The AM Forum

A quick question for the group:

I presently feed the 50 ohm input to my homebrew balanced L-network antenna tuner with a choke balun wound with about 27 feet of RG-213 on a 4.0" O.D. piece of Schedule 40 PVC pipe. It seems to work well, with the bulk of my operating on 75 meters, but I was wondering if there would be any advantage or disadvantage to replacing this existing coaxially-wound balun with a conventional current balun wound on a toroidal core?

Array Solutions of Sunnyvale, TX offers a Model W1JR-50-1 Toroidal 5 KW/10 KW PEP 50 ohm/50 ohm balun that covers the frequency range of 1.8 to 30 Mhz., so this is one possible option if the toroidal current balun is considerered a better solution in this application.

Thanks & 73,

Bruce, W2XR

Bruce,

I personally think that the toroidal balun is a better isolator than a simple RF choke comprised of a coil of coax.  The optimum amount of coils is about four turns for 10M, while for 75M it's about 12 turns - all on a 4" pvc pipe. So, on 20M-10M the reactance degrades if using 12T or more.  It's even worse if you wanna cover 160-10M.

In contrast, the toroidal unun or balun is pretty consistent across this spectrum if swr is any indication. Hook a 50 ohm resistor to the sec and sweep the pri with the MFJ-259B and you'll see.

I also used some of WX0T's unun's here to go from 50 to 75 ohms. (50 ohm Coax to 75 ohm hardline)   

After paying $100 a pop for a few I decided to wind my own using W2FMY's book and parts. Do a search and wind your own for a few bux cost of the toroid and wire. I use Teflon wire on mine.

73,
T

Hi Tom,

Thanks for the reply!

You indicated that the isolation with the toroidal current balun would be better than that of the coaxial choke balun. Why would the isolation be an important or more optimum consideration here? By isolation, I am assuming that you are referring to the isolation between the input to the balanced-L tuning network, and the output of the transmitter.

Is this correct?

BTW, I have done some research on winding my own toroidal current balun, and you are right; I would be better served to wind my own.

73,

Bruce

Hi Bruce,

What I mean by isolation is the balun's ability to keep the unbalanced components from the balanced components... ie, like an effective transformer's primary referenced to ground and the sec floating with a balanced dipole, etc. The output balanced currents are not skewed and phase distorted by the input's unbalanced currents.

An RF choke as created by a coil of coax will vary in impedance depending on it's number of turns. Too many turns on the higher bands and it's performance degrades markedly due to intercapacitance of the turns, etc.. But you need the many turns to cover 160 and 75M, etc.  There was a thread here where I posted the freq vs: turns impedance. It shows the problem clearly when using it on a broad spectrum like 160-10M. (The idea with the RF choke is to stop unbalanced RF currents flowing on the outside shield, as you know.)  Many guys use the proper permability ferrite beads over the coax too, to do the same job.

But I figure, why not start with no unbalance in the first place by using an unbal pri to bal sec RF transformer? Plus, the toroidal balun is pretty flat (broadband) across the whole 160-10M spectrum. Just get the correct core permability and you're all set. The downside is they are prone to permanent saturation damage from lightning pulses. Up on the antenna it's a problem. But in the shack you will probably be OK cuz the ant usually gets disconnected during stroms. You will know your ferrite has a problem if the swr is high... (the core cannot pass much power)

Don't get me wrong...a coax  RF choke works FB for semi-mono band use. I use them on many of my antennas here at the feedpoints except for some 20 ohm Yagis where I need a 4:1 coaxial balun to interface a delta match to give 50 ohms.. 

As a precaution...  Make sure you wind it like a toroid is supposed to be. If you do it like a conventional pri to sec power transformer, the coupling is poor and you will see the core heat up, even though the match may be good. Just follow the instructions and it will work FB..

73,
T

Hi Tom,

Thanks for clarifying this isolation issue for me. Yes, I completely understand the point you are making here, and I understand the concept and need for the isolation you have described.

But out of curiousity, why would the toroidal current balun offer better isolation vs. the coaxial RF choke current balun? And if the choke balun and the toroidal balun were both to be used on one band only, say 75M, and the choke balun was optimized for that band, would the isolation be essentially equivalent between the two devices?

Thanks & 73,

Bruce

I've tried both on my tuner and they both work. mine was built in '83 a few years before measures invented it. I actually have a 1:2 BB steo up transformer with 6 2 inch iron cores. 3 beside 3 wound whth 5 turns of #14 teflon quadfilar. 2 windings in parallel for primary and two in secies CT to ground for the secondary. Warms up silghtly after the 4Cx3000A oln buzzard qro.

One question Bruce, why do you think you need a better/different balun?


If you take a look at the impedance to common mode currents for a ferrite based balun and a non-ferrite based balun, you see the impedances for the ferrite based balun will almost always be higher, especially at the lower frequencies.

So, do you need the higher impedance/better isolation?

I tried the choke balun last winter and it worked just as well as the BB transformer.

Bruce,

Oh, OK - I thought you were trying to use the coaxial balun for multiband use. I see you did say your bulk of operating was on 75M.

So for mono 75M, it really won't make any difference in the real world between the two. Especially in the shack where size does not matter, lightning is not an issue and the freq is single band.

I don't have an explanation of why the toroidal balun would be theoretically better, other than what I've already said - just what's been told to me by some of the guys in the industry who make them, like Jay at Array Solutions, etc.   I've used toroidal baluns up to 6M and they always perfomed perfectly, when designed right. Same for optimized coaxial RF chokes...

I'll do some more reading and let ya know what I find... ;)

T

 a good Lightning hit will explode the cores. Choke balun might hang in if it is a short high frequency pulse. Waveform 4 will fry both configurations. Coax in cheaper than big cores if you are a bean counter.
The tuning was exactly the same with both configurations in my tuner.
I was surprised since one was a 1:2 step up and the other was 1:1. It must have just changed the Q of the network.

If it's for monoband use, why mess with the wideband baluns, choke, core, coax, etc? Seems to me you could just built a simple link and be done with it.

yup.....

Because you need a balanced signal to drive the two L networks out of phase.

Make the link balanced.

The dual L networks need drive out of phase. It would take 2 primary links. then what do you do with the inputs of the l networks? I suppose you could tie them to each other. I think that TMC tuner on epay last week did that.

A center-tapped winding wouldn't work?

it sure will but you change the name to a link coupled tuner.

Sort of. But it doesn't have to be a tuner, just a small link to to the balanced to unbalanced conversion. They're often used at dipole feedpoints for this and broadbanding purposes.

Yup agree. I decided to connect the center tap of my BB transformer secondary to ground for a couple reasons. First to force balance referenced to ground and second to make the antenna see a DC return for static close induced lightning. Still the 10 KW caps will flash over when there is a close one. I would think you could do the same thing with a link coupled tuner. I went a step further by using a pair of 300 pf 10 kv cardwells in series. Cases connected to ground for the output cap. This gives me another point to force balance.
I've tried floating the cases but the tuner seems happier with the cases grounded.
Again last winter I removed the transformer and did a coax balun and it worked fine.
There are many ways to do the deed usually linked to the junk box stock room.

Low pass tuna design, safety choke in amplifrier, antenna at DC ground. Rube Goldberg started out by making antenna tuners.

Guys,

I made the plunge and changed out the RG-213 coaxial choke input balun last night before 75/80M bandwarming activities; it was replaced by a toroidal 1:1 current balun that is rated at 5 KW CW/10 KW PEP. Although the original coaxial choke balun worked well, the toroidal balun allowed me to maintain a unity VSWR all the way down to 3.6 Mhz, and I noticed less RF in the shack, indicating better isolation between the xmtr output and the input to the balanced L-network antenna tuner, and possibly less open wire feedline radiation due to better balance  conditions. The RF  in the shack was never a problem, but this was a nice benefit just the same.

With the coaxial choke balun, the VSWR would gracefully rise to 1.4:1 at 3.6 Mhz., although this was of course,  an acceptable mismatch.

With this balun change, the W2XR homebrew antenna tuner now covers the entire frequency range of 1.8 to 30 Mhz..

Thanks for all of the input!!!!

Best 73,

Bruce

The input L and output C settings did not vary much at all when the new toroidal current balun was installed.

Bruce,
What antenna and feed line are you using? Also what are your L and C values in the tuner. My 160 meter dipole is now a Vee with shorter feed line and 62 feet of 450 on a 250 foot Vee is a problem length for my tuner. I think I have to make the feed line shorter. I lost about 50 feet when the tree yanked it down. Still have to drop 2 trees before I can put it back up.
I'm doing a 1:2 step up transformer with CT of the secondary grounded. Coax balun also has a problem. When the feed line was 110 feet I had no problems.
Odd thing is the KW match box works on 75 m. fc

Hi Frank,

The antenna is a 1/2 wave dipole cut for the 75M band up 45 feet and fed with 600 ohm open wire feeders. The overall length of the antenna is 122 feet.

The balanced L-network I'm using to tune this antenna consists of two high-power 22 uH roller inductors synchronized by a 3/4" wide cogged timing belt, and the output shunt C is a 1000 pf @ 14 Kv vaccum variable. The balun I described in the previous post is a toroidal  1:1 50 ohm current device and is located directly at the input to the network; it is rated at 5 KW CW, 10 KW PEP.

The network is located about 20 feet away from the xmtr, and is fed with RG-213. PITA to retune, now that we'll probably be QSYing frequently with the new 75M frequency privileges. I do have turns counters on the L and C, so this will makes resettability easy, but I have to walk across the basement to do this.

I built the entire assembly on a 32" X 28" piece of 3/4" thk MDF (large footprint necessitated by the large size of the components utilized in the design, and component spacing to maintain high Q) and there are no metallic housing components, etc.; this was done to maintain the Q of the tuner to as high as possible. Everything is right out in the open.

Hope this helps!

Would have like to worked you last night; I was on 3725 and later 3715 with WA2PJP and KA1KAQ for quite some time, until about 2:15 AM EST.

73,

Bruce

cool,
I also have a pair of 22 UH HP inductors with bread slicers. On 1945 I use 22 uH and about 225 uf.  My antenna is 125 foot on a leg Vee at 60 degrees with 62 feet of #10 spaced 4 inches(about 400 ohms). All I have is 250 uf max or 350 uf max depending on padder. Maybe I need more C? What is the inductance when you are tuned up on 75. I should be about double that on 160.
My old set up was 110 foot feed line and 250 foot dipole. It tuned up a lot easier.
I think the optimum feed line length would be 55 feet. 
I crapped out at 1:00 and never got above 3685.

Frank,

For operation on 75/80M, which is the only band I currently operate on HF, the roller inductors are set for approx. mid-position, so this is about  10-12 uH.

The output shunt C is set for around 500 pf.

The length of the 600 ohm open wire feedline feeding the 1/2 wave flat-top is 66 feet, or a little more than 1/4 wavelength on 75/80M.

Hope this helps!

73,

Bruce

TNX Bruce,
Sounds like I need a lot more C.  On 160 I used the whole inductor with  less C than you had on 75. I'm running lower Z feedline too. On 160 I had just a bit
less than double your lengths.
On 75 i got closest to a match with full mesh on the cap and padder added I should have tried the larger padder. fc