Baluns (Brands, Designs, Losses, DIY Loops, etc.)

[300ohm]

Here are the model results for a halfwave of 75-ohm coax. The Y-axis is mismatch loss in dB and the X-axis is frequency in MHz.

Is that RG6 or RG59 in that model ?

[mlord]

Is that RG6 or RG59 in that model ?

It shouldn't matter which, as both are 75ohms. Where it matters is when actually cutting to length for a real (non-model) build, where one has to take the VF into account.

Cheers

[mlord]

Originally Posted by Brian

I decided to build a balun to validate the model.

Excellent work Brian! If this holds up to scrutiny (it ought to), then those little loop-baluns I've been using are probably better than I thought they might be! (despite the sloppy construction of some of them).

Cheers! (and thanks!)

[k6sti]

I didn't mention one other thing. What you really want is a current balun, one that forces equal currents in the balanced load or feedline. A halfwave coax balun and a ferrite balun are voltage baluns that force equal voltages. This is OK as long as the surge impedances of the two conductors of the balanced side are equal. Usually they will be. But if you're feeding twin-lead, for example, and one side is closer to metal than the other, the impedances will differ. In this case a voltage balun will actually cause line imbalance by equalizing the voltages with respect to ground, which the unbalanced impedances promptly turn into unbalanced line currents. It's the currents that count since that's what causes radiation in the transmit case or signal pickup in the receive case.

The unbalanced currents will induce a current on the outside of the coax feedline past the voltage balun. If you can kill that current then the parallel line currents will rebalance themselves. Perhaps the best way to do that at UHF is with a sleeve balun. This is a quarterwave cylinder placed over the feedline. It is open at the feedpoint and shorted to the feedline shield at the far end. It acts like a high impedance to current on the outside of the coax. You can make one with solid metal or with flexible copper braid stretched over the coax vinyl. Google "sleeve balun" for more info. Ideally, you should follow your 4:1 voltage balun with a current balun to do the whole job.

I neglected to mention that the halfwave coax balun model used a lossless transmission line. I checked typical loss values for small coax at UHF-TV frequencies and with the short lengths required, the losses are negligible (a few hundredths of a dB). If you want to use 75-ohm coax, a friend tells me that Belden 1855A is very nice. It is subminiature and bends easily. Its minimum bend radius spec is 1.5". The specs for the Belden RG-59 and RG-6 I checked were 2.5" and 3" respectively. I didn't check RG-62, but I probably violated its minimum bend spec when I made the test balun. I'm not sure what can happen if you don't meet the spec, but years ago I heard horror stories about the center conductor slowly migrating through certain kinds of dielectric at bends, eventually shorting to the shield.

Brian

[mlord]

I'm not sure what can happen if you don't meet the spec

I think the main concern is that the foil sheathing will crack when the bend radius limit is exceeded. Not a problem with most RG59, though! (no foil in that..)

[k6sti]

I found a better load resistor, 1/8 or 1/10 watt, that measured about 320 ohms and should have lower shunt capacitance than the half-watt I was using before. Using the new resistor the balun center frequency rose to 610 MHz and showed a huge return loss dip (more than the splitter shows with a 75-ohm load, so there's some reactance compensation going on). Mismatch losses 100 MHz away were 0.09 and 0.14 dB. The model says they should be 0.17 dB. 115 MHz away, corresponding to the width of the UHF-TV band (counting 698 MHz as 700 MHz), I measured mismatch losses of 0.11 and 0.22 dB. The model says those numbers should be 0.24 dB. This time I averaged both an open and a short as the reference for the return loss measurements. I'm not sure why the physical balun is now beating the model with the new load resistor, but the numbers aren't that far off in absolute terms. In any event, the calculated and measured losses are small, much lower than those of any ferrite balun.

Brian

[300ohm]

I'm not sure what can happen if you don't meet the spec, but years ago I heard horror stories about the center conductor slowly migrating through certain kinds of dielectric at bends, eventually shorting to the shield.

Thats probably using high power transmitters. The voltages and currents in TV receiving antennas are miniscule. But its still a good idea not to force the coax into tight bends, expansion and contraction in it due to temperature changes could be hard on it over time.

[mlord]

One possible weakness in the balun simulations, may be the fact that antennas do not provide constant impedance across the band.

Rather, they vary quite a bit from channel to channel. Eg. 100ohm on some channels, 500ohm on others, 300ohm on a precious few.

I wonder how our beloved coax-loop compares with that in mind ?

[k6sti]

Originally Posted by mlord

One possible weakness in the balun simulations, may be the fact that antennas do not provide constant impedance across the band.

Rather, they vary quite a bit from channel to channel. Eg. 100ohm on some channels, 500ohm on others, 300ohm on a precious few.

I wonder how our beloved coax-loop compares with that in mind ?

The load impedance doesn't affect the balancing function. It just affects SWR and the resulting mismatch loss in the normal way. All a real balun or a simulation of it does is force equal voltages with respect to ground at the balanced terminals. In so doing it changes the impedance by a factor of four. Any impedance variation at the balanced terminals is reflected back to the unbalanced terminal, reduced by a factor of four in absolute terms and not reduced in relative terms. Voltage baluns are equivalent to an RF transformer with a 1:2 turns ratio and a grounded center tap on the secondary. Adding a current balun is equivalent to removing the center tap.

Brian

[k6sti]

Originally Posted by 300ohm

Thats probably using high power transmitters. The voltages and currents in TV receiving antennas are miniscule. But its still a good idea not to force the coax into tight bends, expansion and contraction in it due to temperature changes could be hard on it over time.

From a Belden technical paper describing type 9913F7 cable:

Besides the dissipation factor, old-style chemically-foamed cable typically suffers from another problem: center conductor migration. Because the elasticity of the conductor material and the polyethylene is quite different, over time the conductor tends to creep or cold flow through the polyethylene. Conductor migration is accelerated when the cable is installed with a tight bend radius, continuously flexed, or exposed to elevated temperatures. This migration would radically alter the impedance at that point, causing serious VSWR, and dramatically reduce the voltage-breakdown ratings. The poor mechanical performance can be attributed to the materials used which were typically low or medium density polyethylene.

Some cable made by Times Microwave allows a very small bend radius. For example, the minimum installation bend radius spec for their 75-ohm, quarter-inch LMR-240-75 is 0.75". This is much lower than the Belden specs I quoted. With this stuff you could make really tiny halfwave voltage baluns and coiled-coax current baluns. The latter is an alternative to a sleeve balun. Common coax is too inflexible to make a coiled-coax balun at UHF-TV frequencies, but it should be possible with the Times Microwave cable. They are much simpler than sleeve baluns. An example for 88-108 MHz is here:

http://ham-radio.com/k6sti/balun.htm

Below is an index of the Times Microwave offerings. There are some interesting technical papers at the end of the page.

http://www.timesmicrowave.com/wireless/index.shtml

Brian

[300ohm]

Because the elasticity of the conductor material and the polyethylene is quite different

Thats an older paper I guess. The RG6 Ive been using has the foamy core and Im not sure if its technically polyethylene anymore. But that may make it migrate even more, heh.

[stampeder]

There was a line of braided-core RG6 a long time ago that allowed tight bend radii, but I don't remember the brand and I had doubts about its longevity. Times Microwave is using a solid core.

[d510d180]

Hi,

question #1 Can a 4:1 current balun be exchanged with a common 4:1 voltage balun (would that not work with typical antennas like DB4, SBGH, folder dipole?)


question #2 I can't find the schematics on the web that would be matching the wiring I see in two common 4:1 baluns I openeds, and I paid attention to the core winding directions too. I actually destroyed it on purpose in the process of making sure I took the wiring right. Meanwhile, in the wiring I see, there is a 4th coil which is surprisingly shorted. I would appreciate if someone open a balun and confirm/link a schematic. If you can also explain how the magnetic fields flow in the usual binocular ferrite core in them, that may help a bit too.


question #3 Now, I'm still trying to find a 1:1 balun for a DIY project about making a clearstream c2 clone(2 parallel 160 ohms loop antennas is 80 ohms, requiring a 1:1 balun to 75ohms coax). For this balun, Would I need a current or a voltage one?


question #4 It looks like I won't find a 1:1 balun in common stores. For the moment I would have to recuperate the the binocular ferrite core of my destroyed balun and make a 1:1 balun myself once I get the plans... Schematic anyone?

Thanks.

[holl_ands]

Earlier today, I cited references answering these same questions in a parallel dimension:
http://www.avsforum.com/avs-vb/showt...php?p=15679199

I think the Winegard CB-2130 (VHF only, comes with YA-1713) and CB-8269 (VHF/UHF)
are 1:1 Baluns:
http://search.solidsignal.com/?q=win...m&new_search=1

[d510d180]

I have looked at all links and pdf, twice now (I know how to use google too ). There is no association of any given circuit to the common 4:1 voltage balun we see. Meanwhile they don't answer my questions regarding whether or not the current and voltage balun are interchangeable, and which 1:1 balun I should use for a double loop antenna. At this point I would like a human answer, rather than digging endlessly tons of doc. That's why I came to the forum last