Effects of mounting points on multi bay bowtie antennas

[mclapp]

Over the weekend I did some testing and one of the antennas I tested was a 9 1/2" x 9" 4 bay mounted on a wood frame.

The spacer blocks are about 1" wide so the phase lines are in contact with it for nearly an inch. The elements and phase lines were made of #12 copper wires. When I tested this antenna it was about 1 - 3 db down from the other similar antennas tested.


I removed the phase lines and elements from the wooden frame and attached them to a ½” PVC frame and retested the antenna. It gained .5 – 2db overall and the gain peak moved up nearly 50 MHz. I don’t know for sure if it was due to the wood or the larger contact area the elements and phase lines had with the wood compared to the PVC or both but it definitely made a difference.

I would think the change was mainly due to the larger contact area.

The area where the elements attach to the phase lines is considered a high voltage area relatively speaking and is the most sensitive to contact with other objects. The high voltage area changes with frequency but is mainly centered around the attachment area on this style of antenna. The best solution would be to avoid any contact with anything in this area but that can make construction more difficult.

I also noticed that most of the antennas tested had a peak gain that was somewhat lower in frequency than how they computer model which would fall in line with the loading effect of the attachment points.

The 9 ¾” 4 bay shown in the chart has custom plastic spacer blocks to minimize the contact area and that one performs the best overall and closest to the model predictions.



Also in this test I trimmed the whiskers of the 9 1/2" x 9" to 9"x9" and tested a gen 1 GH (44mm feed gap) and a classic 6 1/4" closed end single bowtie.

A series of 4 folded dipoles ranging from 6 1/2" to 11 1/2" were used as reference to get dbd readings.

The tests were done by mounting a transmitting dipole and the test antennas about 7 meters above ground and spaced just far enough apart to minimize ground reflections. A B&K 2630 spectrum analyzer was used for the measurements and as the signal generator.

I did alot of other tests this weekend as well and I will post that data later but I thought this was one of the most important of all the stuff I learned or confirmed.

[mclapp]

Here are some results of the antenna tests I've been working on. The tests were carried out just like the ones posted earlier but this time using different antennas. Here is a short description of the antennas tested. The 9 3/4" x 9 1/2" uses a 36"h x 40"w curved reflector made of 2"x4" wire fence , the 10" PVC uses #18 galvanized wire strung back and forth between the PVC frame, and the winegard style uses a 40"h x 48"w curved reflector made of 2" x 4" wire fence.
The 12" x 11 3/4" 4 bay uses a 46"h x 44" wide curved reflector made from 1" x 2" wire fence and the Hoverman uses a split screen reflector made from 1/2" x 1/2" hardware cloth.

I tested some of the antennas with different reflector spacings to show what effect that has on different channels. The tests were done on UHF and VHF-HI.


The Hoverman was omitted from the VHF-HI plot because it had negative 11 - 16 db gain.

This last plot is of a DIY 8 bay using 10" whiskers and the effects of different feeders. The first test was feeding the antenna with 2 baluns and 2 equil lengths of rg6 to a combiner. The second test was using 34" of 450 ohm ladder line and attaching a balun in the middle. The 3rd test was attaching a CM spartan3 amp fed with a short length of 300 ohm from the 450 ohm ladder line to the amp, I then subtracted the advertised gain of 22db from the readings. The gain for the amp test may not be 100% accurate but is good for comparing the gain curve with the other tests. I also added 2 other antennas for comparison.

[300ohm]

I'd love to see the gain plot from 4nec2!!

Ill try to whip out a model of it in the next few days. Whats kind of interesting about the DB4 is the phasing lines are 45mm apart, yet the feedgap is 30mm.

Good drawing, but I still have some questions.
How far do the phasing lines extend above the whiskers on the top and the bottom ? And what is the length of the cross over X's (each diagonal) ? And where they cross over, the distance between the wires ? And the width of the whiskers at the point where the phasing lines go over them ? And see if you can give a good guess as to the AWG wire size of the whiskers, phasing lines, and reflector wires. (just take same known wires and compare)

[stampeder]

mclapp, when you say Hoverman in your posts and diagrams, are you referring to the designs from the original patents?

If you mean the new models designed here at this site they are Gray Hoverman antennas, or GH.

I'm not being finicky, I'm asking because the GH antenna designs are copyrighted and GPLv3 licensed so there is a legal reason for clarifying the name, as well as protecting the work of the GH contributors.

Thanks, and thanks also for the great tests and charts!

[mclapp]

I would say it leans more towards the GH version 1 but using a split screen reflector instead of collinear rods and slightly different measurments. I don't have the measurements of an antennacraft hoverman so I couldn't say for sure which it's closer to.

The one I built and tested has a 1 7/8" feed gap and 7 1/4" zig zag section and 5 5/8" stubs on the ends, it's constructed with #10 wire. The reflector spacing is about 4", I also tested it with a 3" spacing but didn't publish those findings as they weren't as good overall as the 4" spacings.

[DjiPi]

Originally Posted by mclapp

Here is a short description of the antennas tested. The 9 3/4" x 9 1/2" uses a 36"h x 40"w curved reflector made of 2"x4" wire fence

mclapp, thank you for your work! If I understand well, the 9¾ x 9½ you are talking about refers to whisker elements length x elements distance at anchor point? What is the feed gap between the "V"s used for your tests ?

[mclapp]

Originally Posted by DjiPi

mclapp, thank you for your work! If I understand well, the 9¾ x 9½ you are talking about refers to whisker elements length x elements distance at anchor point? What is the feed gap between the "V"s used for your tests ?

The spacing between the V attachment points on that antenna is 1 1/4". Here is a diagram of the phase line, the V shaped whiskers are 9 3/4" measured from the attachment point to the tip and the tips of the V are spread apart about 5".

[stampeder]

This thread has been split out of the original OTA Build-It-Yourself Antennas thread, which is now closed.

[Grand Audio]

Hello everyone. I am new to this forum and have some questions I haven't found the answer to yet.

I have went by these plans and built an antennae (9 3/4" 4 bay bowtie). I used 10 gauge solid copper wire for the whiskers and phase lines and soldered all contact points. I then used zip ties to mount to wood frame. Instead of the 300 ohm balun, can I just hook the RG-6 straight to the feed points? How do I go about adding a reflector to this? I see nothing about a reflector on the schematic.

Also, is it possible to hook this antennae up to a Terk TV44 clip-on antennae to get better reception?

Thanks in advance for any response.

[mlord]

Originally Posted by Grand Audio

Instead of the 300 ohm balun, can I just hook the RG-6 straight to the feed points?

You can, but it won't work as well as it might otherwise. The antenna has a nominal 300ohm impedance, whereas RG6 is 75ohm. That 4:1 mismatch could cost significant reception.

But.. if you add a simple 1/2 wavelength coax loop "balun" to the connection, that problem vanishes, and you end up with something even better than a store-bought balun!

I use coax-loop baluns on most of my antennas here, and really like the simplicity and lack of added noise (versus a store-bought balun).

More information near the bottom of this page, and some construction photos (of mine) are here. You can ignore the fancy box construction and just simply attach the wires and hold them with zip ties. EDIT: but you *must* protect the exposed coax ends from moisture somehow!!

For UHF channels, the length of that extra piece of coax should be between 7.0 and 8.5 inches. Long for the low channels, short for the higher channels.

For my own UHF antenna, I designed it to center on channel 24, plus or minus 10 channels. This meant I had to use a loop of RG6 that was 21.2cm in length (8-11/32"). If using RG59 coax instead, then the same loop needs to be shorter, at 18.7cm (about 7-11/32"). There's a calculator at the bottom of the first link above which will figure out those numbers for you.

Cheers

[Porky]

Mlord,

I was thinking of purchasing a more expensive balun version to replace my 'RadioShack' one. But will now build my own. I always read that these home-made baluns were better.

Thanks for the picks and the info !

btw....nice workshop

[mlord]

One caveat: the store-bought baluns perform okay over a wider range of frequencies than a 1/2W coax-loop balun does. Most store-bought baluns work over the entire UHF/VHF bands, but with losses from 1dB to 6dB, depending upon the channel (or frequency).

The coax loop balun should have no more than 0.5dB loss over the band of channels it was designed for (center frequency, plus or minus 10%, or about a 10-20 channel range on UHF).

It will still work for other frequencies, but will have definite "null points" where the loss is higher than a store-bought balun.

Cheers

[300ohm]

Instead of the 300 ohm balun, can I just hook the RG-6 straight to the feed points?

The antenna is an unbalanced line whereas the coax is a balanced line. Hence comes the name the for the balun, BALanced to UNbalanced.


Also on the bowtie in general, Ive found that straight phasing line connections work better than looping the wire around the bolts/screws. On straight copper phasing lines, I use bent copper washers to hold them in place. If using different metals, consult a galvanic corrosion chart.

Also, the point at which the wires touch plastic should also be kept to a minimum. Wires resting on plastic is better than wires resting on wood.

[300ohm]

The antenna has a nominal 300ohm impedance, whereas RG6 is 75ohm. That 4:1 mismatch could cost significant reception.

Impedance on all broadband antennas varies with frequency. Thats why I dont like to get caught up in the impedance thing. The impedance is only normally going to be 300 ohms at just a few frequencies. The mismatch is shown by the increase in the SWR ratio, ie the signal bouncing up and down the coax. Also, even with careful modeling, the impedance of any particular actual antenna is going to depend a lot on the build quality as well. (And we also should take into account the impedance of the TV too where the build quality and testing of the inputs could vary.)

[Davepet]

The antenna is an unbalanced line whereas the coax is a balanced line. Hence comes the name the for the balun, BALanced to UNbalanced.

Ummm,isn't that backwards? I thought the antenna was balanced & the coax unbalanced?

Dave