My new homemade LPDA (and the first)

[prcguy]

I have several commercial lab type log periodics from Electro Metrics, Singer, Racal and all terminate in an unbalanced RF connector and some have a string of ferrite beads on their internal coax harness to isolate the antenna from the outside world and unbalanced feedline. These are not cheap ham or scanner logs, but higher end units for the measurement community and designed to use as is with no external baluns.

As Mr. Wienke mentioned, a good common mode choke balun will isolate both the center and shield of coax when you compare the input to output side of the balun. If it doesn't, then is not a very well made choke balun.

One way to describe a balanced feedline is each parallel run conductor is RF isolated from ground or each other. If you take a coax feedline that is grounded and obviously unbalanced, then feed that through an effective 1:1 choke balun, then you will RF isolate that coax from ground, from the unbalanced input and meet the criteria of a balanced feedline on the outout side of the balun.
prcguy

It will only isolate the shield, stop any current running outside of the shield. You can do that by using toriods or coil up the coax that will inductivly shortcircuit the shields on each turn cancelling out any RF. You have to do that when something is wrong in the antenna system and you get RF on the coax shield.

You must use a transfomer to connect a coax to any type of balanced antenna. All log periodic antennas are balanced.

If you connect a coax directly to the booms of a log periodic antenna it will no longer function as one.

You can stick any metal up in the air and it will receive signals. More metal will receive more. Scanner antennas are often advertised as having a 25MHz-1000MHz range and it will actually receive all frequencies, if the signals are strong enough. When you connect a coax to a balanced antenna without using a balun transformer you will receive signals but the antenna will not perform as it was designed to do.

/Ubbe

 

[RFI-EMI-GUY]

(snip)
As Mr. Wienke mentioned, a good common mode choke balun will isolate both the center and shield of coax when you compare the input to output side of the balun. If it doesn't, then is not a very well made choke balun.

(snip)
prcguy

I am having a hard time with this concept. Yes I understand that the chokes block common mode currents from flowing on the coax (in ether direction), but the concept of it having any influence on the inner center conductor (shielded) escapes me.

Why if I were designing an antenna and had the option of a transformer or choke, would I choose the choke? How does the choke effect a true balance? With a transformer, assuming stray L and C on the radiators is equal, the balance is inherent in the design. I just don't see the same inherent properties with the choke as a balun. I could just be dense.

 

[jonwienke]

I am having a hard time with this concept. Yes I understand that the chokes block common mode currents from flowing on the coax (in ether direction), but the concept of it having any influence on the inner center conductor (shielded) escapes me.

For the simple reason that the choke balun forces the RF current in the center conductor to be equal but opposite the current in the shield. The effect is the same regardless of whether you use a choke balun or a transformer--equal-but-opposite currents on on the output-side conductors.

 

[jonwienke]

nvm

 

[nanZor]

I am having a hard time with this concept. Yes I understand that the chokes block common mode currents from flowing on the coax (in ether direction), but the concept of it having any influence on the inner center conductor (shielded) escapes me.

Looks to be answered but maybe some more detail...

To RF, due to "skin effect", coax has three surfaces:

1) Outer skin of the inner conductor
2) Inner "skin" of the coax braid
3) Outer "skin" of the coax braid.

Differential mode:
Currents on #1 and #2

Common mode:
Current on the outside skin of the coax braid. (so called because it has no opposing reference except for the common universe.

If there is an unbalance, differential mode currents can transition to common mode currents at the feedpoint for transmit, or common-mode transitions to differential mode, which is all your rx/tx gear is concerned with.

Normally this results in unwanted lobes, poor directionality, etc when the common mode skin becomes part of the antenna.

In some cases, like the so-called "shielded loop" made out of coax with a typical braid-cut of about 1 inch at the top, purposely utilize the common mode as the antenna and for directionality. The cut allows for a transition from common mode to differential mode inside. Unfortunately, 60 years of folklore insists on calling it a shielded loop, when in fact it is the common mode that is purposely used, not any sort of shielding. Most of us give up on the explanation.

Ferrite beads of the proper frequency range are the best, since they have no reactance to alter the feedpoint impedance reactance. Coiled coax can introduce reactance, which is usually undesirable, although this may be more of an issue at HF rather than UHF. I'll let prcguy's test gear pick up on that....

 

[RFI-EMI-GUY]

Looks to be answered but maybe some more detail...

To RF, due to "skin effect", coax has three surfaces:

1) Outer skin of the inner conductor
2) Inner "skin" of the coax braid
3) Outer "skin" of the coax braid.

Differential mode:
Currents on #1 and #2

Common mode:
Current on the outside skin of the coax braid. (so called because it has no opposing reference except for the common universe.

If there is an unbalance, differential mode currents can transition to common mode currents at the feedpoint for transmit, or common-mode transitions to differential mode, which is all your rx/tx gear is concerned with.

Normally this results in unwanted lobes, poor directionality, etc when the common mode skin becomes part of the antenna.

In some cases, like the so-called "shielded loop" made out of coax with a typical braid-cut of about 1 inch at the top, purposely utilize the common mode as the antenna and for directionality. The cut allows for a transition from common mode to differential mode inside. Unfortunately, 60 years of folklore insists on calling it a shielded loop, when in fact it is the common mode that is purposely used, not any sort of shielding. Most of us give up on the explanation.

Ferrite beads of the proper frequency range are the best, since they have no reactance to alter the feedpoint impedance reactance. Coiled coax can introduce reactance, which is usually undesirable, although this may be more of an issue at HF rather than UHF. I'll let prcguy's test gear pick up on that....

Yours is a better explanation. The ferrite has influence only on blocking current on the outer skin of the coaxial braid. It (the ferrite) imposes no effect on current flowing inside the coax on the inside skin of the braid or skin of the center conductor. The question is if this is as effective a means of balance when compared with a transformer. .

 

[jonwienke]

Yours is a better explanation. The ferrite has influence only on blocking current on the outer skin of the coaxial braid. It (the ferrite) imposes no effect on current flowing inside the coax on the inside skin of the braid or skin of the center conductor. The question is if this is as effective a means of balance when compared with a transformer. .

Not true. It blocks any current in the braid or center conductor that are not equal and opposite.

 

[RFBOSS]

Kirchoff's law tells us that is not correct.

If there is current flowing on the outside of the shield then the current flowing on the center conductor is equal to the current flowing on the inside of the shield plus the current flowing on the outside of the shield. Simple math.

Just a simple example.

The antenna is connected to a transmitter that causes 1amp of RF current flow.

If the unbalanced coax is connected to a balanced antenna there will be current flow on the out side of the coax.

Lets say just for a simple example that current is 0.25 amps. That is current that is not flowing into the antenna. The current flowing into the antenna will be what is left, 0.75 amps.

The common mode choke will force all of the current to flow into the antenna. The current flowing on the center conductor stays the same.

This is just a simple example, the feed point of the antenna may present complex impedance as well as the feed line presenting a complex impedance.

 

[RFI-EMI-GUY]

Kirchoff's law tells us that is not correct.

If there is current flowing on the outside of the shield then the current flowing on the center conductor is equal to the current flowing on the inside of the shield plus the current flowing on the outside of the shield. Simple math.

Just a simple example.

The antenna is connected to a transmitter that causes 1amp of RF current flow.

If the unbalanced coax is connected to a balanced antenna there will be current flow on the out side of the coax.

Lets say just for a simple example that current is 0.25 amps. That is current that is not flowing into the antenna. The current flowing into the antenna will be what is left, 0.75 amps.

The common mode choke will force all of the current to flow into the antenna. The current flowing on the center conductor stays the same.

This is just a simple example, the feed point of the antenna may present complex impedance as well as the feed line presenting a complex impedance.

I am not sure this is entirely correct. If the action of the choke is to provide a high impedance on the outside of the braid, then there is no diversion of current flow, there is a potential reduction of current flow. The mechanism by which the diversion if any, is accomplished is unexplained.

 

[RFBOSS]

The total current may change if the total impedance changes, that is why I said it is a simple example.

The total current still must be equal to the sum of internal and external current on the shield and that equals the current on the center conductor, even if the total current changes.

It is all about where the current comes from, where it goes and how it gets home.

This is the simple version of Kirchoff's law.

Another example.

Lets say we have folded dipole fed with a balanced fee line. In a perfect world all the power would go into the antenna.

Now lets say that we connect a piece of wire a quarter wave length long to one of the feed points. Current will flow in that wire.

The total current will be the sum of the current going into the antenna and the current in the added wire. The current in the other wire of the feed line will be the sum of the antenna current and the current in the added wire.

The added wire may change the total current but it does not change the principle.

The division of current depends on the impedance of each path, the antenna element and the impedance of the outside of the coaxial cable shield.

Same applies to my other example of adding a wire to the folded dipole antenna.

 

[RFI-EMI-GUY]

Basically, the addition of the ferrite choke, or a tuned sleeve balun will remove the feedline from the equation, and the impedance of the antenna load will likely change. The antenna will appear balanced with respect to the earth and feedline.

Sent from my SM-T350 using Tapatalk

 

[RFBOSS]

That is true, but that does not make what I posted invalid.

 

[RFBOSS]

To be a little more technically accurate.

The choke will dissipate some power depending on its impedance. This is vanishingly small on a receiving antenna.

 

[Ubbe]

I would propose that lab type LPDAs use ferrits to keep the gain/frequency curve flat for measurement purposes and cannot use a transformer balun due to its impedance complexity.

I also believe that choke balun/ferrits absorbs the outer skin current and dissipates the energy as heat.
The current in the innerlead and the inner skin will then differentiate and create an unbalance.

The unbalance will reduce the signal strenght and the directional of the antenna.

/Ubbe

 

[jonwienke]

You are wrong on all counts. A choke balun blocks common mode currents in the center conductor and the shield equally. The individual conductors in the shield braid are too small for there to be a difference in current between the inner surface of the shield and the outer surface. Especially given that the woven nature of the braid ensures that every individual braid wire alternates between being on the inside and on the outside.

FYI shielding works by currents induced in a conductor creating a magnetic field that is equal and opposite to the magnetic field that induced the current. You see this in superconductors at close-to-DC frequencies causing magnetic levitation. If there is no current flow, there is no shielding effect. So if you prevent current flow in the outer braid then a ferrite around the coax can affect current flow in the center conductor.

The choke dissipates some of the common-mode current as heat, but most of it is simply prevented from flowing in the first place due to inductive impedance.

 

[RFI-EMI-GUY]

You are wrong on all counts. A choke balun blocks common mode currents in the center conductor and the shield equally. The individual conductors in the shield braid are too small for there to be a difference in current between the inner surface of the shield and the outer surface. Especially given that the woven nature of the braid ensures that every individual braid wire alternates between being on the inside and on the outside.

FYI shielding works by currents induced in a conductor creating a magnetic field that is equal and opposite to the magnetic field that induced the current. You see this in superconductors at close-to-DC frequencies causing magnetic levitation. If there is no current flow, there is no shielding effect. So if you prevent current flow in the outer braid then a ferrite around the coax can affect current flow in the center conductor.

The choke dissipates some of the common-mode current as heat, but most of it is simply prevented from flowing in the first place due to inductive impedance.

A bit of a sanity check needed here.

If what you say is true, that the copper braid thickness is insignificant to the skin current,

1) the shielding properties would be nil, and the cable would have a high impedance loss..

2) assuming the ferrite choke would be effective to choke the current flowing on the inner as well as outer side of the braid, the VSWR would be significantly raised simply by the introduction of the choke balun. Even if terminated in a resistive load.

I think if you calculate the skin depth vs the mil thickness of the braid for frequencies HF through VHF you will find the braid is sufficiently thick. I could be wrong, but I did this exercise at VHF freqs just a day or so back.

 

[jonwienke]

Wrong on all counts.

Shielding is based on equal and opposite currents, which create equal and opposite magnetic fields, which cancel each other out, which means they essentially do not exist.

A ferrite has no effect on balanced (equal but opposite) currents because there is no net magnetic field to interact with it. The balanced fields cancel out, and therefore do not exist as far as the ferrite is concerned. It does have an effect on unbalanced currents, because they have a net magnetic field which it can interact with. Thus, a ferrite (and a choke balun) presents a high inductive impedance to unbalanced currents, but has little or no effect on balanced currents. It doesn't care whether the imbalance is in the shield or the center conductor.

Skin effect applies to individual conductor size (each strand of the braid), not the size of the braid as a whole. And you didn't explain how there could be separate currents flowing in the inside and outside of the braid when the braid conductors are woven, and alternate being on the inside and outside of the braid.

What I am saying is easily verified experimentally. Get a piece of coax and a decent sized toroid, and wrap as many turns of coax around the toroid as you can. Measure the inductance from one end of the shield to the other, and from one end of the center conductor to the other. Both will exhibit equally high inductance. Then short the shield to braid on one end of the coax, and measure the inductance between shield and braid on the other end. It will be low.

 

[prcguy]

What I do know is a 1:1 ferrite choke balun can provide between 20 and 40dB of isolation between the input and output of the coax shield and I can measure that. There will be some isolation of the input and output of the center conductor and I don't know how to measure that individually. My mentor on this subject is now dead and I'm too lazy to to further research on my own.

With this isolation comes a bridge between the unbalanced feedline and balanced antenna, thereby allowing the unbalanced antenna to operate with some degree balance and at the very least, common mode currents on the feedline will be reduced to a point where the unbalanced feedline will have little to no effect on the resulting antenna pattern.

When I get back from travel I can easily run the experiment that mr wienke proposes, but I'm not sure what that will prove. I suppose you would want to measure inductance between the center conductor and shield with one end shorted before wrapping the coax around ferrite then after. Again, not sure what that will prove.
prcguy

Wrong on all counts.

Shielding is based on equal and opposite currents, which create equal and opposite magnetic fields, which cancel each other out, which means they essentially do not exist.

A ferrite has no effect on balanced (equal but opposite) currents because there is no net magnetic field to interact with it. The balanced fields cancel out, and therefore do not exist as far as the ferrite is concerned. It does have an effect on unbalanced currents, because they have a net magnetic field which it can interact with. Thus, a ferrite (and a choke balun) presents a high inductive impedance to unbalanced currents, but has little or no effect on balanced currents. It doesn't care whether the imbalance is in the shield or the center conductor.

Skin effect applies to individual conductor size (each strand of the braid), not the size of the braid as a whole. And you didn't explain how there could be separate currents flowing in the inside and outside of the braid when the braid conductors are woven, and alternate being on the inside and outside of the braid.

What I am saying is easily verified experimentally. Get a piece of coax and a decent sized toroid, and wrap as many turns of coax around the toroid as you can. Measure the inductance from one end of the shield to the other, and from one end of the center conductor to the other. Both will exhibit equally high inductance. Then short the shield to braid on one end of the coax, and measure the inductance between shield and braid on the other end. It will be low.

 

[jonwienke]

What my proposed experiment will prove is that a choke balun doesn't just affect currents in the shield/outer braid of the coax, it has the same effect on currents in the center conductor as well. If the currents are balanced (equal and opposite) then they can pass unimpeded. If they are unbalanced, the choke blocks the unbalanced portion of the current, regardless of whether it is in the outer or inner conductor.

 

[prcguy]

I can test the inductance and choking effectiveness of the shield on a choke balun but I can't think of a good way to test the choking effectiveness on the inner conductor separately. I don't think testing the inductance between the shield and center conductor of a choke balun with the other end shorted will be of much benefit because that will be be a complex circuit with internal cable capacitance, etc.
prcguy

What my proposed experiment will prove is that a choke balun doesn't just affect currents in the shield/outer braid of the coax, it has the same effect on currents in the center conductor as well. If the currents are balanced (equal and opposite) then they can pass unimpeded. If they are unbalanced, the choke blocks the unbalanced portion of the current, regardless of whether it is in the outer or inner conductor.