't that hard to assemble.
The bigger ones do require a bit more in the way of cranes and steel handling
expertise. (These dishes aren't cheap by amateur standards. The 10m dishes
cost about the same as a new luxury car. The 5m jobs are a bit more reasonable.)
Multiple small dishes can be laid out as an interferometric array. This will
greatly improve resolution (proportional to the effective baseline diameter),
but won't approach the signal gathering ability of a single large dish. You'd
need the same surface area as the single large dish to capture the same
amount of signal, and the losses from the combining and phasing networks
would require total dish area to actually be several times larger than that
of the single large dish. It'd be a nightmare to tune up and point too.
A *serious* 10 GHz EME antenna wouldn't need to be bigger than 5m. A 10m
dish would make a world class 10 GHz EME antenna. I think you're setting
your sights quite a bit higher than your wallet can afford by contemplating a
30m dish. AFAIK there is only one 30m dish in the world owned by an amateur,
and I'm not sure it has ever been fully re-installed (it's an ex-Comsat dish
bought at scrap metal prices, and even that is a tidy sum).
Cathryn Mataga
Gawd, it seems like a shame that you'd need all that steel just to hold up a
reflector. I sort of imagine, and this is totally stupid Usenet speculation,
some kind of inflatable spherical baloon with reflective material stuck on half of
it -- and with the feed inside the baloon? The outside hemisphere of the baloon
would need to be made of material that was invisible to microwaves.
The idea would be to use air pressure to keep the shape of the reflector? Or maybe,
to keep the feed on the outside of the baloon, some kind of inflatable donut
type of shape?
Gary Coffman
On Fri, 24 Apr 1998 23:17:09 -0700, Cathryn Mataga <ke...@junglevision.com> wrote:
>Gary Coffman wrote:
>> Now a 30m truss dish is *huge*, and weighs many tons. As a benchmark,
>> the 34m dishes at Goldstone weigh in at about 100 tons. Yours won't be
>> much lighter. A support back design is required to keep your trusses from
>> deforming under gravity loads (and wind loads) at various pointing angles.
>> (This is relatively standard bridge engineering.)
>
>Gawd, it seems like a shame that you'd need all that steel just to hold up a
>reflector. I sort of imagine, and this is totally stupid Usenet speculation,
>some kind of inflatable spherical baloon with reflective material stuck on half of
>it -- and with the feed inside the baloon? The outside hemisphere of the baloon
>would need to be made of material that was invisible to microwaves.
>The idea would be to use air pressure to keep the shape of the reflector? Or maybe,
>to keep the feed on the outside of the baloon, some kind of inflatable donut
>type of shape?
Well, a balloon would form a spherical reflector instead of a parabolic
reflector. That doesn't totally kill the idea, but does require a different
sort of feed. What kills the idea is anchoring and pointing that 30 meter
balloon in the wind and under gravity loads. Even a relatively light wind
would generate a lot of force against that much sail area. You also need
to allow for the effect of gravity on the balloon. It will tend to sag under
its own weight. And it needs to be steerable to a pointing accuracy within
a fractional degree, which isn't easy for a more conventional dish, and
would be more difficult for an anchored balloon.
You'd need to inflate the balloon to pretty high pressure to avoid too
much sag and deflection by gravity and wind (a deflection of 1 inch
over the surface would defocus it at 10 GHz, that's a tolerance of less
than 0.01%). And that in turn requires a strong balloon, which means
a very heavy balloon. We're talking about a balloon made of material
heavier than that of a truck tire. I'd estimate it would weigh about
395,000 pounds. A steel truss dish would actually be lighter.
This balloon antenna would be bigger than the Goodyear blimp, and
the blimp ripples and distorts in the wind more than would be acceptable
for a dish. Large airships had to go to rigid construction to maintain
structural integrity in the wind. Even so, the Los Angeles broke up in
a thunderstorm. Very large light structures aren't very durable in
atmosphere. In space, a large inflatable antenna makes a lot more
sense. The pressure can be low and the structure can be very thin
and low mass. The material of Echo I was such a thin mylar that
a piece of it would dance on an air current like smoke. But on Earth,
we need to build more robustly.
The requirement that the dish be steerable is the real killer. If we
can make do with a fixed dish, we can make very large dishes that
aren't so massively built. An example is the 1000 foot Arecibo dish.
It was a massive engineering project, but it was much less massive
than if it were required to be steerable. I'm dubious that our civil
engineers are up to the task of making a steerable dish that size.
I expect that they'd start to run into strength of materials limits
trying to maintain surface accuracy while swinging it around at
all angles.
Ian White, G3SEK
Gary Coffman wrote:
>
>The requirement that the dish be steerable is the real killer. If we
>can make do with a fixed dish, we can make very large dishes that
>aren't so massively built. An example is the 1000 foot Arecibo dish.
>It was a massive engineering project, but it was much less massive
>than if it were required to be steerable. I'm dubious that our civil
>engineers are up to the task of making a steerable dish that size.
>I expect that they'd start to run into strength of materials limits
>trying to maintain surface accuracy while swinging it around at
>all angles.
>
Tha Arecibo dish is spherical, which allows it to be semi-steerable over
a few degrees. The penalty is a very fancy line-source feed system which
compensates for the fact that a sphere doesn't have a single focal point
like a paraboloid. (I'm still struggling to see how a steerable feed
could be implemented inside a balloon, but it's fun to speculate...)
A 100ft amateur version of the Arecibo dish was built by W1FZJ/KP4 who
was working on the full-size project. The feed was mounted on a 50ft
pole and steered by changing the guy lengths; obviously Sam Harris had
unique access to the feed design information too. Unfortunately he died
before it could be fully commissioned, and the dish ended its days as a
chicken coop.
The 1000ft dish has a 400-450MHz wavegude feed, and has been used a few
times for amateur moonbounce. 5W puts out a storming signal - so strong
that I thought it was local QRM and almost ignored it!
Cathryn Mataga
Gary Coffman wrote:
> On Fri, 24 Apr 1998 23:17:09 -0700, Cathryn Mataga <ke...@junglevision.com> wrote:
>
> >Gary Coffman wrote:
> >> the 34m dishes at Goldstone weigh in at about 100 tons. Yours won't be
> >> much lighter. A support back design is required to keep your trusses from
> >> deforming under gravity loads (and wind loads) at various pointing angles.
> >> (This is relatively standard bridge engineering.)
> >
> >reflector. I sort of imagine, and this is totally stupid Usenet speculation,
> >some kind of inflatable spherical baloon with reflective material stuck on half of
> >it -- and with the feed inside the baloon? The outside hemisphere of the baloon
> >would need to be made of material that was invisible to microwaves.
> >The idea would be to use air pressure to keep the shape of the reflector? Or maybe,
> >to keep the feed on the outside of the baloon, some kind of inflatable donut
> >type of shape?
>
> Well, a balloon would form a spherical reflector instead of a parabolic
> reflector. That doesn't totally kill the idea, but does require a different
> sort of feed. What kills the idea is anchoring and pointing that 30 meter
> balloon in the wind and under gravity loads. Even a relatively light wind
> would generate a lot of force against that much sail area. You also need
> to allow for the effect of gravity on the balloon. It will tend to sag under
> its own weight. And it needs to be steerable to a pointing accuracy within
> a fractional degree, which isn't easy for a more conventional dish, and
> would be more difficult for an anchored balloon.
Yeah, I'm not sure how far I want to go with this. But, what the hell. Maybe
the baloon would be fastened to nylon guys connected to phone poles. Some kind of
electric winch system on each of the guys would steer the thing. There'd
have to be enough guys connected to the baloon, obviously, to keep it from moving
under light winds. And, air pressure, probably with a pump going all the time
would keep the basic shape.
I sort of imagine using a material something like the canvas used
in sailboats -- somehow modified to be reflective
at RF. And, judging from the Macy's day Parade, I don't think baloons
necessarily have to be spherical. Still, sewing a canvas baloon into a perfect
parabola -- it's not obvious to me how to do this. It'd be a big advantage to
figure out a shape where the focal point was outside of the baloon.
>
> You'd need to inflate the balloon to pretty high pressure to avoid too
> much sag and deflection by gravity and wind (a deflection of 1 inch
> over the surface would defocus it at 10 GHz, that's a tolerance of less
> than 0.01%). And that in turn requires a strong balloon, which means
> a very heavy balloon. We're talking about a balloon made of material
> heavier than that of a truck tire. I'd estimate it would weigh about
> 395,000 pounds. A steel truss dish would actually be lighter.
I'm not sure that heaviness is required for something to hold it's shape.
Fabric-type materials are going to be pretty in-elastic once the thing
is inflated to the maxiumum size. Still, when the wind kicks up, obviously
a 30m baloon is gonna' go where it wants to go. Really the issue is whether
under milder winds how much air pressure is required for it to hold its shape.
And is there a spot on the planet where the wind is mild enough, on enough days
to make the thing usable.
> This balloon antenna would be bigger than the Goodyear blimp, and
> the blimp ripples and distorts in the wind more than would be acceptable
> for a dish. Large airships had to go to rigid construction to maintain
> structural integrity in the wind. Even so, the Los Angeles broke up in
> a thunderstorm. Very large light structures aren't very durable in
> atmosphere. In space, a large inflatable antenna makes a lot more
> sense. The pressure can be low and the structure can be very thin
> and low mass. The material of Echo I was such a thin mylar that
> a piece of it would dance on an air current like smoke. But on Earth,
> we need to build more robustly.
Is 30M really that big -- in terms of inflatable stuff? I always thought
the Goodyear blimp was bigger. And there, the forces and risk to human
life are whole 'nother ballgame. A balloon reflector also has the advantage
of not having to be light.
>
> The requirement that the dish be steerable is the real killer. If we
> can make do with a fixed dish, we can make very large dishes that
> aren't so massively built. An example is the 1000 foot Arecibo dish.
> It was a massive engineering project, but it was much less massive
> than if it were required to be steerable. I'm dubious that our civil
> engineers are up to the task of making a steerable dish that size.
> I expect that they'd start to run into strength of materials limits
> trying to maintain surface accuracy while swinging it around at
> all angles.
>
Gary Coffman
On Sun, 26 Apr 1998 08:47:58 +0100, "Ian White, G3SEK" <G3...@ifwtech.demon.co.uk> wrote:
>Gary Coffman wrote:
>>
>>The requirement that the dish be steerable is the real killer. If we
>>can make do with a fixed dish, we can make very large dishes that
>>aren't so massively built. An example is the 1000 foot Arecibo dish.
>>It was a massive engineering project, but it was much less massive
>>than if it were required to be steerable. I'm dubious that our civil
>>engineers are up to the task of making a steerable dish that size.
>>I expect that they'd start to run into strength of materials limits
>>trying to maintain surface accuracy while swinging it around at
>>all angles.
>>
>
>a few degrees. The penalty is a very fancy line-source feed system which
>compensates for the fact that a sphere doesn't have a single focal point
>like a paraboloid. (I'm still struggling to see how a steerable feed
>could be implemented inside a balloon, but it's fun to speculate...)
The Arecibo dish steers the feed, but not the dish. I don't think it is
feasible, from a civil engineering perspective, to physically steer
such a large dish. As to steering a feed inside a balloon, I would
assume that a three wire suspension with 3 tensioning motors
would be used. These would be anchored to an interior surface
great circle of the balloon.
Gary Coffman
On Sun, 26 Apr 1998 14:29:36 -0700, Cathryn Mataga <ke...@junglevision.com> wrote:
>Gary Coffman wrote:
>> Well, a balloon would form a spherical reflector instead of a parabolic
>> reflector. That doesn't totally kill the idea, but does require a different
>> sort of feed. What kills the idea is anchoring and pointing that 30 meter
>> balloon in the wind and under gravity loads. Even a relatively light wind
>> would generate a lot of force against that much sail area. You also need
>> to allow for the effect of gravity on the balloon. It will tend to sag under
>> its own weight. And it needs to be steerable to a pointing accuracy within
>> a fractional degree, which isn't easy for a more conventional dish, and
>> would be more difficult for an anchored balloon.
>
>Yeah, I'm not sure how far I want to go with this. But, what the hell. Maybe
>the baloon would be fastened to nylon guys connected to phone poles. Some kind of
>electric winch system on each of the guys would steer the thing. There'd
>have to be enough guys connected to the baloon, obviously, to keep it from moving
>under light winds. And, air pressure, probably with a pump going all the time
>would keep the basic shape.
Remember the weight of the balloon. If it is sitting on the ground, the bottom
will be flattened by that weight. And unless the antenna is pointing down, the
bottom will be part of the reflector surface. You can't let that deflect more than
0.01% or you'll ruin the surface accuracy of the reflector.
>I sort of imagine using a material something like the canvas used
>in sailboats -- somehow modified to be reflective
>at RF. And, judging from the Macy's day Parade, I don't think baloons
>necessarily have to be spherical. Still, sewing a canvas baloon into a perfect
>parabola -- it's not obvious to me how to do this. It'd be a big advantage to
>figure out a shape where the focal point was outside of the baloon.
Yes, but any shape other than a sphere will be very hard to maintain by
air pressure alone. Again I point out the surface accuracy required is
a deviation of less than 1 inch over the entire surface of the reflector.
I'm talking about a systematic error in the surface shape here, not just
a small dent or two. A gravity loaded balloon will be an oblate spheroid,
not a sphere.
>> You'd need to inflate the balloon to pretty high pressure to avoid too
>> much sag and deflection by gravity and wind (a deflection of 1 inch
>> over the surface would defocus it at 10 GHz, that's a tolerance of less
>> than 0.01%). And that in turn requires a strong balloon, which means
>> a very heavy balloon. We're talking about a balloon made of material
>> heavier than that of a truck tire. I'd estimate it would weigh about
>> 395,000 pounds. A steel truss dish would actually be lighter.
>
>I'm not sure that heaviness is required for something to hold it's shape.
>Fabric-type materials are going to be pretty in-elastic once the thing
>is inflated to the maxiumum size. Still, when the wind kicks up, obviously
>a 30m baloon is gonna' go where it wants to go. Really the issue is whether
>under milder winds how much air pressure is required for it to hold its shape.
>And is there a spot on the planet where the wind is mild enough, on enough days
>to make the thing usable.
This balloon has a surface area of 2827 square meters. A 10 kph wind will
load one face of it with a force of about 12,720 newtons. That's the equivalent
of a large cement truck sitting on the balloon. It must not deflect as much as
1 inch under that load. Do you want to reconsider the stiffness required?
>> This balloon antenna would be bigger than the Goodyear blimp, and
>> the blimp ripples and distorts in the wind more than would be acceptable
>> for a dish. Large airships had to go to rigid construction to maintain
>> structural integrity in the wind. Even so, the Los Angeles broke up in
>> a thunderstorm. Very large light structures aren't very durable in
>> atmosphere. In space, a large inflatable antenna makes a lot more
>> sense. The pressure can be low and the structure can be very thin
>> and low mass. The material of Echo I was such a thin mylar that
>> a piece of it would dance on an air current like smoke. But on Earth,
>> we need to build more robustly.
>
>Is 30M really that big -- in terms of inflatable stuff? I always thought
>the Goodyear blimp was bigger. And there, the forces and risk to human
>life are whole 'nother ballgame. A balloon reflector also has the advantage
>of not having to be light.
And it can't be light. The Goodyear blimp is about 30m long, but it is much
smaller in diameter. And as I mentioned, its envelope deflects far too much
to be usable as a reflector. I've acted as transmission engineer on the blimp,
and the amount of sway, twist, and flex of the balloon in normal flight is a little
frightening. If you've watched the Macy parade, you should have noted how
much the balloons nod, sway, ripple, and *bend* even when there is very little
wind. What we need is something that can maintain steel bar like rigidity. That
requires much higher pressure, and that requires much heavier envelope
material (more on the order of a heavy truck tire). Those mass roughly 70 kg
per square meter of surface.
qwa...@hotmail.com
hmm, very interesting. say i made my dish on the ground, largely immobile
using the stressed surface technique - how could I go about steering it just a
small amount? just enough that over the course of a year I will be able to see
most of the sky with it. could I use hydraulics or even an inflatable lifting
system? how could I move the whole dish without bending it and still be able
to take advantage of it being low to the ground? or is it not worth the
effort?
In article <35424468...@news.atl.bellsouth.net>,
ke...@bellsouth.net (Gary Coffman) wrote:
>
> On Fri, 24 Apr 1998 23:17:09 -0700, Cathryn Mataga <ke...@junglevision.com>
wrote:
>
> >Gary Coffman wrote:
> >> Now a 30m truss dish is *huge*, and weighs many tons. As a benchmark,
> >> the 34m dishes at Goldstone weigh in at about 100 tons. Yours won't be
> >> much lighter. A support back design is required to keep your trusses from
> >> deforming under gravity loads (and wind loads) at various pointing
angles.
> >> (This is relatively standard bridge engineering.)
> >
> >Gawd, it seems like a shame that you'd need all that steel just to hold up
a
> >reflector. I sort of imagine, and this is totally stupid Usenet
speculation,
> >some kind of inflatable spherical baloon with reflective material stuck on
half of
> >it -- and with the feed inside the baloon? The outside hemisphere of the
baloon
> >would need to be made of material that was invisible to microwaves.
> >The idea would be to use air pressure to keep the shape of the reflector?
Or maybe,
> >to keep the feed on the outside of the baloon, some kind of inflatable
donut
> >type of shape?
>
> Well, a balloon would form a spherical reflector instead of a parabolic
> reflector. That doesn't totally kill the idea, but does require a different
> sort of feed. What kills the idea is anchoring and pointing that 30 meter
> balloon in the wind and under gravity loads. Even a relatively light wind
> would generate a lot of force against that much sail area. You also need
> to allow for the effect of gravity on the balloon. It will tend to sag under
> its own weight. And it needs to be steerable to a pointing accuracy within
> a fractional degree, which isn't easy for a more conventional dish, and
> would be more difficult for an anchored balloon.
>
> You'd need to inflate the balloon to pretty high pressure to avoid too
> much sag and deflection by gravity and wind (a deflection of 1 inch
> over the surface would defocus it at 10 GHz, that's a tolerance of less
> than 0.01%). And that in turn requires a strong balloon, which means
> a very heavy balloon. We're talking about a balloon made of material
> heavier than that of a truck tire. I'd estimate it would weigh about
> 395,000 pounds. A steel truss dish would actually be lighter.
>
> This balloon antenna would be bigger than the Goodyear blimp, and
> the blimp ripples and distorts in the wind more than would be acceptable
> for a dish. Large airships had to go to rigid construction to maintain
> structural integrity in the wind. Even so, the Los Angeles broke up in
> a thunderstorm. Very large light structures aren't very durable in
> atmosphere. In space, a large inflatable antenna makes a lot more
> sense. The pressure can be low and the structure can be very thin
> and low mass. The material of Echo I was such a thin mylar that
> a piece of it would dance on an air current like smoke. But on Earth,
> we need to build more robustly.
>
> The requirement that the dish be steerable is the real killer. If we
> can make do with a fixed dish, we can make very large dishes that
> aren't so massively built. An example is the 1000 foot Arecibo dish.
> It was a massive engineering project, but it was much less massive
> than if it were required to be steerable. I'm dubious that our civil
> engineers are up to the task of making a steerable dish that size.
> I expect that they'd start to run into strength of materials limits
> trying to maintain surface accuracy while swinging it around at
> all angles.
>
> Gary
> Gary Coffman KE4ZV | You make it |mail to ke...@bellsouth.net
> 534 Shannon Way | We break it |
> Lawrenceville, GA | Guaranteed |
>
qwa...@hotmail.com
five watts?!?! that's incredible! that is the reason why i'm asking all these
questions about big dishes, I want to be able to do fairly low power voice
modes, or maybe digital via moonbounce (among other things).
In article <FvESgYAu...@ifwtech.demon.co.uk>,
"Ian White, G3SEK" <G3...@ifwtech.demon.co.uk> wrote:
>
> Gary Coffman wrote:
> >
> >The requirement that the dish be steerable is the real killer. If we
> >can make do with a fixed dish, we can make very large dishes that
> >aren't so massively built. An example is the 1000 foot Arecibo dish.
> >It was a massive engineering project, but it was much less massive
> >than if it were required to be steerable. I'm dubious that our civil
> >engineers are up to the task of making a steerable dish that size.
> >I expect that they'd start to run into strength of materials limits
> >trying to maintain surface accuracy while swinging it around at
> >all angles.
> >
>
> Tha Arecibo dish is spherical, which allows it to be semi-steerable over
> a few degrees. The penalty is a very fancy line-source feed system which
> compensates for the fact that a sphere doesn't have a single focal point
> like a paraboloid. (I'm still struggling to see how a steerable feed
> could be implemented inside a balloon, but it's fun to speculate...)
>
> A 100ft amateur version of the Arecibo dish was built by W1FZJ/KP4 who
> was working on the full-size project. The feed was mounted on a 50ft
> pole and steered by changing the guy lengths; obviously Sam Harris had
> unique access to the feed design information too. Unfortunately he died
> before it could be fully commissioned, and the dish ended its days as a
> chicken coop.
>
> The 1000ft dish has a 400-450MHz wavegude feed, and has been used a few
> times for amateur moonbounce. 5W puts out a storming signal - so strong
> that I thought it was local QRM and almost ignored it!
>
> 73 from Ian G3SEK Editor, 'The VHF/UHF DX Book'
> 'In Practice' columnist for RadCom (RSGB)
> http://www.ifwtech.demon.co.uk/g3sek
>
qwa...@hotmail.com
I've thought about spherical reflectors before. I was brainstorming an idea of
how to get a *huge* radio telescope into space and construct it cheaply. what
could be better than a huge ballon that inflates itself and is easy to point
where you want it? I also was wondering if there are plastics that can
alternate between rf refection and transparency according to electric
currents, which would make it even easier to steer by just changing which side
is reflective and which is transparent. I thought that that was a bit far
fetched and probably a way off, so what about a balloon with a fine wire mesh
on one side? that may work for a home station too, eh? I could put up wind
breakers that would also serve to keep the background noice down.
In article <3543A740...@junglevision.com>,
Ian White, G3SEK
wrote:
>five watts?!?! that's incredible! that is the reason why i'm asking all these
>questions about big dishes, I want to be able to do fairly low power voice
>modes, or maybe digital via moonbounce (among other things).
>
Sure - they could have done it with less than one watt. A thousand-foot
dish is all it takes.
Gary Coffman
On Mon, 27 Apr 1998 11:57:22 -0600, qwa...@hotmail.com wrote:
>five watts?!?! that's incredible! that is the reason why i'm asking all these
>questions about big dishes, I want to be able to do fairly low power voice
>modes, or maybe digital via moonbounce (among other things).
The gain of the Arecibo dish at 70cm is on the order of 60 db. So 5 watts
of excitation gives an ERP on the order of 5 megawatts. Antenna gain is
a wonderful thing. But really high gains get really expensive. A 1.5 kW
transmitter and a much more modest antenna would be considerably
cheaper (and smaller).
Cathryn Mataga
Cool. All I need is the a 1000 ft dish for my HT. Btw, is there
any reference for feeds for spherical dishes? I assume this is buried
in someone's Phd thesis or something. Or is there like a standard
text that covers this kind of stuff.
Ian White, G3SEK wrote:
>
> wrote:
> >five watts?!?! that's incredible! that is the reason why i'm asking all these
> >questions about big dishes, I want to be able to do fairly low power voice
> >modes, or maybe digital via moonbounce (among other things).
> >
>
Josh J. Haviland
ok, i've been following this for awhile, and have picked up some pretty
good info, and have a few questions to ask, 1 if one did make a dish (solid
not mesh surface) how would you figure where to mount the antenna? and what
would the possible gain of a dish, say 8' - 12' dish have at 144 and 440mhz
general ideas...
thanks
Josh
--
__/////////\ Budgreen, ea...@cleveland.freenet.edu /\\\\\\\\\__
\\\\\\\\\/ Josh...... IS...@hotmail.com \/////////
=================================================================
klo...@toad.net
In article <6ihm55$ss0$1...@alexander.INS.CWRU.Edu>#1/1,
ea...@cleveland.Freenet.Edu (Josh J. Haviland) wrote:
>
>
> ok, i've been following this for awhile, and have picked up some pretty
> good info, and have a few questions to ask, 1 if one did make a dish (solid
> not mesh surface) how would you figure where to mount the antenna? and what
> would the possible gain of a dish, say 8' - 12' dish have at 144 and 440mhz
> general ideas...
> thanks
> Josh
Josh,
One way to estimate the gain in dBi is by using 20*log(D/lambda) + 8, where D
is the diameter of the dish in the same units as lambda.
Bill
Ian White, G3SEK
wrote:
>In article <6ihm55$ss0$1...@alexander.INS.CWRU.Edu>#1/1,
> ea...@cleveland.Freenet.Edu (Josh J. Haviland) wrote:
>>
>>
>> ok, i've been following this for awhile, and have picked up some pretty
>> good info, and have a few questions to ask, 1 if one did make a dish (solid
>> not mesh surface) how would you figure where to mount the antenna? and what
>> would the possible gain of a dish, say 8' - 12' dish have at 144 and 440mhz
>> general ideas...
>> thanks
>> Josh
>
>Josh,
>One way to estimate the gain in dBi is by using 20*log(D/lambda) + 8, where D
>is the diameter of the dish in the same units as lambda.
Unfortunately that won't work for such a small dish diameter - small in
terms of wavelengths, that is.
The gain formula should include an efficiency factor, which is partly
about effectiveness of illumination by the feed antenna, and partly
about the efficiency of the reflector itself. The formula above is based
on optics and assumes that the reflector is many wavelengths in diameter
- which ain't true.
To take the extreme case, the gain and radiation pattern of an 8ft dish
at 144MHz will be pathetic. I can't put any numbers on it, because the
only way to evaluate it is by computer modeling, but would guarantee
that even quite a small yagi will be better.
At the other extreme, a 15ft dish on 440MHz is just about beginning to
work as it should. Experience with moonbounce shows that the gain will
be in the low twenties of dBi, about the same as four 15-20ft yagis.
Bottom line: Josh will do better with yagis.
Gary Coffman
On 3 May 1998 11:58:29 GMT, ea...@cleveland.Freenet.Edu (Josh J. Haviland) wrote:
> ok, i've been following this for awhile, and have picked up some pretty
>good info, and have a few questions to ask, 1 if one did make a dish (solid
>not mesh surface) how would you figure where to mount the antenna?
Depending on what you want to do with it, either as high as possible, or
with a clear view of as much of the sky as possible.
>and what
>would the possible gain of a dish, say 8' - 12' dish have at 144 and 440mhz
>general ideas...
>thanks
>Josh
You'd be hard pressed to get 11 dbi from a 3m dish at 2m. Any good long
boom yagi would beat it. At 70 cm, you could get about 20 dbi. That starts to
compare to a small stacked set of yagis. (Both of those figures depend on
getting good feed illumination, which would be very difficult on 2m, and
still difficult at 70 cm.) Small dishes like that don't start to become worthwhile
until you move up into the GHz region. At S band, it will have a gain of about
35 dbi. And at X band, it would have a gain of about 48 dbi, and that's worth
talking about.
klo...@toad.net
In article <in5byTAV$WT1...@ifwtech.demon.co.uk>#1/1,
> >>
> >>
> >> ok, i've been following this for awhile, and have picked up some pretty
> >> good info, and have a few questions to ask, 1 if one did make a dish
(solid
what
> >> would the possible gain of a dish, say 8' - 12' dish have at 144 and
440mhz
> >> general ideas...
> >> thanks
> >> Josh
> >
> >One way to estimate the gain in dBi is by using 20*log(D/lambda) + 8, where
D
> >is the diameter of the dish in the same units as lambda.
>
> Unfortunately that won't work for such a small dish diameter - small in
> terms of wavelengths, that is.
>
> The gain formula should include an efficiency factor, which is partly
> about effectiveness of illumination by the feed antenna, and partly
> about the efficiency of the reflector itself. The formula above is based
> on optics and assumes that the reflector is many wavelengths in diameter
> - which ain't true.
Ian, since the actual optics equation is (pi*D/lambda)**2, there is an
efficiency term of about -1.95 dB (64%) built in. The equation is right out of
a CCIR report and is used to *estimate* the gain of a circular aperture
antenna when no other info about the antenna is available.
> To take the extreme case, the gain and radiation pattern of an 8ft dish
> at 144MHz will be pathetic. I can't put any numbers on it, because the
> only way to evaluate it is by computer modeling, but would guarantee
> that even quite a small yagi will be better.
So from the above, an estimate of the *best possible gain* of an 8' dish at
144 MHz would be about 9 dBi. No argument as to the gain being pathetic. I'll
bet if you did model such an antenna, the gain would be much less than 9 dBi,
but you would spend at least a couple of hours doing the modeling.
> At the other extreme, a 15ft dish on 440MHz is just about beginning to
> work as it should. Experience with moonbounce shows that the gain will
> be in the low twenties of dBi, about the same as four 15-20ft yagis.
>
> Bottom line: Josh will do better with yagis.
No argument, but that wasn't Josh's question.
73, Bill N3WK
Josh J. Haviland
<snip>
>Depending on what you want to do with it, either as high as possible, or
>with a clear view of as much of the sky as possible.
>
>>and what
>>would the possible gain of a dish, say 8' - 12' dish have at 144 and 440mhz
>>general ideas...
>>thanks
>>Josh
>
>You'd be hard pressed to get 11 dbi from a 3m dish at 2m. Any good long
>boom yagi would beat it. At 70 cm, you could get about 20 dbi. That starts to
>compare to a small stacked set of yagis. (Both of those figures depend on
>getting good feed illumination, which would be very difficult on 2m, and
>still difficult at 70 cm.) Small dishes like that don't start to become worthwhile
>until you move up into the GHz region. At S band, it will have a gain of about
>35 dbi. And at X band, it would have a gain of about 48 dbi, and that's worth
>talking about.
>
i see, well thanks for the info, i now see that a dish of the size i
could get/make for 144Mhz wou;d not have the gain of my quagi.. hmm maybe
i'll just throw together another or 3 quagi's and just phaze them :), but
if i ever go get any equiptment in the Ghz range i'll keep this in mind.
thanks for the help
Ian White, G3SEK
N3WK wrote:
>In article <in5byTAV$WT1...@ifwtech.demon.co.uk>#1/1,
> "Ian White, G3SEK" <G3...@ifwtech.demon.co.uk> wrote:
>> The gain formula should include an efficiency factor, which is partly
>> about effectiveness of illumination by the feed antenna, and partly
>> about the efficiency of the reflector itself. The formula above is based
>> on optics and assumes that the reflector is many wavelengths in diameter
>> - which ain't true.
>
>Ian, since the actual optics equation is (pi*D/lambda)**2, there is an
>efficiency term of about -1.95 dB (64%) built in. The equation is right out of
>a CCIR report and is used to *estimate* the gain of a circular aperture
>antenna when no other info about the antenna is available.
>
frequencies estimate of overall inefficiency, but it still takes no
account of the special inefficiencies of dishes that are electrically
very small.
My point was that there are three major factors that can reduce the gain
from the CCIR estimate:
1. Feedhorn inefficiency (failure to illuminate the reflector correctly
and with no overspill)
2. Inefficiency of an electrically small reflector (shortfall below the
value assumed from optics)
3. Loss of gain due to blockage of reflector aperture by the feed
antenna and its supports.
In most microwave dishes you're doing well to achieve 64% overall
efficiency. With very small dishes, factors 1, 2 and 3 all get markedly
worse, and there's no way around it because the three factors are
interlinked. The geometry of the main reflector dictates the beamwidth
of the feed, which in turn dictates the feed antenna's own aperture
size.
The best available non-waveguide feed for lower frequencies (the "ring
radiator" dipole) has a 2-wavelength subreflector diameter - which
almost completely blocks a small dish.
>> To take the extreme case, the gain and radiation pattern of an 8ft dish
>> at 144MHz will be pathetic. I can't put any numbers on it, because the
>> only way to evaluate it is by computer modeling, but would guarantee
>> that even quite a small yagi will be better.
>
>So from the above, an estimate of the *best possible gain* of an 8' dish at
>144 MHz would be about 9 dBi. No argument as to the gain being pathetic. I'll
>bet if you did model such an antenna, the gain would be much less than 9 dBi,
>but you would spend at least a couple of hours doing the modeling.
>
That was my whole point: you'd waste a lot of time, only to discover
that the CCIR estimate is misleadingly high.
>> Bottom line: Josh will do better with yagis.
>
>No argument, but that wasn't Josh's question.
Running a monthly Q&A column in a magazine, I quickly learned that it's
sometimes a disservice to answer only the question that was asked,
because it often isn't the right question. The later posting from Josh
suggests that "use yagis" was the answer he really *needed*.
klo...@toad.net
In article <BKHDZIAs...@ifwtech.demon.co.uk>#1/1,
"Ian White, G3SEK" <G3...@ifwtech.demon.co.uk> wrote:
>
> >So from the above, an estimate of the *best possible gain* of an 8' dish at
> >144 MHz would be about 9 dBi. No argument as to the gain being pathetic.
> >I'll bet if you did model such an antenna, the gain would be much less than
> > 9 dBi, but you would spend at least a couple of hours doing the modeling.
> >
> That was my whole point: you'd waste a lot of time, only to discover
> that the CCIR estimate is misleadingly high.
>
Actually, if I used the CCIR equation and got a value of 9 dBi, I would stop
right there and look for an another antenna with better gain. I guess I
assumed Josh would do the same. You're right about providing an alternative
solution. 73,