... I'm thinking I should probably use the frequency of the weakest transmitter as my balun frequency.


For the coax balun, that is what I did and it seems to work well across the band (range from real channel 22 to 58 - my weak channel is 33). However, I didn't notice a lot of difference when I made "channel cut baluns" for 58 and 22 either.

... tvfool also seems to suggest my noise margin is *better* if I have the antenna lower. (12.1dB for 10' AGL, 10.1dB for 30' AGL) This seems very weird, but does it have anything to do with my primary reception being double-edge diffraction rather than LOS?


My TV Fool does the same thing, but I didn't have the energy to dismantle anything and put it near the ground to test it out. I'm curious how real-world testing of this would pan out in our situations, considering the conflicting "higher is better" thoughts.

Does anyone have any suggestions whcih would help improve the stability of reception?

Buy a 91XG and a Kitztech coax powered amp. Mount it as high as you can with a rotor.

Buy a 91XG and a Kitztech coax powered amp. Mount it as high as you can with a rotor.

I may end up doing that yet, but I'm pretty excited that I was able to watch not only CBC but also the Buffalo stations when the antenna was in my bedroom. Rotor's not something I want, I am not interested in looking at other transmitters at this time.

Any reasons for the Kitztech vs the CM-7777? The specs on the CM seem pretty damn good (26dB gain on UHF with 2.0dB noise figure). I'm definitely a newbie at this, but I do have a pretty solid electrical background (I design industrial embedded systems). One thing I do know though, is that RF is a very strange mistress, indeed. :-)

(edit after reading up on Kitztech) Ahh. 1.0dB noise figure across the band, so a not insignificant advantage! Not quite as high a gain, but the noise figure is more important, as I understand things.)

Ok, ANT-2109 seems to have a craptastic balun. I bought a $2 no-name balun from Orion on Lancaster in Kitchener and swapped the PC board balun found in the folded dipole element for it.

Sitting in my upstairs bedroom I was only pulling in 11.1 (CHCH out of Stoney Creek, the station I can almost get using nothing but a paper clip for an antenna). Swapping out the baluns instantly pulled in CBC HD (5.1). Putting the antenna up in the attic and doing a half-assed aiming job on it got CBLT (5.1), CFTO (9.1), CHCH (11.1) and something up on 36.1 I can't quite lock on.

Not bad for $2 and 15 minutes of my time. The balun's 300 ohm leads are not proper twinlead either; I did a quick job of trying to keep the leads approximately the right distance apart, but I'm sure they're twisted and causing an impedance mismatch up there.

I also bought an RG61 female end and a length of copper-braid RG6; I'll build the coax loop balun this weekend and give it a go as well.

If anyone's interested, here are the pics of the balun that came with the ANT-2109. It's crap, and could be a major reason why the antenna performs so poorly. It doesn't appear that the board was even washed off.

http://www.mixdown.ca/~andrew/ant2109balun/front.jpg
http://www.mixdown.ca/~andrew/ant2109balun/back.jpg

Quote:
Originally Posted by akohlsmith
... tvfool also seems to suggest my noise margin is *better* if I have the antenna lower. (12.1dB for 10' AGL, 10.1dB for 30' AGL) This seems very weird, but does it have anything to do with my primary reception being double-edge diffraction rather than LOS?

My TV Fool does the same thing, but I didn't have the energy to dismantle anything and put it near the ground to test it out. I'm curious how real-world testing of this would pan out in our situations, considering the conflicting "higher is better" thoughts.

Same in my case, I can confirm TVFool seems to have been bang-on, maximum NM's were at 8-10 ft. The NMs decrease by 1 dB per ft increase in heigth. I have tested this repeatedly, raising the antenna 2 INCHES at a time. Most consistent reception was in fact at 8 ft, but this is an impractical height to actually use on a tiny property. After 14 ft, at least up to 35 ft, I gradually LOSE all reception.:eek:

If anyone's interested, here are the pics of the balun that came with the ANT-2109. It's crap, and could be a major reason why the antenna performs so poorly
The circuit traces there look like a rather poorly constructed loop balun. I very much expect your home-made coax loop balun will outperform it across the band.

Cheers

This looks very much like the balun from the 91XG. I had earlier reported that my DTVPal signal went from something like (I'm too lazy to look up my post) ~65-75 to ~73-75 when I made a coax-loop balun. So, my max signal didn't really go up, but it was far more consistent.

I had been wondering if this was real or some type of coincidence. I thought about swapping the PC balun back in and out, but with this info, laziness again prevails, and I'll just stick with the coax-loop one.

Same in my case, I can confirm TVFool seems to have been bang-on, maximum NM's were at 8-10 ft. The NMs decrease by 1 dB per ft increase in heigth. I have tested this repeatedly, raising the antenna 2 INCHES at a time. Most consistent reception was in fact at 8 ft, but this is an impractical height to actually use on a tiny property. After 14 ft, at least up to 35 ft, I gradually LOSE all reception.

Very weird. I played with the AGL and it looks like picking up the weakest station on the CN tower for me occurs at 5' (2dB NM). I might end up mounting this antenna in the garage instead of the attic.

Any RF gurus care to comment? I'm curious as to why this would be? I also realise that the NM figures do not take into account other buildings/trees/etc. so it may in fact be incorrect when these factors are taken into account.

The circuit traces there look like a rather poorly constructed loop balun. I very much expect your home-made coax loop balun will outperform it across the band.

I'll be constructing the loop this weekend if I can find a few hours of time. I'll definitely report back here. CBC still came in strong this morning with the $2 balun, which is a first for this antenna. :-)

just to be clear:
the philips outdoor baluns are better than the winegard tv-2900 which was the flavor of the month last month ?


If you look at the two they appear to be identical including the boot and fitting that comes with it. I also bought some baluns from skywalker communications which came as "generic" and all three have the same case..stampings etc. I wonder if all three are made at the same Chinese plant???

Hmmm. I have both the Winegard TV-2900 and Magnavox (Philips) baluns. Mine are quite different.

I'm about to make my coax loop balun. I have some "75 OHM RG-59/U COAXIAL CABLE" but the insulator is most definitely NOT foam; it looks and feels like hard nylon.

The calculator on K7MEM's site has a checkbox for RG59 and RG59U (foam) -- I definely don't have and can't obtain the datasheet for this specific piece of wire. Out of curiosity, what should I consider it?

I say say out of curiosity since it's the difference of 14 or 16.7cm for 707MHz, the higher end of the band. Since mlord indicates that loss across the band is on the order of 0.5dB I am going to try to tune for the weakest station (CITY-DT) and let the rest of the chips fall where they may. As you come down in frequency in the band, the loop length gets longer, so if I use the RG59/U (foam) velocity factor and I'm wrong, I end up tuning for a little lower in the band... not a big deal. I'm just curious to know what this cable is, since as far as I can tell it's definitely not foam, even though it's RG59/U.

Mil-Specs for RG-59 mandate the use of that hard, slippery, plastic stuff....NOT FOAM.

If it walks like a duck and quacks like a duck....then it's probably RG-59....

I had better luck with RG-59 Nylon then foam for my homemade baluns.

I did check off the right kind of rg-59 on K7MEM website.

Maybe it's just a fluke.

Still haven't built the 4:1 coax loop voltage balun, but have gone back and read over the discussion on the current baluns. I think I have successfully confused myself. :-)

a 4:1 coax loop balun connects the 300 ohm active element on the yagi and presents an unbalanced (voltage) to a 75 ohm coax feedline. So far so good. Where I start getting lost is how I then use a 1:1 current balun (copper/brass tubing apparently) to make a 75 ohm to 75 ohm balun... unbalanced back to balanced? Can you have a balanced voltage and unbalanced current? whrrrr...

Where I start getting lost is how I then use a 1:1 current balun (copper/brass tubing apparently) to make a 75 ohm to 75 ohm balun... unbalanced back to balanced? Can you have a balanced voltage and unbalanced current? whrrrr...
No.

The 4:1 coax-loop will give unbalanced voltage at 75ohms.
A 1:1 current balun after that will give unbalanced current as well.
The two together make as "perfect" a transformation as possible.

Cheers

The 4:1 coax-loop will give unbalanced voltage at 75ohms.
A 1:1 current balun after that will give unbalanced current as well.
The two together make as "perfect" a transformation as possible.

Hmm okay. Where did the 150 ohm impedance coax requirement come from? I know that 150 is twice 75, but where's the factor of two coming from? I am definitely feeling a facepalm moment coming up, but I'm just not seeing it yet...

edit: I think I see it now; the current balun described on page 15 shows the two quarter-wavelength runs with a characteristic impedance twice that of the desired feedline impedance. Facepalm in 3..2..1..

edit 2: facepalm averted. The diagram at the bottom of page 15 shows a 4:1 current balun, where I'd be trying to make a 1:1 current balun. Would that not mean I'd want a characteristic impedance of 35 ohms on each the 1/4wave segments? Can I cheat and use two equal segments of 75 ohm coax in parallel (shield-to-shield, center-to-center) for each of the legs on the unbalanced side?

Now I'm more confused. I was thinking that you would need either the voltage balun or the current balun - one or the other - not both.

I've gotten as far as finding 18-gauge solid insulated wire and the copper tubes cut to length (my 2 tubes are each 4-1/4" long). I'm dragging my feet on how to hold the wire inside the tubes (I think that the yogurt cut-outs will be too loose) - that, and the fact that the loops are working SO WELL. Maybe I'll use the yogurt holders and squirt some foam in there for a more robust assembly.

Also, just today, I received my order of copper shielded RG-6. Maybe that will get me more motivated. All my coax-loops had been made with copper shielded RG-59, that I've found to work great, but I've recently noticed that the impedance is only 71 Ohm. I'm guessing that doesn't make much difference, but perhaps every little-bit helps !

kkoceski,

All my coax-loops had been made with copper shielded RG-59, that I've found to work great,

im very curious how you quantify this, the ONLY way for us to know for sure if they work better than commercial balun is with an A/B test - have you done this...

so far in my testing results are very mixed, variable, and INCONCLUSIVE

Now I'm more confused. I was thinking that you would need either the voltage balun or the current balun - one or the other - not both.

You can certainly get by with only one; in fact that's what most baluns are doing. I'm going for what mlord calls the "ultimate" in matching: matching both current and voltage. I'm going after the deep fringe (hitting a 2.2kW transmitter (CITY-DT) from 60 miles away) so I need every dB I can get. :-)