Thanks MLord,

I will try it out tonight.

I will post my results here compared to my RG-6 Coax-Balun and Channel Master Balun.

I will post pictures of the ugly setup also.


Intravino

I did swap out the original balun on my CM4221HD with an RCA balun, the signal strength drop anywhere from 10-15% (on my TV's signal meter) across all channels from 2.x to 66.x.

So the RCA balun is not very good...

The RG-59 was easy to do because I did not make a box for it.

I had Amphenol rg-59 for the Coax Balun.

I did not solder any 300 twin lead because the wires are short from the coax.

Results:

TivoHD tuner

CM balun: 0-32 % no lock and bouncing

Rg-6 balun: 15-47 % almost a lock with pixelization.

Rg-59: 10-37 % no lock and bouncing.


So my Rg-6 is king for now.

I've attached some 1600x1200 pictures also:

http://img132.imageshack.us/img132/9412/rg59coax.jpg

http://img132.imageshack.us/img132/7053/rg6coax.jpg

Well, your Rg-6 version is certainly prettier. :)

Bob Morton suggested bunching the three coax ends as close as possible together, and then soldering the ground braids into a tight bundle. Neither of your baluns do exactly that, but it does sound as if the Rg-6 one is working rather well!

The coax-loop is a rather forgiving design, making it perfect for us amateurs. :)

Cheers

I'm disappointed with RG-59 Results.

I will try to modify the Rg-59 Balum and I will put also a 300 twin lead at the end instead to see.

Maybe I could be lucky.

What kind of baluns would be worth getting then? I read the post about Tin Lee baluns quoted at $50 each. 2 baluns cost as much as the whole antenna:rolleyes:

Ive personally found the old made in Japan baluns, found at flea markets, to be of excellant quality, especially the ones from old Japanese game systems. Of course, some may be broken. And they would need to be weatherproofed, silicon sealant etc.

I'm disappointed with RG-59 Results.
Well, you shouldn't be disappointed with it! RG-6 is the better coax to use. I only use RG59 because it has a copper braid (for soldering), whereas the RG6 I have here has aluminium braid (not suitable for soldering).

Still, the two ends of the coax loop (whichever kind) should be bundled close together, for a 300-ohm spacing. That is supposedly the ideal way to do it. But wider probably doesn't hurt much -- my own early baluns look just like your RG-6 one, and they do seem to work rather well! ;)

Cheers

Since we are trying to reduce signal losses over a wide VHF/UHF band how about a balun design using:

1. http://www.minicircuits.com/pdfs/TCM4-19.pdf
2. Short piece of 300 ohm wire to hook up to the feedpoint
3. Custom length* of RG6 cable
4. One crimp/compression F - connector
5. a small weather resistant box
6. Some weather sealing goo


* Normally (in the different layout concepts I have been playing with at least) there is a relative short lead of RG6 running from the balun to either a preamp, diplexer or splitter mounted close to the antenna. The custom length referred to here is the same length that would be used to get to the first connection.

That way one save the F connector connections at the balun and its associated losses.

or do I miss something?

Since we are trying to reduce signal losses over a wide VHF/UHF band how about a balun design using:

1. http://www.minicircuits.com/pdfs/TCM4-19.pdf
...
or do I miss something?
"50-Ohm", at the top of the linked pdf.

And an "insertion loss" of 1.13-1.24dB for UHF, from just the transformer itself. Add the connections for the real world and that becomes 1.5-1.7dB, perhaps. Plus the impedance mismatch loss (another 0.5dB ?). Much higher (3X) the band-specific coax-loop baluns.

But yes, if you're not optimizing for any specific reception, you could use one of those, or a store-bought balun with a 75-ohm version already inside of it. The "half-dead" balun I had here (from the test at Bob's range) had something just about identical to this one inside of it: tiny PCB with a ferrite transformer on it.

Mmm.. I wonder why it has an "operating temperature" that only goes down to -20C ?? If true, then that wouldn't work up here.

Cheers

I now understand why some clone bowtie don't perform well - cheap balun!

"50-Ohm", at the top of the linked pdf. Yup, saw the 50 ohm - although the website has a 75 ohm search option I could not find a 75 ohm, but I am sure there must be something out there for 75 ohm . The goal is to have control and optimize what is happening outside the transformer - from there the idea to get rid of the Female - Male F- connector combo.

I wondered about the -20C too - could it be that the ferrite's EM characteristics change? - what will those low temperatures do to off the shelf baluns also using the same technology?

Since we are trying to reduce signal losses over a wide VHF/UHF band how aboutAnd so another crusader is hooked into the search for the Holy Grail of baluns... :D

Earlier in this thread is a Hybrid Balun design from a fellow in the UK who thought he'd found the perfect balun design, but nobody with testing capabilities seems to have followed up on it here: http://www.digitalhome.ca/forum/showpost.php?p=872629&postcount=51

I've seen a few HAMS recommend using four turns of RG-6 coax on a four inch mandrel as a 300 ohm - 75 ohm balun. I tried this suggestion, using a 2 inch length of 4 inch outside diameter pvc drain pipe as a mandrel. I tie-wrapped the assembly in four places to prevent the coax from unraveling on the mandrel . The device seems to work and it appears to have significantly lower loss than the el-cheapo baluns available at the hardware store.

The coax balun seems to give me about 30% more signal strength than the el-cheapo baluns -- this test occurred on the same day with about 30 minutes separating the two tests. (using signal strength meter on ATSC tuner, testing 3 VHF and 3 UHF channels).

:D :D

I've seen a few HAMS recommend using four turns of RG-6 coax on a four inch mandrel as a 300 ohm - 75 ohm balun.
That sounds more like a 1:1 current balun for VHF.

But heck, a TV antenna will work without any balun, too, mismatches and all. It's only an issue for the weak/distant stations.

I think the best/easiest balun design thus far here, is the 1/2 wavelength 4:1 coax-loop voltage balun, followed by a 1:1 sleeve current balun. As suggested much earlier in this very thread.

Simulations and measurements give the coax loop a definite two thumbs up, and adding the sleeve balun ought to make it about as good as is possible. For a single band, anyway.

Cheers

Humor aside :):p;)

I just modified the RG-59 balun and I found a very good one. It beats all of the CM, Philips, Addison, Rona and Japanese baluns for my quest to receive WVNY-DT.

I did want MLord suggested.

I've redone the balun, took of the tyr-raps and cut the length of the copper core wire of the coax to Bob Morton specs. I soldered a small piece of 300 twin lead.

I did a tighter loop also of the coax.

Now the good news:

Specifications:

Amphenol 03554 rg-59 coax.

Tuned for CH13 VHF-Hi

TivoHD tuner (microtune chip)

The results :

CM balun: Bouncing from 0- 31 %, no picture, no lock.

RG-6 balun: Bouncing from 10-43 %, sometines a pciture, heavy pixelization.

New modified RG-59 , Bouncing from 40-57 %, picture, lock, pixelization.

There you have it, my new king of balun is here.

http://www.digitalhome.ca/forum/picture.php?albumid=262&pictureid=1196

I reused my rg-6 plastic project box from Addison to put the rg-59 coax loop with RTV to block the old holes. When I will purchased with compression tool for snap/seal connectors, I will upgrade the rg-59 connector.


Thanks to Mlord for his advice.


Intravino

And so another crusader is hooked into the search for the Holy Grail of baluns...


I think mlord's experience at Bob's was kind of the final straw. Two things made me looking into a better balun. One is a broadband design to get the most out of a NAROD antenna and secondly, I would like to get repeatable performance. I am not dx-ing (yet) but in the short term I would like to join the outputs from two antennas facing in different directions. From what I have read one have to make sure that both antennas are the same and that perspective what is the use to measure the build to the nearest mm but ending up with two baluns that are the same on the outside but with 6 db difference in losses (not to mention the different electrical parameters causing this which in effect will make the antenans different).

I like the fact that minicircuits publish their specifications and I still have to read more about them (or other similar suppliers) but if any one of them is a 6 sigma (or even a 4.5 sigma) company then I know what manufacturer to buy from. So at this point of the game repeatable results 1st and then minimum losses 2nd, provided that the losses are say below 3dB. I also believe that a properly done solder joint will give a more stable performance over time than a F- connector connection, so in this aspect taking the F connector out of the balun will hit both goals.

I just modified the RG-59 balun and I found a very good one. It beats all of the CM, Philips, Addison, Rona and Japanese baluns for my quest to receive WVNY-DT.
...
I've redone the balun, took of the tyr-raps and cut the length of the copper core wire of the coax to Bob Morton specs. I soldered a small piece of 300 twin lead.
Fancy! And congratulations on the performance!

Now we just need to get a low-noise (pre-)amp for you to try.

Cheers

Intravino,

Great work! Do you have a picture of the insides of that box? What is the final length of your loop?

Earlier in this thread is a Hybrid Balun design from a fellow in the UK who thought he'd found the perfect balun design, but nobody with testing capabilities seems to have followed up on it here: http://www.digitalhome.ca/forum/show...9&postcount=51

Thanks for the link :)

I like his approach, (Andrew is from Cape Town, South Africa). I still have to get my head around the concept but it appears as if his design pretty much removes the antenna impedance sensitivity that you would get with a voltage balun - which is what you would need for a wide band antenna where we all now that the impedance is all over the place across the frequency band.

I have done a quick google and did not find any further work on this well written 2005 article.