Hey 300ohm,
are you sure it was just 7 foot ? No I'm sure some were a lot larger than 7 ' I just picked 2 meter for ease of building
the model,close to same size as p-7 or cm 4251, and its close to the half I picked up.
Also I think 2 meter is about min. size to compete with easier to build panel
type antennas like 8 bays and dbghs.

should I limit the size of the reflector and elements to something like a 2 bay? The one's I ended up with on the model were 12.5 inches.

And I'm probably stating the obvious here but yes the larger the feed
elements reflector is the more "shade" it casts on the parabola, but to small
and the f/b ratio suffers and that back-side direct signal is probably out of
phase and multi-path (with exceptions).

So for a feed antenna you want a beam wide enough to illuminate the whole dish
at the distance of the focal point and a good f/b ratio. Smaller beamwidth
higher gain feeds just concentrate their sensitivity to the spot that's shaded
by them. 8-10 dbi is in the right range IMHO.

The clearstream2's specs seem pretty close to the biquad I used.

I'm going to swap the feed on the posted model with other types for
comparison. I'm the curious sort like kc2cbd.

I'll play with that to when I get a chance (post 187),it looks easier to build
only 3 feet wide though? And what kind of feed cuz I'm not up to modeling
that feed (maybe one like the p5 feed 300ohm posted.

300ohm did you notice the model of the p-5 feed has a Quite high SWR at
channel 22 and below. Before I put it in front of a dish I'd want to make the
folded dipole longer (probably tuned for the higher channels we don't have
now) oh and turn it around (and I don't like folded dipole models)
Do you have anything like that, did you play with it after you made the
one from measurements of the original?
IF not I guess I could modify the feed you posted or cut off all but one
director on a yagi.
Just trying to save some time.

oh and turn it around (and I don't like folded dipole models)
Do you have anything like that, did you play with it after you made the
one from measurements of the original?
Nope, I dont.

You can get good folded dipole measurements for the models with the Yagi Calculator program, to save time on optimizing.

So for a feed antenna you want a beam wide enough to illuminate the whole dish at the distance of the focal point and a good f/b ratio.

A bi-quad actually fits that bill better than a bowtie or folded dipole. (but not as wideband as a bowtie) Its got a nice round almost perfect bubble pattern.
Some 4nec2 biquad analysis for 2.4 Ghz here, just rescale for the uhf frequencies you want.

http://www.lecad.fs.uni-lj.si/~leon/other/wlan/biquad/index.html

Well, I'm hoping I could use half of an AD C4, that is collecting dust, making it a C2.
W/ the advertised pattern of 70deg, should I position it on my 10' parabolic, so the pattern reaches the edges?

I've already added a few inches to the center support to compensate for the reflector distance.

I based this on the standard parabolic focal distance formula.

Thanks, people!

Craig

Stereocraig
Well, I'm hoping I could use half of an AD C4, that is collecting dust, making it a C2.
W/ the advertised pattern of 70deg,

The c2 ( clearstream 2) is very close to the biquad I used (for the model) in published specs.

should I position it on my 10' parabolic, so the pattern reaches the edges?

Hopefully when the feed (c2) is at the focal point it's beamwidth will also hit close to the edges.
If the two can't happen at the same time then I'd have to build a model to be sure where to sacrifice for best results.
My guess would be toward focal point being more important. At least then the system would definitely act as a parabola
(of smaller diameter) and only lose the unlit edges.
Ten feet! Wow. Quick estimate a ten foot dish would be 3 dB higher than the 2 meter I posted so (average net gain of ~22+ Dbi)

What is the focal length you calculated?

How high up are you going to try and hang/mount this monster?
How are you going to hang/mount this monster?

If/when you do this please send me a photo:)

Hey 300ohm , that was an interesting link you posted and I appreciate it.
although I'm a little confused.
The model I referred to in post # 254 used a biquad that was optimized by nikiml for uhf already.
Don't you like the biquad used? Is there something wrong with it?
Seems like reinventing the wheel to rework the biquad from microwave frequencies instead of using
nikiml 's model that is already optimized for uhf and 300 ohm impedance not 50 ohm.

Maybe it's my handle of ghz24 ? Which was conceived due to an interest in high gain 2.4 ghz wireless lan antennas.
If you want to compare construction techniques ect. I'm game but I'll have to measure mine it's been 10 or 11 years since I built it
(there was no 802.11 g back then).:) hows here for that discussion ? http://www.digitalhome.ca/forum/showthread.php?t=115777&page=3

On to post #187
Considering the limited measurements available for the antenna pictured on post 187
and only wanting an idea of potential performance anyway I created a model that approximates the reflector of this trough parabola.
2 meters high with a focal point of 0.5 meters one 1 meter wide.
remembering the feed 300ohm posted for the p5 on post #73 of this thread and how
the pattern struck me as tall and narrow to be feeding a round parabola I flipped the feed you posted around
and put the driven element at 17 inches (as a starting point) and got this as the first run.

These are raw gain not net
max 16.87
min 15.46

Average 16.42


High SWR kills all the channels from 21 and below I've been unable to get the swr down to acceptable values across the uhf band yet.
I'm considering alternative feed designs but the raw gains surprised me for a two dimensional parabola.

The model I referred to in post # 254 used a biquad that was optimized by nikiml for uhf already.
Don't you like the biquad used? Is there something wrong with it?
Nah, its fine, I just forgot about it.

High SWR kills all the channels from 21 and below I've been unable to get the swr down to acceptable values across the uhf band yet.
I'm considering alternative feed designs but the raw gains surprised me for a two dimensional parabola.
Thats not bad, and compares to these single channel Scala antennas:

http://www.kathrein-scala.com/catalog/PR-TV.pdf

http://<a href="http://s25.photobucket.com/albums/c92/amptek/?action=view&amp;current=DSCF3047.jpg" target="_blank"><img src="http://i25.photobucket.com/albums/c92/amptek/DSCF3047.jpg" border="0" alt="4251 Clone, Half of AD-C4 mounted to satellite pipe"></a>GHZ: The( f )on this dish, is 42.666", and I needed to rig a bracket to move the C2 back farther.

I will probably use a decent length of square tubing to spread out the torque and clamp that to the Rohn.
I probably won't be putting this up on the tower too high, maybe just enough to clear the house.

It sounds like there should be an attachment button somewhere, but all I see is the pop up to link to photobucket.

How to create a photo album for use on the Digital Forum:

http://www.digitalhome.ca/forum/showthread.php?t=92167

Thank you, 300!

http://www.digitalhome.ca/forum/picture.php?albumid=921&pictureid=4852

42.6" now I see why you wondered where you should place the feed.
I played with the numbers and from trig I came up with a 10 foot dish with a focal point
of 42.6 needing ~109 degree beamwidth ( inv. tan of 60/42.6) X 2
my trig is a little rusty so if I messed that up I hope someone corrects me.
I'm pretty sure that's right. So I backtracked and figured how much of the dish
is the "beamwidth" illuminating and I don't remember the exact answer I got but it was close to 6 foot
of the 10 foot dish. Then, wait the 2 meter model I just studied has a similar focal point 42 vs 39
so I applied the same test to the 2 meter dish and it failed the test
and should have "needed" almost 100 deg beam width as well. :confused:

So I built a model of the dish you intend to use 10 foot diameter focal point
of 42.6 . Set the reflector on the biquad to 5 inches (like a c2) instead of
8.75 and let the optimizer pick the best distance (for 585 MHz)
it came up with 39.2 inches to the driven elements and 5 inches to the reflector makes 44.2 inches.
Again it optimized to a spot where the focal point is between the reflector and driven elements.
I haven't run a sweep yet but it doesn't look like any lack of illumination is
hurting the performance to much. I got 21.4 dBi @ 473 MHz and 22.9 dBi @ 585 MHz and 22.57 @ 695 MHz. I guess the arbitrary 3dB beamwidth is not as meaningful for a feed.(8 -3 still is 5 plenty for a feed
about 3 dB better than the 2 meter as expected (swr is above 3 here though)

I can let the optimizer try at 695 tonight and see how far it moves the feed for the higher frequency.

I finally got the p-5 feed to have swr less than 2.7 across the band so I'll be posting the results of post 187 soon.

wait don't cut that pipe/tube yet. I just ran the sweep and all the things from earlier are true but the swr is unacceptably high (>2.7) from 17 to 29 and 42 on up. but gain stayed constantly above 21 dBi

I'll rerun the optimizer tonight to sweep every other or third channel and look for a better distance
for all channels.

The existing pipe is stepped down to fit the opening in the plates, so the (f) travel is limited.

I also remember reading something about using plastic VS metal pipe, so I may just pick up some SCH80 components.

Any recommendations from anybody on a budget real time signal meter,or should I just pick up a converter box and a cheap TV to bring outside?

Any recommendations from anybody on a budget real time signal meter,or should I just pick up a converter box and a cheap TV to bring outside?
A lot, if not most, of the converter boxes just have stupid bars, no numbers. Two converter boxes that have both signal strength and signal quality meters displayed in numbers are the Sansonic FT300a and the Apex DT502.
The FT300a has a further advantage of being able to be powered by a 5 volt battery.

So for a feed antenna you want a beam wide enough to illuminate the whole dish at the distance of the focal point and a good f/b ratio. Smaller beamwidth higher gain feeds just concentrate their sensitivity to the spot that's shaded by them.

Oldsparks' uhf lpda http://www.digitalhome.ca/forum/showthread.php?t=148830 post 3, looks like a good candidate for a feed element for a Kathrein Scala type half parabola.
http://www.kathrein-scala.com/catalog/PR-TV.pdf
This type of half parabola frame could be built by DIYers using grey pvc pipe and a heat gun. :)
(design note: the grey electrical pvc pipe is sold is 10ft long pieces, but about 2.5" of one end is flared out to allow coupling, making the effective length about 9 ft 9 inches. )

http://img35.imageshack.us/img35/7042/oldsparkslpdapattern.jpg

Stereocraig: I've found the problem that's causing the high swr on the
10 foot model. I don't know if it pertains to what you are trying to do
because the c2 isn't the biquad in the model so it may react differently.
When I moved the reflector to 5 inches behind the driven element instead of
8.75, that caused the swr problem. On a stand alone biquad (without the dish)
the swr is elevated but acceptable ~(2.4 to 2.0).
So with out a model of a c2 I think I'm at the limits of the assumption
of the c2 and the biquads similarity. They maybe very similar, a 5 inch
spacing to the reflector raises the gain Of the biquad especially in the
upper part of the band.
So here's my advise place the c2 at the ~ 39.2 inch and if you have a problem
on the channels I noted above try moving the reflector of the c2
~3.75 further from the driven elements (should be about 5 inches now)
or 8.75 total. Hope that helps.

300ohms: You always bring me the coolest links. Just before I got the modded
p5 feed swr under control I was thinking about a 2 element lpda with a reflector (rod)
as a possible alternate feed.
It seems like the p5 feed and the lpda you suggest would cast a smaller
shadow (aperture calculations would maybe disagree)
The post 187 trough parabola is hard to make wider
if I use the p5 feed (3 element yagi) feed then the wider the parabola gets
the wider the horizontal beamwidth (until the lobe splits)
And trying to feed it with a biquad gets results like this.
I'm not sure what happened here maybe the biquad is shading the parabola
the focal point is only 0.5 meters and it's 6 feet wide.

http://www.digitalhome.ca/forum/picture.php?albumid=924&pictureid=4877

I'll have to try a trough with a focal point of 1 meter to better compare
and find out if it's the trough shape or the short focal length causing this.
There are patterns in the same album.
Thoughts?

GHZ24 I did a model of the post 187 parabola and whenever I set a symbol for the width and use that symbol the evolve goes loses the parabola.out file in temp.
I made another like it with a 12 inch wing on each side, so it will be 5 feet wide, but I haven't tried it yet.
I used a biquad on my first parabola model, but I was pondering the original bowtie and a curtain quad on it.

Thoughts?

Maybe an additional horizontal Paraflector type parabolic reflector would corral the beams and add gain and still could be built by a DIYer using mesh.

kc2cbd: I'm not sure what problem you're having. I posted the model (I referred to in post 265 of this thread) as trparab187.nec (http://ghz24.freeshell.org/antenna/models/newermodels/trparab187.nec) if you want to try the evolver on it or compare , the width is the "SY py=18" 18 is half of parabola width. All the other variables are in the feed part (the first 100 tags) so you can erase them if you install a feed of your choice.
That is the parabola starts at tag 100 to leave room for a feed. What do you want/hope the evolver to do?
Would you post the model that's giving you problems? What value did you use for a focal point?
As the parabola gets wider in the model posted the forward gain lobe gets wider until
the lobe splits/forks
Also posted a page here (http://ghz24.freeshell.org/antenna/trparab187.html) with net and raw gains and swr for uhf frequencies.

300: Where are you thinking of a parabola at? A second big parabolic?
Maybe bend the para-reflector rods like a chevron? Or a little one behind/for the feed?
I'm pretty sure the focal point is to close to benefit from the biquads qualities.
A dish/trough 6.5 feet high with a focal point of .5 meters maybe a little close.
When I get some time I intend to recreate a trough like this only with a longer focal length and see
if the biquad is useable with a trough at all.

I have pretty much the same trough parabola with a biquad feed with a reflector behind it.
The dimensions were 36 inches wide by 7ft tall to match the one in post 187 but I didn't model the bowtie feed.
I used the build program to make the parabola with the above dimensions and a 0.5 meter for the focal point.
All I was evolving was to see if a wider parabola would be any better and the best point for the biquad because I really didn't have a clue where to put it in relation to the front of the parabola.
The evolver can't losing the temp file for some reason, and I'm pretty sure it was with the width symbol cause when I ran it without that symbol it worked fine.

300: Where are you thinking of a parabola at? A second big parabolic?
No, two identical Paraflector type parabolic reflectors (like post 187), one mounted vertically and one horizontally connected with crosses, with 2" X 4 mesh going in the 2" up spacing on both.

Two other ideas that may trigger some more.

On another website I saw that someone had used what looked like an old aluminum shovel (imagine an old fashioned aluminum shovel for clearing snow) as a reflector. clear off any paint -- make a nice shine and eliminate the handle -- drill a few holes for mounting and you are there.

Also, I have personally done the same thing with a one-quater section of an old aluminum "clothes dryer" tub. Cut in half first -- then in half again. Removed the coating (not paint), but paint removers can make some of the soft plastic cover softer and easier to work with. Remove and then flip on an angle. Used currently on one of my Toronto "yagi" Antennas that is in the attic -- distance is about 11 inches away from the driven element. Receive all Toronto stations at 100% including 9.1.