The wire model in post # 354 has 6 GA wire not 12 GA. Switching to 12 GA lost about .5 dB. Actual beamwidth is 13 degrees. We should get some or all of that back due to smoother horizontal curve ( gave up 0.6 dB from the chevron approximation)
I have fixed the models to be 12 gauge even though they have served their purpose of providing these numbers.


8x12 optimized for 585 MHz

SY f=9
SY rl=47.2
SY x=44.36679 *
SY a=3.65
SY b=9.059382
SY c=16.9456

3.65 =a or ~3 5/8"
12.709382 =a+b or ~12 3/4"
29.655022 =a+b+c or ~29 5/8"

8x12 optimized for 473 MHz

SY f=9.272709
SY rl=47.2
SY x=39.20542 *
SY a=3.65365
SY b=9.290371
SY c=16.23624


3.65365 =a or ~3 5//8"
12.944021 =a+b or ~12 15/16"
29.180261 =a+b+c or ~29 3/16"

And for comparison here are measurements from the pre-chined chevron (perfect vertical curve) 2,4 and 6 feet.

3.65
14.626
32.908

* I would have thought x would be the opposite (longer focal point for longer wavelengths)

The ~fractional numbers are the height of the reflector (laying on its back) along the vertical seam (the X value along the Z axis) at 2 foot, 4 foot, and 6 foot (the top and bottom) from the center.
I intend to use them as a guide in building the full size model.
So if it's laying flat on its back I'd overlap the fence enough to raise the now reflector to 3 5/8 inches at 2 feet above (and below) the center
Tie the fence at 2 foot then overlap more to raise the 4 foot mark to 12 15/16 inches above the floor.
Again tie or affix the fence at the +and- 4 foot points.
Finally overlap the final amount to raise the top and bottom to 29 3/16 inches.
This would be for the 473 MHz optimized numbers.
I'm going to swap the two x values in the two models to see what difference is attributable to the reflector changes the optimizer made. To help decide where between the low and middle optimized values to use for the actual build. Because I'm only building one version at this size.


I have been working on ldpa feed designs, even though I'm very happy with oldsparks model.
I was thinking how close the 8x12 size is to the finco 400 and possibly integrating the reflector of the finco as a support/skeleton if stronger support is needed and as a plus maybe add vhf high to the mix for sort of an extreme deep fringe combo.
Then it occurred to me if the finco reflector at approximately 8 foot wide can
reflect VHF high then maybe an 8 foot wide parabola is big enough to work (at least some) for VHF high.

So I whipped up a 3 element high VHF lpda (http://ghz24.freeshell.org/antenna/models/newermodels/parabolas/feeds/lpdavhfhigh.nec) dropped it about 3 feet in front of the reflector and...
Here's a peek at some potential. raw net SWR

max 16.15 max 14.9778369841 max 2.2808298055
min 13.35 min 9.3974272956 min 1.644934137

Average 14.9314285714 Average 11.9683895836 Average 2.0058468817


I haven't tried to make a single both bands lpda yet.

Then it occurred to me if the finco reflector at approximately 8 foot wide can reflect VHF high then maybe an 8 foot wide parabola is big enough to work (at least some) for VHF high.

8 ft wide is actually big enough to work for 1/2 wavelength vhf-low down to channel 3. :p
For vhf-hi its in the area of a 1.5 to 2 wavelengths reflector, which could mean funny lobes.

That is a good idea if it doesnt interfere with the uhf gain, which it might not. Then hook the two driven elements together with a UVSJ.

I used the wrong spreadsheet template the equation for mismatch loss was 300 ohms instead of 75.
so the VHF numbers for net gain should have been

12.8652391904 at 7
13.3347430919
13.8950247199
14.563243644
15.2037878107
15.5819174877
15.4318546745 at 13


14.4108300884 average


When I optimized the 8x12 for 585 and 473 the optimizer changed the angles of the facets and the feed distance (x ) by 5 inches.
To decide which reflector to pursue (for the broadest utility) I swapped the feed distance on both and ran a scan.

opt for 585 (x=39.20542)
net max 23.76 at 45
min 21.08 at 14

Average 22.92
swap x (x=44.36679)
net max 23.06 at 44
min 21.47 at 51 21.615 at 14

Average 22.52
opt low (x=44.36679)
net max 23.52 at 41
min 21.60 at 51 21.91 at 14

Average 22.87
low swap (x=39.20542)
net max 24.18 at 43
min 21.54 at 14

Average 23.36


I'm thinking the second set for the reflector (feed distance easily adjusted to match local needs/wants ).

But that reflector seems to drop off quick around channel 50. about a dB

Opinions ?

Was considering if something like this, rescaled to UHF, might work for co-channel rejection....

maybe the screen can be even tighter to act also as a Faraday Cage...

At high UHF this is buildeable, but for my VHF #13, it would be too HUGE.

http://qsl.net/yu1aw/ANT_VHF/Shortened%203D%20Corner%20Reflector%20Antenna.pdf

From reading the 3 D Corner reflector article:

Seems like that design gives both very good forward gain and very good rear rejection (!).

At one point, the rear null is touching near the -50 dB line (!)

But at UHF rf ch 50, 3 wavelengths ( 3 lambda ) is near 1.3 metres.
So looking at a half box reflector around 4 feet on a side.

Reminds me of a KISS (keep it simple) Reflector design being worked on in another thread. But on that one they started to use a 4 element LPDA - log periodic - antenna as the "illuminating" "receiving" antenna.

But the KISS reflector is trying to simplify a parabolic reflector design for UHF - if I remember correctly.

This might be a good way to simplify the KISS reflector design EVEN MORE !
(because the KISS reflector had 4 reflector planes - I think - and this one has 3 )
... and the conclusion that you could cut off the bottom plane, without many negative effects - helps too.

I wonder what would happen if you used this reflector on the KISS design.
ie. Used the 4 element LPDA feed - like the KISS does.

mrvanwinkles,
Reminds me of a KISS (keep it simple) Reflector design being worked on in another thread. But on that one they started to use a 4 element LPDA - log periodic - antenna as the "illuminating" "receiving" antenna.

I thought that was this thread ? Is there another thread developing the kiss reflector ?
I used an 8 element lpda in the KISS reflectors I modeled for UHF and a 3 element for the results in post #363 ( if you count each dipole in the array as an element )

I still plan to develop these quasi parabolas. But my next model will be from measurements of a full scale prototype.

I thought that was this thread ? - Yes it is ...

Is there another thread developing the kiss reflector ? - NO ...


No, they [Moderators?] moved these few posts here - to this thread - from another thread - because I mentioned about the KISS design being worked on here - and how much the half cube reflector reminded me of the KISS reflector and work .

In the other Thread, there was someone working on trying to get a far off station, with a relatively strong co-channel / same channel station (rf 50) - interfering station to the rear - and they were looking for an antenna design to reject signal from the rear / give a very high front to back ratio ... but also a high gain from the front ... and another member posted the half cube reflector article / link.

They were looking at Phase Cancellation techniques in that other thread.
Antenna Stagger Stacking and such ...

Is there a way to calculate the gain of my antenna by the voltage that is produce by it...87,000 micro volts ?:rolleyes:

Is there a way to calculate the gain of my antenna by the voltage that is produce by it...87,000 micro volts ?
The voltage at the antenna terminals is from all frequencies, TV or not.

I just aquired a used Channel Master 4250 6' parabolic uhf tv antenna.
The antenna is in great condition other than rusted rivets, which one would
expect from an antenna of it's age. I will be dismantling the antenna and clean it up, then replace all the rivets with either stainless steel or aluminum ones so they won't rust.
I noticed upon measuring the bow tie elements that they are 2.5" shorter
than the elements on the CM-4221. Being that the CM4250 was designed for channelas 14-83 and now we only use 14-51, what would be the best length
for the bow ties. I can either buy new aluminum & make new bow ties or just use aluminum rods like the CM4221HD uses I guess. Would there be a reduction in efficiency if I use rods over a solid bow tie ?
Also If I use longer elements I guess I would also have to resize the reflector grid which is behind the bow ties now.
Is there anyone who knows where I can get the aluminum re-anodized or is there a conductive paint available. As I clean off the rust from the rivets I'm rubbing off the anodizing protection on the aluminum rods.
Any help would be greatly appreciated, I really want to make this antenna like new & get it up in the air to see what she can do.

Bruce.

Toronto.

Wow, great score!

I'm working on a 4251 clone and found measurements at this guy's site.

I made my bow ties out of circuit board material.

http://www.rocketroberts.com/cm4251/cm4251.htm

Hey: Craig.

Thats great bud.

I thought about trying to replicate these antennas as well.
I figure it would not be real hard as long as I could find aluminum
cheap enough then it would just be figuring out how to bend it.
A co-worker suggested using a peice of 2x6 lumber an lying one
of the elements on the broad side of the wood tracing out the
radius then cut the wood with a jigsaw. Once you have the curve
in the wood it could be used as a template to bend new aluminum
stock. I'm not sure if heating the aluminum with a torch would be
better or if there is an easier way to get the curve.
I guess I could go to metal supermarkets & they would cut & bend
the stock for me.
I'm looking around for cheap stock, maybe the stuff they use to support
tents or car ports as long as it's aluminum. If weight was not an issue copper
pipe would be a great conductor.

Let me know how your antenna project works.


Take care.
Bruce

Sorry Bruce! I forgot to mention that I was using a mesh dish.
I originally started out w/ a 10 footer, but am going to try a 7' first, before hoisting up a 10.

They can usually be had very cheap, or for scrap prices.

I promised around (post 257) to weigh the mesh dish I saved from the scrap yard. I finally made it to the farm (during daylight and with a scale and tape).
It is an 8 foot diameter dish and each quarter weighs 5 Lbs. so the whole reflector would be 20 Lbs. That doesn't include the support arm for the LNB.

Craig & Bruce : depending on the channel in question you may be better off over all with the lpda feed. Due to it's very stable SWR, flat gain, and high tolerance to focal point error.

It may also have a smaller shadow (on the dish) I'm not sure why but the lpda beat all the other feeds I tried.

It looks pretty easy to build I'll find out how easy soon. I got some square brass tube the other day.

Just a suggestion from what the models say.

Craig & Bruce : depending on the channel in question you may be better off over all with the lpda feed. Due to it's very stable SWR, flat gain, and high tolerance to focal point error.
Yep, as long as the LPDA is kept to about a foot or so in length. Using a longer LPDA to get another db or so in gain on the feed doesnt translate to better overall gain.

I got some square brass tube the other day.

I got some 1/4" OD round copper tube that the ID fits the stripped down unbraided coax perfectly. :p The trouble is, its going to be more difficult to solder on the elements than a square tube. Ill have to build a jig to hold it and file grooves in the tube. Making spacers is also trickier :(

Would I be able to remove some of the panels for weight, wind and ice?
If so, how many degrees should be kept?

Thank you!

Are you asking about your 7 or 10 ft dish ? How big are the panels ?

If you could post some pictures of your dishes, maybe we could give you some ideas on how to reduce their wind load for uhf frequencies. :D

http://www.digitalhome.ca/forum/picture.php?albumid=921&This is the ten footer.
I have the 7' mounted on the garage, just not completely bracketed yet.


The hub on this is 13.5"d and I have already removed 10 of the sections, leaving about 160 deg of surface.
The head, is a copy of a 4251 w/ dimensions taken from the tribute page.

I am bolting two 1.5" pipes to the hub, so I can mount it to my tower, via drum cage clamps.

http://www.digitalhome.ca/forum/picture.php?albumid=921&pictureid=5572

I have already removed 10 of the sections, leaving about 160 deg of surface.

So its only going to be a partial parabola then ?

A picture could describe a thousand words. :D

Yes, just a partial, is what I'm hoping I can get buy with. Is the entire radius even necessary?


Is there something else you'd like me to post a pic of?
It sounds like it's not showing up for you?
In my album here, it says it's still being moderated, if that means anything.