From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Professional sound production (Rated C-class, Low-importance)
WikiProject icon This article is within the scope of WikiProject Professional sound production, a collaborative effort to improve the coverage of sound recording and reproduction on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 Low  This article has been rated as Low-importance on the project's importance scale.
WikiProject Electronics (Rated C-class, Low-importance)
WikiProject icon This article is part of WikiProject Electronics, an attempt to provide a standard approach to writing articles about electronics on Wikipedia. If you would like to participate, you can choose to edit the article attached to this page, or visit the project page, where you can join the project and see a list of open tasks. Leave messages at the project talk page
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 Low  This article has been rated as Low-importance on the project's importance scale.

Funny looking transformer[edit]

The transmission line choke balun depicted is not any sort of transformer. A transformer transforms. The common mode choke with coaxial windings around the ferrite works fine as a balun but isn't a transformer. Think about it for a second. The inner conductor of the coax only has a magnetic effect on the inner surface of the shield and vice versa. The current on the inner conductor MUST be exactly equal and opposite the current on the inner surface of the shield.

Any differences in current from the inner conductor to the braid MUST manifest itself in currents on the outer surface of the braid. By winding the coax into a coil around some ferite as depicted, these differeneces are eliminated (or at least greatly reduced) by a choking action thus balancing the signal on one end of the device regardless of what is happening on the other.Euc (talk) 12:50, 17 June 2012 (UTC)

The choke balun does act as a transmission line transformer. It is 1:1, meaning that the ratio of the windings is the same. There are two wires both wrapped around a core, like in any transformer. It just happens to be that the two wires are part of the same piece of coax, in the same insulating jacket. If it was only one wire, I'd agree with you that it's not a transformer. Then it would be an electromagnet or a plain inductor. --ssd (talk) 20:04, 30 June 2012 (UTC)

Rubbish. The 1:1 refers to impedance not the number of windings. A 4:1 impedance matching transformer uses windings in the ratio 2:1 NOT 4:1. Sure a piece of coax can be used as a transformer such as a piece used to drive both sides of a push-pull RF amplifier (terminated by the input impedance of the transistors), but in this case it isn't. Only the outer surface of the braid has any effect. The inner surface of the braid and the inner conductor simply behave as a longer piece of coax, nothing more. They have absolutely no magnetic effect on the outside world and only the same effect on each other as they normally would in a straight piece of coax. The effect is to choke currents on the outer surface. It is a common mode choke NOT a transformer.

In fact, an ordinary electromagnet is exactly what it is made only from the outer surface of the braid - one with a high impedance intended to choke currents on the outer surface of the braid. Exactly the same effect can be achieved by surrounding a straight piece of coax with magnetic material such as ferrite without winding it into any sort of coil.

Put another way, there are three conductors involved. There is no mutual inductance whatsoever between one turn of the inner conductor and the next. There can't be because any magnetic field generated by one turn is contained by the braid. There is no mutual inductance between one turn of the inner surface of the braid and the next. There can't be because any magnetic effect is exactly balanced by the inner conductor. The only place where there is any mutual inductance between one turn and the next is on the outer surface of the braid. This has nothing to do with the skin effect.

In yet another effort to be clear, the windings, if you can call them that, are in series with the applied voltage not parallel as with transformers. Euc (talk) 19:58, 3 July 2012 (UTC)

Skin effect is important, because it gives three conductors rather than two (i.e. the inner and outer surface of the coax braid, plus the centre conductor). The coupling you are looking for, if we are trying to describe this as a transmission line transformer, is between the centre conductor of the coax and the inner surface of the braid. Assuming the coupling is strong (k=1), i2=-i1, where i1 is the current on the centre conductor and i2 is the current on the inside surface of the braid. The coax itself is therefore incapable of transmitting a common mode signal, so any common mode must be carried on the outside surface of the braid, which is acting as a simple wire. Anything that blocks current flow on the coax outer (magnetics, looping the coax and resonating with a C, lambda/4, etc) will block the common mode. So, as you say, this is a common mode choke. If applied in the right place (i.e. close to the coax-to-non-coax transition), it also works as a balun. GyroMagician (talk) 14:37, 4 July 2012 (UTC)

That said, I would agree that there is no need to try to describe this as a TLT.

Sorry GyroMagician - I found something where we disagree. Skin effect and proximity effect have little to do with it. The inner surface of the braid and inner conductor are very strongly coupled (one is completely surrounded by the other) regardless of length. If you check the thickness of the skin effect (using copper) the skin depth is still far greater than the thickness of the braid up to about 1GHz or so (using decent coax). Skin effect is expressed as the depth (into a conductor) where the probability is 63% (or something I have forgotten exactly) that an electron will move in the direction of an applied voltage.

Proximity effect is also completely neutralised because for currents to be equal (and opposite), so must the voltage from place to place. Skin effect is specifically created by eddy currents in a single conductor. (Look Skin_effect up in Wiki.) Eddy currents are irrelevant to the strong coupling between the two inner conductors.

Perhaps skin effect was not very well named. It is not a simple rule that all AC is transmitted on the surface of conducting meterials. Skin and proximity effects are specifically used to calculate the resistance of a conductor given the conductor's conductivity, permeability (skin), frequency and geometry (proximity). They are created by certain conditions not the other way around.

In any case, the so called "transformer" shown is NOT any sort of transformer. Transmission Line Transformers are typically (but need not be) between 0 and 1/4 wavelength. Because the input impedance varies with length to a maximum at 1/4 lambda, you can use a piece of coax or even twin speaker wire to match just about anything to anything else.

Here is where we do agree, the thing shown is NOT such a transformer simply because length is totally irrelevant. The impedance at both ends of this transmission line are the same regardless of frequency (up to a point) or length. You can have 2 turns or 1000 (min about 30 recommended) and the action remains the same. There is zero (or near to it) winding impedance associated with the inner conductors but a high impedance (choke) on the outside of the braid. Any winding impedance present, and one reason why coax is very lossy at higher SWRs while balanced line isn't, is because the velocity factor of the inner conductor is not the same as the inner surface of the braid. Euc (talk) 07:26, 6 July 2012 (UTC)

I'll agree that we can ignore proximity effect here (I didn't mention it), but if you don't bring skin depth into the conversation, how do you have three conductors? Without a skin depth effect, the potential of the inner and outer surface (and so current density) of the braid would be the same. A quick skin depth calculation (based on the example on the wiki page), for copper at 100MHz gives 6.5μm, which is significantly smaller than the braid thickness of any coax I'm using. It's small enough that I'm happy to consider the current as flowing only at the surface, although this is, of course, a shorthand.
But we all know that coax will transmit signals down to DC, so maybe there is more to this. For a pure differential mode, the magnetic field generated by current on the centre conductors is exactly cancelled by the magnetic field generated by the braid - hence the is no H-field outside the coax. Therefore, any H-field interacting with the core must be related to common-mode. Do you prefer that description?
But, we also know the two conductors are strongly coupled, so any current flowing on the braid (if we exclude shielding by skin effect) must induce an equal and opposite current on the inner conductor. Can we say that the common-mode will cancel itself out? I.e. we have two equal currents I1 and I2 flowing on the inner and braid. I1 induces a current -I1 on the braid, and I2 induces current -I2 on the inner. Our conductors are now carrying currents I1-I2 and I2-I1, which is zero if I1=I2. I think that gives us a low frequency model, that says the common-mode simply won't propagate. As the frequency increases, the skin effect begins to separate the braid into two surfaces, and a common-mode may be propagated on the 'third' conductor that is the outer surface of the braid, without interacting with the inner braid surface.
Thoughts? GyroMagician (talk) 17:47, 6 July 2012 (UTC)

What's all this about either side of ground ?[edit]

Too often people say balanced signals are two signals either side of ground or some other reference. Balanced signals work against each other and have nothing to do with ground or any other reference. Twisted pair communications lines are a typical example. Suppose one end of the line is referenced to either side of ground. Changing magnetic fields, typically 50 or 60Hz mains generated fields, change the currents in both but, because they are twisted, affect both by the same amount thus, on the other end of the line there could be many volts of common mode (both the same in magnitude and phase) 50Hz signal. By examining the difference between the two and NOT the absolute voltage (referenced to anything), the original signal can be retrieved. Frequently in various computer communications systems, both signals fluctuate either side of 2.5 volts and not 0.Euc (talk) 21:32, 7 June 2012 (UTC)


There have been persistent attempts to add a link to this page pointing to the home page of an Italian electronics company that offers several models of balun. These have been posted by a number of different IPs, but all making the same errors of wikitext and English syntax and all linking to the same page.

It appears to be wikispam, and the claimed revolutionary nature of their products combined with the pseudo-scientific terminology on their website is not comforting. Andrewa 6 July 2005 17:57 (UTC)

transformer ground[edit]

with the earth ground or chassis ground left floating or unconnected on the balanced side.

All this really means is that the secondary of the transformer is not connected to ground, right? There's no ground in a transformer that isn't connected. — Omegatron 00:48, 16 December 2005 (UTC)
I have not yet really figured out how transformer baluns work and are wired, but I do understand what they do. The function of the balun is more in the transformer than it is in the connections on the outside of the transformer. The goal is to prevent certain current flows (common mode current in the coax shield) while allowing others (opposing currents in the shield vs. center conductor). I should post a good schematic picture. Each type of balun does its job slightly differently. The choke balun prevents the outside current while the coax allows the inside current. The narrow band ("hairpin") coax balun works by shorting out the common mode current with current 180 degrees out of phase -- which is why it is narrow band (you have to get the phase right). The transformer does something goofy with passing two out of three currents in opposing directions; the polarity of the winding connections is critical, but more than that I have not groked. --ssd 05:19, 3 January 2006 (UTC)

pronunciation guide for 'twin-lead'[edit]

I think possibly the pronunciation guide for 'twin-lead' is out of place here; possibly better suited to being moved into the twin-lead article itself? (anon)

I agree. Mostly. Actually, this is an encyclopedia, not a dictionary; I'm not sure the pronuciation guide belongs in the twin-lead article either. But I'll at least remove it here. And proof read the article for anything else I've missed. --ssd 04:32, 3 January 2006 (UTC)

balun picture[edit]

I don't know how to edit it myself, so I thought I'd just mention it here in case anyone wants to fix it. The picture of a 4:1 balun using a ferrite core has mispelled balanced (it's spelled in the picture 'balenced').

You edit the main page in the same way as you edited this one. Go for it.Euc (talk) 20:02, 3 July 2012 (UTC)

A nice white paper describing balanced signals[edit]

I'm too lazy to integrate its info into the article, but other editors may find this white paper on balanced video at Extron's web site a useful reference. It deals particularly with video over UTP, but the description and diagram of balanced signalling on page 4 seems clearer than anything in this article right now. --rcousine (talk) 20:40, 14 April 2008 (UTC)

Alternatives to Baluns[edit]

This section describes what I would call a cable trap. However, I think the described mechanism is wrong. Looping the coax, and adding a ferrite core, creates a large inductance on the coax braid. The high impedance blocks current on the braid, but the mechansim is reactive rather than resistive, so it is not lossy. The lossiness of the ferrite core is irrelevant. Am I missing something? GyroMagician (talk) 02:34, 23 November 2008 (UTC)

Looping the coax or using cores of ferite, iron or anything else, affects only currents on the outer of the braid. Currents on the inner of the braid and on the central conductor behave only as a longer piece of coax with no magnetic interaction. Currents on the central conductor and the braid's inner surface are already magnetically neutral to the outside world.Euc (talk) 21:39, 7 June 2012 (UTC)

Lossyness of the ferrite core is important, because if it is lossy, it will absorb some fraction of the power and will heat up. --ssd (talk) 15:20, 14 April 2013 (UTC)

Ratios never explained[edit]

There is no explanation of the difference between a "1:1 balun" and a "4:1 balun", or what those ratios mean. The page Transformer gives no clarification either. —Preceding unsigned comment added by (talk) 00:55, 28 March 2009 (UTC)

The ratio refers to the impedance on both sides of a device. It really has nothing to do with baluns (balanced <-> unbalanced) unless a transformer is used as a balun in which case the device serves two purposes, 1) to convert between unbalanced and balanced line and 2) to transform the impedance. As well as baluns, the ratio 4:1, 1:1 or x:1 might also be used with a plain transformer used to transform an impedance in that ratio or change voltage in the square root of the expressed ratio.

Euc (talk) 21:54, 7 June 2012 (UTC)

Amateur radio impedance[edit]

The impedance of coax used for amateur radios is 50 ohms; 75 ohms is used for tv. Rosattin (talk) —Preceding undated comment added 08:30, 12 October 2009 (UTC).

1:1 Choke Balun[edit]

The picture of the 1:1 choke balun was removed with the comment 'a choke is not a balun'. In this case, it is. Think of a coax cable as three conductors - the central conductor, the inside surface of the braid and the outside surface of the braid. A differential signal travels as a TEM mode between the central conductor and the inside surface of the braid. A common mode signal travels on the outside surface of the braid. Wrapping the coax around a ferrite introduces a large inductance to the common mode signal, without affecting the differential mode (which cannot 'see' the ferrite). The coax is unbalanced. A balanced line can be connected directly after the ferrite. This makes a nice simple broadband balun. For more details, look for anything written by Jerry Sevick GyroMagician (talk) 07:06, 29 April 2010 (UTC)

I agree totally - chokes can be used as baluns and you are right about common mode signals being on the outside of a conductor, although this has nothing to do with skin effect as often stated. Simple physics says the current in the central conductor MUST be equal and opposite the current on the shield inner. The magnetic field generated by the two currents and between them says it can't be anything else. Any attempt to change one will change the magnetic field between them and thus change the other by the same amount. Differences will therefore manifest themselves on the outer surface of the shield because there is no other place for them to flow. If you choke the currents on the outside of the shield, one end can be connected to a balanced load (equal and opposite currents) and the other end to unbalanced line (still equal and opposite currents but one side referenced to ground). Chokes can be used as baluns as can transformers.
Those who disagree with us just don't understand what is happening.

Euc (talk) 21:16, 7 June 2012 (UTC)

How about a transformer taxonomy?[edit]

I read this article, and encountered: "Baluns can be considered as simple forms of transmission line transformers. A more complex (and subtle) type results..."

On the whole, I found this article to be quite unhelpful. It seems to assume a huge amount of knowledge. This make it more like an "experts forum". I know quite a bit of basic electronics (little RF though), however I came away none the wiser. Sorry. — Preceding unsigned comment added by (talk) 03:59, 8 June 2011 (UTC)

Proof that wrapping magnetic material around coax is in use[edit]

See this catalog page from DX Engineering:

Similar units may be found on the signal cables of a vast array of computer equipment.

True, these clip-on ferrites are typically seen on computer equipment to prevent interference. But they don't usually offer enough attenuation to use them on an antenna - that either takes several turns of coax on the ferrite (as in the 1:1 balun photograph shown in the article), or a very large stack of ferrites. Either way, I think it is confusing to talk about 'wrapping' brittle ferrite material around anything.
On reading a bit more closely, I don't understand why this is offered as a balun alternative - coax wound on a ferrite core is a balun. GyroMagician (talk) 12:17, 23 January 2012 (UTC)
I edited the text to refer to passing the coax through the material rather than wrapping the material around the coax. And I tend to agree that it might be better to just refer to it as another version of a balun rather than a balun alternative. Ham radio operators routinely use a small stack of ferrite beads rather than wrapping the coax around a ferrite core. Some kinds of coax do not respond well to tight bends, and the coax with beads may be easier to mount than a coax wrapped around a core. The stack recommended in the ARRL Antenna Handbook cited is 7 beads. Jc3s5h (talk) 15:09, 23 January 2012 (UTC)
I like it - it's clearer now. GyroMagician (talk) 19:45, 23 January 2012 (UTC)
I like what GyroMagician is saying. (S)he has it right. I am a ham operator but don't even bother with the beads or the turns because they introduce losses by increasing the length of the feedline. I simply leave the upper end of the coax outer ungrounded (not connected to anything except the antenna) and tape (wire ties) the entire length of coax to the steel/iron tower and, where I can, thread it through any iron pipes or tubing forming part of the structure. Ordinary iron is no good for RF transformers because it's too inefficient (too many eddy currents) but for chokes who cares? A bolt or any old piece of iron works fine. Eddy currents don't matter, in fact, improve the performance as a choke.Euc (talk) 20:28, 3 July 2012 (UTC)

Citation style[edit]

I have undone GyroMagician's change in citation format for two reasons:

  1. WP:CITEVAR requires consensus for a change in citation style
  2. Presumably the general citations given in the article were used in writing the article. To relabel them as "Further reading" does not give proper credit to the sources used in creating the article.

I retained GyroMagician's other changes (adding a source and changing Ohm to Ω). Jc3s5h (talk) 20:56, 23 January 2012 (UTC)

Well, it was a work in progress. There is currently no way to tell what sources were used where, hence the label FR. Note the banner at the top of the article. In my opinion, the article currently lacks a citation style - which is why I started to add one. GyroMagician (talk) 21:01, 23 January 2012 (UTC)
When I added the first inline citation I established a style and documented it with a comment at the top of the "References" section:
<!--Citations follow Chicago Manual of Style documentation 2 style.-->
As for deciding which parts depend on the general citations, the only way to avoid plagiarism is to scrap the article and start over; otherwise passages based on the general citations may remain. Jc3s5h (talk) 21:09, 23 January 2012 (UTC)
I was not making any accusation of plagiarism, I was simply pointing out that currently there is no link between the contents of the article and the sources at the end. I would not define these as references. The 'inline citation style' is surely not really a style - isn't it better suited to printed text, rather than an electronic document? Wikipedia uses a variety of different referencing style - I am happy with any of them - but pure text is not suitable. GyroMagician (talk) 22:39, 23 January 2012 (UTC)
As a follow-up to my earlier message, here are some links. For a basic introduction to citing sources on Wikipedia, look at the tutorial. For much more detail, see WP:CITE. This article is currently extremely short on inline citations, something I would like to fix. Name-year style citations (Chicago or Harvard style) are less common on WP, but certainly allowable. Personally I do not like this stle, as the page quickly becomes crowded as more citations are added, but I am happy to discuss. However, whatever style is used, all references should be hyperlinked, allowing the reader to click on the citation in the body text to see the full citation details at the bottom of the page. I personally like citation templates (WP:CT) but I am aware not everybody does. The big advantage they provide is that various bots will come through and automatically fill in missing details such as ISBNs, DOIs, etc. One possibility is to use
in the body text (Straw 2005, p. 26-25), and place a citation template in the references section like so (edit this page to see the source):
More details here. I have not found a sensible way to achieve this without citation templates - maybe someone else knows one? These are only suggestions - but can we please agree on something designed for the web, rather than a print-standard.
I was intending to go through the article to add inline links to the provided (or further) references, but clearly there is no point if I am to be reverted before I can complete the task. The whole article is currently a mess, and needs some attention. GyroMagician (talk) 12:02, 24 January 2012 (UTC)
When your read WP:CITE you apparently missed this part:

A number of citation styles exist (some relevant Wikipedia articles include Citation, APA style, MLA style, The Chicago Manual of Style, Author-date referencing, Vancouver system and Bluebook). Examples can be found at Wikipedia:Citing sources/Example style.

As for what is or isn't a reference, any work consulted during the writing of the article should be listed as a reference if any ideas were used from the work. Since all we know is some previous editor decided to list them as references, we should assume good faith and presume the previous editor really did use them as references.
If you would like further help using Chicago style see the last two items at the bottom of the [[Wikipedia:Citing sources#Further reading|Further reading] section of WP:CITE.
Be aware that the effect on page load performance of citation templates has been questioned. Jc3s5h (talk) 16:22, 24 January 2012 (UTC)
I did in fact read the links I provided. I understand how to use the CMOS for print. Wikipedia is not primarily a printed source. What is not clear to me is how to implement CMOS links here. I do not think plain text is the best way to show citations on a page using HTML - we can do better. Do you have any suggestion?
I am aware of no significant negative impact of citation template on page load performance. I have used them on many small and large pages, and not yet run into any problems. As there are currently 75,000 transclusions of the Citation template, and WP is still running, they appears to work. However, if you have a suggestion that does create hyperlinks, does produce consistent formatting, and does not use citation templates, I am happy to discuss.
As to what does or does not constitute a reference, I think you misunderstood my point. In general, it is better to include inline citations - they tell the reader exactly what information comes from where. I intend to try to transform the current block into inline citations - this would improve the article, and would be much easier to do as a team. But I do not want to attempt to do until we can agree upon an adequate cite style. GyroMagician (talk) 18:28, 24 January 2012 (UTC)
I do not agree to use citation templates. I have edited the article to provide links from the parenthetical citations to the "References" section. Jc3s5h (talk) 19:06, 24 January 2012 (UTC)
Okay, this is a step forward. The citations are now links. But why are these transcludes are better than citation templates? They put the same load on the servers (negligible), but do not provide any of the auto-formatting benefit, which is useful to keep things consistent. I have learned that you do not like citation templates, but I still have no idea why. GyroMagician (talk) 19:47, 24 January 2012 (UTC)
Read Wikipedia:Centralized discussion/Citation discussion. That reminds me, this is a perennial proposal and I should document it as such. Jc3s5h (talk) 19:55, 24 January 2012 (UTC)
For a specific example where citation templates fail in an article with many citations, see
Jc3s5h (talk) 19:00, 3 February 2012 (UTC)

__Always__ use electromagnetic coupling?[edit]

"They always use electromagnetic coupling for their operation." Probably, this should be clarified. Fully Differential Amplifiers can be used to the same effect, numerous TI datasheets illustrate that. — Preceding unsigned comment added by (talk) 23:21, 27 April 2012 (UTC)


I see that 'baluns' are listed as being used to convert twinax to cat5. Since the signal remains balanced on both sides, is such an adapter really a balun? — Preceding unsigned comment added by (talk) 15:07, 28 February 2013 (UTC)

There's multiple factors involved. First, it's probably also an impedance transformer, as twinax is coax and cat5 is twisted pair, which have very different impedances. Second, if coax is involved, there is a chance for common mode current, and a balun would block that. --ssd (talk) 15:24, 14 April 2013 (UTC)