PDA

View Full Version : How does a J-pole work



NC4CW
11-30-2008, 12:03 PM
I'm going to build a 2 meter j-pole out of 1/2 copper tubing today. Can someone explain to me in very simple terms why the coax is not a direct short through the copper tubing? It appears to me that the coax conductor and shield are common through a short piece of copper. How does this contraption radiate?
Thanks:confused:

G3TXQ
11-30-2008, 12:56 PM
I'm going to build a 2 meter j-pole out of 1/2 copper tubing today. Can someone explain to me in very simple terms why the coax is not a direct short through the copper tubing? It appears to me that the coax conductor and shield are common through a short piece of copper. How does this contraption radiate?
Thanks:confused:

The abbreviated answer to your question is that the path through the tubing is not a direct short because it is a significant fraction of a wavelength long at 2m. An extreme example of this is a quarter wavelength transmission line - it can have a short-circuit at one end and yet present a very high impedance at the other end.

There's a detailed description of how the J-pole works on Cebik's website:
http://www.cebik.com/content/a10/vhf/jp1.html

73.
Steve

KX0Z
11-30-2008, 01:40 PM
Some people say its works like an overly complicated end fed half wave antenna with 1/4 wave impedance matching section, that works equally as well as a simple 1/4 wave ground plane.

K0CMH
11-30-2008, 01:40 PM
You may have trouble seeing what appears to be a direct short because of the difference between Alternating Current and Direct Current. If you already know the difference and why the J-pole would not have a short, read no further.

If you apply DC current to the coax, you will indeed have a short circuit.

When applying AC, at the frequencies we use, the path that appears to be a short actually has a very large amount of impedance to the AC.

Let us look at one cycle of the AC (in our case, the polarity of the voltage is changing once every 1/146,000,000 second).

As the voltage starts to rise from zero, it forces electrons into the parts of the antenna. As the voltage continues to increase, the electrons are accelerated.

Now (very important), accelerating electrons build both a magnetic field and a electric filed around the antena parts. These fields move off the parts in waves. This takes work to build these fields and waves. Just like pulling a weight with a rope, there is resistance to getting it moving, and so do the electrons offer resistance to getting moving. We call this resistance "impedance".

Looking at the shape of a J-pole, one can see that the waves will overlap eachother. The waves will interact differently in different places around the J-pole. In some places they will add and double their strength. This make is very easy for the electorns to move at that point on the antenna part. In other places they will subtract and cancel eachother out. And there can be all kinds of other combinations; adding only 1/2 strength, loosing only 1/10 strength, etc. At these points, the waves are pushing back on the electrons in the antenna parts. All this can also add and subtract to the "impedance" the AC current is seeing as it moves along the J-pole parts.

All the different impedances on different places on the J-pole add up to a total impedance seen at the point where the signal (the signal is our AC current changing polarity at 146,000,000 times per second) enters the antenna.

Therefore, the end of the coax sees a total impedance value, rather than the simple DC short.

In a J-pole, the feed point is at the place where it appears there is a short. But remember, the AC is building fields, going to zeor, then building fields in the opposite direction, ect. The attachment point of the coax sees all the resistnace (remember, we call it "impedance") to accelerating and decelerating those electrons, and making and collapsing those fields.

And another point of AC current. Just like it takes work to build those fields, when they collaps as the voltage returns to zero, the work it took to build those fields gets dumped back into the antenna parts. This is another source of impedance to the flow of the AC current.

The amount of impedance in the antenna is very dependent on the frequency. If we apply 60 Hz current (which is our household current), the J-pole would act like a short. The time that 60 Hz (changing voltage 60 times per second) has to go before it returns to zero voltage (that is, the time it takes to start at zero, go to +117 and return to zero) is so long that a whole lot of current can pass through the short before a lot of impedance can build up. There is enouh time that you would blow a fuse if you connected the J-pole to your wall outlet -- but it offers a lot of resistance (impedance) to VHF radio frequencies (146,000,000 Hz).

Another way to look at it is at household current, the antenna offers 60 bursts of impedance per second, while at your radio's frequency, the antenna offers 140,000,000 bursts of impedance per second.

Just to tie up some possible losse ends; when the switch is first turned on in a DC circuit, there is a rise in voltage from zero to the value of the source (let us say a battery). During that time, the battery sees the same "impedance" that an AC current would see. But (here is the BIG difference), once the DC current comes up to the max voltage of the battery, the electrons are no longer being accelerated. They are just running along at a steady speed. There is no more "impedance" in the circuit for DC current. Now the battery can send everything it has through that circuit. Ditto for the moment the switch is turned off. The circuit presents "impedance" due to the electrons slowing down to rest. It is the exact same impedance except in the opposite value, and it only lasts for that instant after the switch is turned off.

So, a short for DC current (and slow AC current) can actually be a resistance to AC current.

Man kind has observed and measured enough of the various ways AC current and radiating elements of antennas interact, they can predict things about it. This is what antenna designing is about.

Summary: AC current acts very differently from DC current, because of the rising and falling of voltage, and acceleration/deceleration of electrons.

The next step: Your radio is designed and built to work properly with a 50 ohm impedance. We can design an antenna to develope that 50 ohms impedance, but as you can imagine, it is 50 ohms only at a specific frequency. If we change the frequency, then the impedance offered by the antenna will change. This is why our antennas have to have specific shapes and lengths of its parts. A shape and length at one frequency may not work good at another (i.e., it would not provide 50 ohms impedance at the frequency we want).

And to tie up another loose end: unfortunately we use the same name for units of DC resistance as we do for AC impedance. The world would be better if a different name had been picked for impedance units. I believe this fact has caused unmeasureable confusion for those starting out in electronics.

W5DWH
11-30-2008, 01:52 PM
I built one out of copper several months ago. I was not happy with the results. The SWR varied greatly with height and surrounding objects. I could never get the SWR to a range that suited me (1.5:1 or less across the band). Every length of coax I used changed the SWR. I tried a choke and it didn't help.

I then built a ground plane antenna and it has a SWR of less than 1.2:1 across the band. I will stick with it.

K0CMH
11-30-2008, 02:09 PM
DWH:

I can offer that something may be wrong with the j-pole you built.

I have built any number of these j-poles, for both 2 meters and 70 cm, and they all work find. Friends of mine have also built many and they also work fine.

Possibly there is something wrong in the lengths of the parts of your j-pole. You may try downloading a design from the internet and double check all the measurements of your j-pole.

I know that I once had to take one of mine apart because of the same problem you mentioned, and found that I had purchased elbows with different lengths of the flared section, so that when I soldered them together to make the "U" part of the antenna, the two upright parts were to close together. A trip to a different hardware store found elbows that were like the ones I had used on my first j-pole, and reassembly resulted in another perfect j-pole at 146 MHz.

And also, yes, J-poles are very sensitive to metal objects that are close and at or above the bottom of the U. They tend to ignor metal objects below the bottom of the U.

Also, are you aware that a j-pole often has to be "tuned" by moving the attachment points on each of the two upright parts. Also, when tuning, one must move their body below the bottom of the U, or better, many feet away from the antenna, so that they do not interfer with the SWR measurements. Again, I found this out the hard way. The first design I down loaded said nothing about this. I found that me being within a foot of the antenna really threw off the swr reading. When I finally found a different design which mentioned this, I found that my SWR was just fine. I had to step down four steps on the ladder so that I was well below the bottom of the U.

Also, the j-pole is a "50 ohm antenna". If you accidently use 75 ohm TV coax, the SWR will be poor.

Hope this helps.

KA4DPO
11-30-2008, 03:45 PM
Some people say its works like an overly complicated end fed half wave antenna with 1/4 wave impedance matching section, that works equally as well as a simple 1/4 wave ground plane.


I have to agree on that. A J-pole has a little gain over the ground plane but not much, about 1 db which is not significant. The quarter wave ground plane is easy to build and is much easier to match to coax than a J-pole. In fact, the ground plane is almost fool proof.

NA0AA
11-30-2008, 06:24 PM
Mine work well - I built two of them, breakdowns for portable use - one slip joints one threaded joints. Expensive buggers with the price of copper, but it's a physically robust design in copper pipe.

But while it works OK, it's not a miracle antenna - some claim 3 db over a 1/4 wave vertical. I don't know myself. But mine work just fine for what they are.

W8JI
11-30-2008, 06:34 PM
I'm going to build a 2 meter j-pole out of 1/2 copper tubing today. Can someone explain to me in very simple terms why the coax is not a direct short through the copper tubing? It appears to me that the coax conductor and shield are common through a short piece of copper. How does this contraption radiate?
Thanks:confused:

The short is 1/4 wave from the open ends, and as such reflects an open circuit away from the short. It has to do with standing waves. You should read about them.

The J-pole is a pretty poor antenna design. The "antenna" is an end-fed 1/2 wave. It is UNbalanced.

The UNbalanced antenna connectes to a BALanced stub. Bad news for common mode currents.

The BALanced stub is fed by an UNbalanced line, the coax!!! Again that's bad news, because again we have an balabced to UNbalanced junction that is improperly treated.

So the truth is the antenna radiates, the matching stub radiates, and the feedline radiates with a J-pole. Anything conductive the J-pole is mounted on will also radiate. Sometimes you get lucky and the unwanted radiation isn't an issue, but most times it is an issue. Sometimes people don't notice it, but it always happens.

You can see how the J-pole evolved here:

http://www.w8ji.com/end-fed_vertical_j-pole_and_horizontal_zepp.htm

73 Tom

WA9SVD
11-30-2008, 06:53 PM
One issue is that J-Pole users will say their antenna "works fine." But Compared to WHAT??? Have they compared it even to a standard ground plane?

The truth is, a J-Pole CAN be an acceptable antenna; it's easy to construct (if not always easy to adjust) and in some cases may be more easily installed mechanically thay other antennas, including those that require a proper ground plane. (I.e., the "real estate" for a J-Pole is almost nil, whereas a traditional 1/4 wavelength ground plane antenna will require radials.)
But the J-Pole is not a "miracle" antenna. If "works fine" means accessing the local repeaters, and nothing more, then that's all that's needed. And in most cases, a simple 1/4 Wave vertical or ground plane antenna might function as well.
Too many people think their antenna (ANY antenna) works well (or even spectacularly) because they have no reference for comparison.
If a J-Pole suits your needs, by all means, use it. But consider other alternatives, and don't claim "great" results without a meaningful comparison.

(JUst FYI, I DO use a couple of J-Poles of the "Copper Cactus" Variety; for 2 M and 1.25 M bands; but that's because I only operate FM on those bands to a very few, very local, repeaters for club net check-ins.)

M3KXZ
11-30-2008, 07:00 PM
I don't think it matters that the matching section radiates. But I do think it should have a 1:1 current balun at the feedpoint to choke off the common mode currents on the feedline.

N3MQM
11-30-2008, 07:27 PM
With threads like this the "uneducated", like me, move up a step.

Thanks

KA9VQF
11-30-2008, 09:57 PM
Yes, the ‘U’ part of your J-pole is really a transformer at the frequency it is used at. I know that’s a little hard to grasp when you can see it is a direct short at DC but radio waves are just that waves not direct current.

I’ve built any number of J-poles and they have all worked to a degree. Some better than others even though to look at them side by side you can detect no difference. Its all in the placement of the finished antenna.

Personally I like simple žth wave ground plains because they are easily made and adjusted.

K7FE
11-30-2008, 10:03 PM
I don't think it matters that the matching section radiates. But I do think it should have a 1:1 current balun at the feedpoint to choke off the common mode currents on the feedline.

When you use a choke on the feedline you limit one of the sources of "balance" that a typical J uses. It would probably perform worse. Most J designs count on the outer portion of the shield to act as a counterpoise to the end fed half wave.

73,
Terry, K7FE

AE6ZM
11-30-2008, 10:12 PM
Whille I do not have any lab grade analysis, I do have some anecdotal observations on my J-pole performance. My antenna is the so-called "Super J" with a second half wave added with a matching stub, base at about 20 feet above ground level. Using my FT-60 set to 5 watts, I am full quieting into a 2M repeater located at about 4000 feet 80 miles away. So, while this may be a "poor design", mine seems to "work fine". It was easy to build, easy to mount, and easy to feed. In my book, that makes it pretty near perfect.

Wes
AE6ZM & VE7ELE
ARRL Technical Specialist
Lincoln, CA
Placer County ARES

WB2UAQ
11-30-2008, 11:03 PM
A useful characteristic of a J ant and other stub matched antennas: The stub matching arrangement can be quite freq selective and reduce intermodulation problems.

In my case I live near a weather service transmitter (164.4 MHz) and a number of paging transmitters in the 152 - 153 MHz range. By making the spacing closer and closer beween the conductors in matching section, I was able to reduce the antenna's bandwidth considerably. Unfortunatly the adjustments became very critical BUT I was able to reduce the intermod considerably. I don't have any weather channel images at all and very little paging service issues.
73, Pete

W8JI
11-30-2008, 11:29 PM
I don't think it matters that the matching section radiates. But I do think it should have a 1:1 current balun at the feedpoint to choke off the common mode currents on the feedline.

It only matters because the unwanted common mode cancels low angle radiation, makes the antenna sensitive to how it is mounted and what it is mounted on, and makes SWR vary with the environment around the feeder and what the antenna is mounted on.

This is why we generally don't see commercial systems using J-poles, even though they are very popular for Hams. In ham radio the ease of construction and low cost outweighs the electrical disadvantages.

73 Tom

M3KXZ
11-30-2008, 11:34 PM
When you use a choke on the feedline you limit one of the sources of "balance" that a typical J uses. It would probably perform worse. Most J designs count on the outer portion of the shield to act as a counterpoise to the end fed half wave.

73,
Terry, K7FE

Maybe I'm misunderstanding something, but I can't agree that the j-pole needs to use the outer shield of the coax as a counterpoise. Think about it, would you expect an off-centre fed dipole to require a counterpoise? No, I guess, as it has antenna element both sides of the feedpoint. But it would need a balun as it's an unbalanced antenna and would result in common mode currents on either a coax feeder or on a twin line feeder. Now, when I look at a j-pole, I see an OCF dipole that has been bent into a "j" shape, and has a shorted transmission line matching stub. Being nothing more than a bent OCF dipole, it doesn't need a separate counterpoise.

I've attached a picture to illustrate! What have I missed?

Pete

AI6DX
12-01-2008, 12:14 AM
DWH:

I can offer that something may be wrong with the j-pole you built.

I have built any number of these j-poles, for both 2 meters and 70 cm, and they all work find. Friends of mine have also built many and they also work fine.

...Also, are you aware that a j-pole often has to be "tuned" by moving the attachment points on each of the two upright parts.

I agree with your statements above.... I built a nice 222mHZ copper J-pole. Tuning it was very simple by moving the attachment parts and watching the antenna analyzer bring me to a 1.2:1 match at 223.5 mHZ.

They are really simple, dependable and durable antennas.

WB2WIK
12-01-2008, 02:23 AM
I agree with your statements above.... I built a nice 222mHZ copper J-pole. Tuning it was very simple by moving the attachment parts and watching the antenna analyzer bring me to a 1.2:1 match at 223.5 mHZ.

They are really simple, dependable and durable antennas.

::Here's a good statement:

It only matters because the unwanted common mode cancels low angle radiation, makes the antenna sensitive to how it is mounted and what it is mounted on, and makes SWR vary with the environment around the feeder and what the antenna is mounted on.

This is why we generally don't see commercial systems using J-poles, even though they are very popular for Hams. In ham radio the ease of construction and low cost outweighs the electrical disadvantages.

73 Tom

::I agree with this. The j-pole radiates, as does anything conductive. It's a lousy antenna design, and the only ones using it are amateurs. You'll never see a "commercial j-pole" because it's a terrible antenna design. A 1/4-wave ground plane with 4 radials, made from an SO-239 and five pieces of 19" copper wire (about $2 to build) is, in general, a better antenna.

Of course those of us who think so have only tried everything there is over 40 years or more.

Feel free to keep trying stuff.:p

But when you come up with a new idea that works better, try to make sure it's actually a new idea. The j-pole's been around a long time and nobody has ever proven it to work better than a 1/4-wave ground plane -- ever.

WB2WIK/6

NC4CW
12-01-2008, 10:07 AM
I appreciate the responses and explainations. I built the j-pole and seems to be working well. The swr is slightly over 1:1 and I can hit a repeater over 100 miles away.

Thanks
Dave WB4DLR:)

W8JI
12-01-2008, 02:44 PM
Maybe I'm misunderstanding something, but I can't agree that the j-pole needs to use the outer shield of the coax as a counterpoise. Think about it, would you expect an off-centre fed dipole to require a counterpoise? No, I guess, as it has antenna element both sides of the feedpoint. But it would need a balun as it's an unbalanced antenna and would result in common mode currents on either a coax feeder or on a twin line feeder. Now, when I look at a j-pole, I see an OCF dipole that has been bent into a "j" shape, and has a shorted transmission line matching stub. Being nothing more than a bent OCF dipole, it doesn't need a separate counterpoise.

I've attached a picture to illustrate! What have I missed?

Pete

Pete,

I know the final test for a Ham is if they can access a repeater hundreds of feet in the air with a high gain antenna, or if the SWR is low or can be made low. That tells them the antenna works perfectly.

Commercial people have to do things differently. They can't have an antenna that changes tuning depending on what is is mounted to or how long the feedline is. It has to come out of the box and work the same every time, so they are more concerned about common mode and having predictable SWR and the best gain at low angles.

What you are missing is explained at this link:

http://www.w8ji.com/end-fed_vertical_j-pole_and_horizontal_zepp.htm

The end-impedance of a 1/2 wave is NOT infinite. Depending on length to diameter and what is around the antenna, it can be a few hundred ohms to a few thousand ohms. This means whatever end-current you force into the antenna has to be matched by an EQUAL common mode current flowing down the stub at that junction. This means the stub always radiates.

How much the stub radiates depends on what attaches to the stub and the impedance it presents. The stub can radiate as much or more than the antenna radiates, or it can radiate an insignificant amount with no other change than what is attached to the stub or how long something is attached to the stub!

Sometimes the radiation can be harmful, sometimes harmless. It all depends.

Notice none of this says it won't radiate something. Some people will just fall in love with their antenna because it has a low SWR or works a repeater high above earth a long distance away. What it really says is it isn't a good design for making a repeatable antenna. This is why Motorola doesn't sell J-poles, they sell and install ground plane antennas that are repeatable and stable.

To NOT have common mode with an end fed antenna you must have two things:

The antenna must have a counterpoise of something for the antenna to push against.

The antenna must have something that isolates or has a very low impedance to "sink" the unwanted common mode currents generated by end feeding an antenna.

There is no way around those two rules, but the J-pole "design" pretends they don't exist. It really is a very poor design, which is not the same as saying it won't radiate something. It is just a poor design.

Read the link and you will see why.

73 Tom

PY4LC
12-01-2008, 03:12 PM
The short is 1/4 wave from the open ends, and as such reflects an open circuit away from the short. It has to do with standing waves. You should read about them.

The J-pole is a pretty poor antenna design. The "antenna" is an end-fed 1/2 wave. It is UNbalanced.

Indeed for the J-pole and the End Fed Zepp to present a real impedance (with no reactance) on the 1/4 wl line to connect the transmitter line, the radiator must be slightly capacitive, or a little less than 1/2 wl. I published an article on this issue in my home page, www.geocities.com/py1ll.

Luiz - PY1LL/PY4LC

M3KXZ
12-01-2008, 08:01 PM
Pete,

<snip>
What you are missing is explained at this link:

http://www.w8ji.com/end-fed_vertical_j-pole_and_horizontal_zepp.htm

The end-impedance of a 1/2 wave is NOT infinite. Depending on length to diameter and what is around the antenna, it can be a few hundred ohms to a few thousand ohms. This means whatever end-current you force into the antenna has to be matched by an EQUAL common mode current flowing down the stub at that junction. This means the stub always radiates.
<snip>
To NOT have common mode with an end fed antenna you must have two things:

The antenna must have a counterpoise of something for the antenna to push against.

The antenna must have something that isolates or has a very low impedance to "sink" the unwanted common mode currents generated by end feeding an antenna.

There is no way around those two rules, but the J-pole "design" pretends they don't exist. It really is a very poor design, which is not the same as saying it won't radiate something. It is just a poor design.

Read the link and you will see why.

73 Tom

Tom, I've read the link again and again previously but it still doesn't answer my query. What I'm getting at though is this. If there is a good current balun AT THE ANTENNA BASE that provides a choking impedance of several thousand ohms, then what? Surely the antenna will behave no differently from a vertical off centre fed dipole with a balun at the antenna feedpoint. The same with the Zepp - with a GOOD CURRENT BALUN of several thousand ohms choking impedance at the antenna base (i.e. the bottom of the 1/4 wave twin lead section), would there still be a problem? I'm not saying I like the j-pole design, especially for VHF and higher as it's so easy to get a GP antenna or vertical dipole at a decent height above ground. But I am curious as to whether the j-pole design with a GOOD current balun at its base would be ok. Is the problem that people pretend there's no need for this?

73

Pete

M3KXZ
12-01-2008, 08:06 PM
Also, I don't see the j-pole as an end fed antenna. It IS a bent off centre fed dipole. As such, I still think it's fine for the twin section to radiate. But people should bear this in mind and not bolt the metal directly to a mast as I understand many people do, or have this section close to metal objects. It really would need an insulated stand off AND a GOOD CURRENT BALUN at the feedpoint. Which I guess makes it not a very good antenna for someone who just wants to whop something up onto a metal mast!

Pete

KA0GKT
12-02-2008, 05:18 AM
Also, I don't see the j-pole as an end fed antenna. It IS a bent off centre fed dipole. As such, I still think it's fine for the twin section to radiate. But people should bear this in mind and not bolt the metal directly to a mast as I understand many people do, or have this section close to metal objects. It really would need an insulated stand off AND a GOOD CURRENT BALUN at the feedpoint. Which I guess makes it not a very good antenna for someone who just wants to whop something up onto a metal mast!

Pete

Look at the J-pole as a Zeppelin (Zepp) antenna. The radiator portion of the antenna is a halfwave section. It is fed via a quarterwave length of balanced feed. The high impedance of the end-fed 1/2-wave is transformed to a more managable impedance at the end of the 1/4-wave . That impedance is then transformed to match the charecteristic impedance of the coaxial cable by a shorted stub. Look at the explanation of the Zepp in the ARRL antenna book for a better explanation of why the "J" or "U" portion of the antenna doesn't radiate.

At best, a J-pole is just another halfwave antenna. It has a gain of zero dBd.

W4CBJ
12-02-2008, 05:30 AM
Besides the correct dimensions...you need a common-mode choke made from the same coax that feeds the J-pole, A few tuns of coax ...WHERE IS PUSHRAFT WHEN YOU NEED HIM ? HI 73 Joe W4CBJ

W4CBJ
12-02-2008, 05:43 AM
Actually, it is NOT a balun. A balun is a BALANCED to UN-BALANCED device. On the J-pole, you have the coaxial cable and the J-pole....both un-balanced. It is a COMMON-MODE choke. Prevents RF on the outer conductor of the coax. Best if you read up on the type of antenna in the ARRL antenna book. All the ones that I have built ...worked fine. In fact, I have some copper foil ribbon and plan to attach it to a window. (It has adhesive on one side.) You can "tap up" on the antenna (about two (2) inches from the "U" to achieve the lowest SWR at your design frequency. Would be nice to use an MFJ 259B antenna analyzer ....if you have one. 73 Joe W4CBJ

K7FE
12-02-2008, 06:18 AM
Also, I don't see the j-pole as an end fed antenna. It IS a bent off centre fed dipole. As such, I still think it's fine for the twin section to radiate. But people should bear this in mind and not bolt the metal directly to a mast as I understand many people do, or have this section close to metal objects. It really would need an insulated stand off AND a GOOD CURRENT BALUN at the feedpoint. Which I guess makes it not a very good antenna for someone who just wants to whop something up onto a metal mast!

Pete

An off center fed dipole is "split." The coax center conductor goes to one side and the shield connects to the other side of the split element.

The J antenna is not split. The lower quarter wave section or "Q section" is for matching. (Read up on matching sections.) When you remove that matching section, the remaining antenna is just a 1/2 half wave vertical element. That 1/2 wave vertical element will need radials (or counterpoise) to balance the antenna and carry a current equal to the current on the 1/2 wave vertical portion. Since there are no radials, the RF will seek other paths to achieve balance. According to Kirchoff's Current Law. At every node, the sum of all currents entering a node must equal zero. You cannot drive current in that vertical element without an equal return current on "the other half of the antenna." Since it has no radials or "the other half of the antenna", common mode currents appear on places like the outside of the coax shield, metal mast, tower and the Q section.

These variable factors in current balance make the J-pole very unpredictable and installation dependent. Some people may have a reasonable performer and the next, not so good. That why I always suggest a 1/4 wave ground plane be used rather than a J-pole. The 1/4 wave ground plane is lower in cost, easy to build, lighter, easier to match and has a consistent performance that is equal to and at times better than a J.

73,
Terry, K7FE

M3KXZ
12-02-2008, 06:49 AM
Actually, it is NOT a balun. A balun is a BALANCED to UN-BALANCED device. On the J-pole, you have the coaxial cable and the J-pole....both un-balanced. It is a COMMON-MODE choke. Prevents RF on the outer conductor of the coax. Best if you read up on the type of antenna in the ARRL antenna book. All the ones that I have built ...worked fine. In fact, I have some copper foil ribbon and plan to attach it to a window. (It has adhesive on one side.) You can "tap up" on the antenna (about two (2) inches from the "U" to achieve the lowest SWR at your design frequency. Would be nice to use an MFJ 259B antenna analyzer ....if you have one. 73 Joe W4CBJ

Cheers Joe, "common mode choke" was my intended term - I did the dumb thing that loads of people do and used the term "balun" as a generic one.

Pete

M3KXZ
12-02-2008, 07:05 AM
An off center fed dipole is "split." The coax center conductor goes to one side and the shield connects to the other side of the split element.

The J antenna is not split. The lower quarter wave section or "Q section" is for matching. (Read up on matching sections.) When you remove that matching section, the remaining antenna is just a 1/2 half wave vertical element. That 1/2 wave vertical element will need radials (or counterpoise) to balance the antenna and carry a current equal to the current on the 1/2 wave vertical portion. Since there are no radials, the RF will seek other paths to achieve balance. According to Kirchoff's Current Law. At every node, the sum of all currents entering a node must equal zero. You cannot drive current in that vertical element without an equal return current on "the other half of the antenna." Since it has no radials or "the other half of the antenna", common mode currents appear on places like the outside of the coax shield, metal mast, tower and the Q section.
<snip>
73,
Terry, K7FE

Terry, I know about matching sections. BUT you cannot consider the single element 1/2 wave part of the j-pole as an antenna on its own into which one is trying to force current.

It is NOT a 1/2 wave end fed on it's own. It is connected to one side of the 1/4 wave matching section. At the other side of the feedpoint is another 1/4 wave matching section. If the feedpoint has a GOOD common mode choke on the rig side, then the antenna consists of a 3/4 wave section on one side and a 1/4 section on the other. There IS somewhere for the return currents to go - the 1/4 wave section on the opposite side of the feedpoint from the 3/4 wave section. The currents are not equal thus the "matching section" (i.e. the bottom 1/4 wave of the 3/4 wave section and the other 1/4 wave section) radiate - the summing of the currents on this section (i.e. of the current magnitude and phase) in this section will be zero at the bottom (provided the antenna feedline is properly choked).

Then we have the shorted part - this is no different from a shorted transmission line stub to aid matching. This is the only part that differentiates the j-pole from a "bent off-centre fed dipole" or "Zepp" (which is also a bent off-centre fed dipole). In fact, such a shorted stub could be added to an off-centre fed dipole to match it, couldn't it?

I'm not trying to be cantankerous, but I still don't see the j-pole (or Zepp) as an end-fed 1/2 wave element on its own with a matching section. Neither though, for situations where the antenna can be mounted a good height above ground (which is all situations really at VHF), do I see any advantage to j-pole over 1/4 wave GP or vertical dipole, so please don't think I'm trying to say its good for these situations!

73

Pete

G3TXQ
12-02-2008, 09:19 AM
I'm not trying to be cantankerous, but I still don't see the j-pole (or Zepp) as an end-fed 1/2 wave element on its own with a matching section.
Pete, I hesitate to post to this thread because I don't operate VHF and I've no experience of the J-pole. However it does seem to me it's easier to visualise the J-pole as a derivitive of the Zepp than as a derivitive of an OCF full-wave.

Simplistically, if I picture moving the feedpoint of a half-wave dipole away from the centre, and at each stage introduce a transmission-line section to match the increasing feedpoint resistance, the Zepp is what I get "in the limit".

From the J-pole modelling results I've looked at it's clear that the 1/4 wave stub is carrying some common-mode component of current as well as differential-mode, and therefore will radiate to some extent; but the differential mode is clearly the more dominant; so, put simply "it's acting more like a transmission line than an antenna".

That cannot be said for the OCF full-wave in your earlier drawing. The current in the "fed half-wave" is all common-mode and all contributes to radiation. So for me, the Zepp model is the much better analogy - it needs little "tweaking" to represent the J-pole characteristics quite accurately, whereas the OCF full-wave would have to be so contorted that it loses any resemblance to its original form and characteristics.

But I suppose we all view things in the way we best can understand them.

Steve

W8JI
12-02-2008, 11:58 AM
Pete, I hesitate to post to this thread because I don't operate VHF and I've no experience of the J-pole. However it does seem to me it's easier to visualise the J-pole as a derivitive of the Zepp than as a derivitive of an OCF full-wave.

Either way works.

Consider it a Zepp ...... or consider it an OCF dipole with the two wire feeder overlapping the shorter half of the OCF dipole and directly connected to that half.


Simplistically, if I picture moving the feedpoint of a half-wave dipole away from the centre, and at each stage introduce a transmission-line section to match the increasing feedpoint resistance, the Zepp is what I get "in the limit".

That's actually what it is...... but if you can picture one of the two wires of a balanced feeder being integrated into the short leg of the ocf dipole, that works too. It COULD be considered an OCF with a balanced two wire feeder with one wire of the feeder bonded to the short end of the dipole, then you tap a coax up that shorted feeder! That shows what a MESS that antenna is.


From the J-pole modelling results I've looked at it's clear that the 1/4 wave stub is carrying some common-mode component of current as well as differential-mode, and therefore will radiate to some extent; but the differential mode is clearly the more dominant; so, put simply "it's acting more like a transmission line than an antenna".

That's true, but the comon mode is the same as it is into the radiator at that end of the line, and it can go UP with distance. To help people understand that it actually might be better to picture the J-pol or Zepp as a OCF dipole with the balanced feeder integrated into the antenna and the coax tapped up or down on that balanced line that is also common wth the radiator.

That might help them see how grounding the shorted part of the J causes a large change in tuning, and how the coax has common mode issues and detunes the antenna.

I don't think the mental model of an OCF dipole is bad. It actually might help people understand how poor the feed system is on an end fed zepp and a J-pole.

Not many people would feed an OCF dipole with a balanced line, and then bond one conductor of that line to the short half of the OCF dipole, make the OCF antenna 3/4 wave long, and then hang coax from it and ground the short end.

The J-pole 'works' in spite of the silly design, not because it was well planned, and viewing it as an OCF dipole actually might help explain it even better than the common mode model I use.



But I suppose we all view things in the way we best can understand them.

Steve

With that in mind, does the OCF model work better for people trying to understand the J-pole??? Here it is......

The J-pole is a 3/4 wave long OCF dipole, with one wire of the two wire balanced feed integrated into the short end of the OCF dipole. From that two wire line, we hang a piece of coax. We also might ground the short end of the OCF dipole to a random length and size tower or mast.

That might explain why the J-pole isn't favored as a "good design" for commercial application, although anything that will radiate, not blow up radios, and hit sensitive repeaters will make Hams happy.

73 Tom

G3TXQ
12-02-2008, 12:10 PM
The J-pole is a 3/4 wave long OCF dipole, with one wire of the two wire balanced feed integrated into the short end of the OCF dipole.
Tom - yes, but Pete's starting point was a linear full-wave OCF which needed just one "bend" too many for my imagination :)

Steve

W8JI
12-02-2008, 12:55 PM
Tom - yes, but Pete's starting point was a linear full-wave OCF which needed just one "bend" too many for my imagination :)

Steve

If he had a 3/4 wave OCF dipole (which is a dumb idea for an antenna), and then if he had fed it with a two wire line and then folded the two wire line into the shorter end of the OCF antenna (which is a dumb idea for a feeder) and then shorted that feeder and tapped coax into it (which is a dumb way to feed the system) and then connected the tower or mast to the shorted end of the OCF dipole, he would have a typical J-pole.

I agree with you now that I see he had it a full wave long. Since he made the dipole the wrong length (a full wave instead of 3/4 wave) it was a bad model.

It could be considered an OCF vertical 3/4 wave long with a horrible feed system and a tower or mast tied to the shorter leg of the OCF dipole. That would make a good model of what it really is. Probably easier to understand than the Zepp model is, but I'd have to see how people follow that description.

73 Tom

KA4DPO
12-02-2008, 04:47 PM
Terry, I know about matching sections. BUT you cannot consider the single element 1/2 wave part of the j-pole as an antenna on its own into which one is trying to force current.

It is NOT a 1/2 wave end fed on it's own. It is connected to one side of the 1/4 wave matching section. At the other side of the feedpoint is another 1/4 wave matching section. If the feedpoint has a GOOD common mode choke on the rig side, then the antenna consists of a 3/4 wave section on one side and a 1/4 section on the other. There IS somewhere for the return currents to go - the 1/4 wave section on the opposite side of the feedpoint from the 3/4 wave section. The currents are not equal thus the "matching section" (i.e. the bottom 1/4 wave of the 3/4 wave section and the other 1/4 wave section) radiate - the summing of the currents on this section (i.e. of the current magnitude and phase) in this section will be zero at the bottom (provided the antenna feedline is properly choked).

Then we have the shorted part - this is no different from a shorted transmission line stub to aid matching. This is the only part that differentiates the j-pole from a "bent off-centre fed dipole" or "Zepp" (which is also a bent off-centre fed dipole). In fact, such a shorted stub could be added to an off-centre fed dipole to match it, couldn't it?

I'm not trying to be cantankerous, but I still don't see the j-pole (or Zepp) as an end-fed 1/2 wave element on its own with a matching section. Neither though, for situations where the antenna can be mounted a good height above ground (which is all situations really at VHF), do I see any advantage to j-pole over 1/4 wave GP or vertical dipole, so please don't think I'm trying to say its good for these situations!

73

Pete


Pete, in the case of the double extended Zep, a two half wave flat top. I built several of these for two meters back in the day and used a 1/4 wave matching section and choke Balun to feed them. It worked and theoretically at least, the matching section should not radiate. I still view the J-pole as an end fed 1/2 wave antenna.

W8JI
12-02-2008, 05:47 PM
Pete, in the case of the double extended Zep, a two half wave flat top. I built several of these for two meters back in the day and used a 1/4 wave matching section and choke Balun to feed them. It worked and theoretically at least, the matching section should not radiate. I still view the J-pole as an end fed 1/2 wave antenna.

That's another way to view it, so long as you remember the matching section has to have the same common mode current as excites the radiator at the junction of the two.

This means the matching section will ALWAYS be part of the antenna and radiate, as will anything attached to the matching section.


I think the OCF dipole is a better way as long as were remember the short leg is 1/4 wave long and the long leg is 1/2 wave, and that the short leg is attached to other things.

I actually like the OCF model. It is much more accurate if we picture it properly.


73 Tom

M3KXZ
12-02-2008, 07:03 PM
Tom, me 'ead's hurting!

Here's a thought....I'd be grateful if you could have a look at it - perhaps an EZNEC model to see how it compares to mine, which I can email you if you with. A vertical FULL wave dipole for 2m, fed 25% from the bottom end, with bottom of bottom leg 3m from ground. I've modeled this as bottom wire (wire 1) 0.485m long (1/4 wave), connected to a short wire (wire 2) of 0.03m with this short wire connected to the top dipole wire (wire 3) of 1.485m long (3/4 wave). The source is placed 50% up length of wire 2. This gives a full wave dipole at approx 145MHz, 580 ohm SWR somewhere around 1.7 (I haven't bothered "tuning" it). The elevation pattern shows RF at high angles as expected from a vertical full wave dipole. I've ignored feedline and am making an assumption that a good common mode choke would be in place at the feedpoint, and that it's not connected to any nearby ground or whatever.

Now, I converted this to a J-pole with the source at same height above ground (near enough) by turning wire 2 horizontal and wire 1 no going up alongside wire 3, so we have a 3/4 wave vertical section alongside a 1/4 wave vertical section, in typical J-pole or Zepp style. SWR is "similar", overall gain is lower, but max RF is now at much lower elevation angles. Again, assumption made that proper common mode choke is used. All that has happened is that current summing has taken place on the portions of the antenna that are now folded alongside each other - the currents are pretty much 180 degrees out of phase, and are of similar magnitude at the feedpoint.

That's how I folded a full wave OCF dipole to get a J-pole.

Performance (modeled) though is crap compared with a centre fed half wave vertical dipole at the same height, especially considering all the effort that goes into such an antenna. And with them being so small at VHF, I can't help but wonder why!

73

Pete

K0CMH
12-02-2008, 07:59 PM
No matter if the design is poor or not, j-poles for vhf use have a reputation of working reasonably well.

Their attribuites seem to outweigh their negatives.

I am involved in my local ARES group and many, many of our members use the old j-pole for emcom purposes, for the following reasons:

1. They are extremely cheap and easy to make.
2. As stated in an earlier reply here, they take up very little horizontal space when deployed.
3. They can be made to break down into very managable sections for transportation and storage.
4. They do work well for local vhf communications, both repeaters and simplex. Many of our members can communicate with j-poles, on simples, 20 miles.

My personal experience, and my observation of any number of others who I have helped get on the air, indicate that the vhf j-pole is dependable and repeatable, provided we are talking about the 20 mile radius at 10-20 watts of xmit power.

I agree that they have little or no gain over a dipole or 1/4 wave ground plane, however, they do perfom good enough.

My group has over three years experience of both home and field deployment with j-poles to feel confident that they will do the job for local VHF work.

And yes, some of us have "beams" and other higher gain deployable antennas if/when we need longer ranges. But VHF is not reliable for 100 mile communications without specialized antennas. If you want to go that far and further, then NVIS on the HF bands is the way to go.

Edit:regarding "only Hams use j-poles and are the not used in commercial applications" (paraphrased by me): I don't see many wire center-fed dipoles in commercial use either, but just about every Ham on HF is using one, or has unsed one at one time or another.

M3KXZ
12-02-2008, 08:15 PM
<snip>Edit:regarding "only Hams use j-poles and are the not used in commercial applications" (paraphrased by me): I don't see many wire center-fed dipoles in commercial use either, but just about every Ham on HF is using one, or has unsed one at one time or another.

There are millions of wire centre fed dipoles in commercial use at HF frequencies and lower. I've only got to drive a few miles from here to see a whole load of them on a tower at Shoreham, and I can look out of my aunty's window in London and see one on top of the Swiss Embassy. And the army use them.

Pete

W8JI
12-02-2008, 08:57 PM
Tom, me 'ead's hurting!

Here's a thought....I'd be grateful if you could have a look at it - perhaps an EZNEC model to see how it compares to mine, which I can email you if you with. A vertical FULL wave dipole for 2m, fed 25% from the bottom end, with bottom of bottom leg 3m from ground. I've modeled this as bottom wire (wire 1) 0.485m long (1/4 wave), connected to a short wire (wire 2) of 0.03m with this short wire connected to the top dipole wire (wire 3) of 1.485m long (3/4 wave). The source is placed 50% up length of wire 2. This gives a full wave dipole at approx 145MHz, 580 ohm SWR somewhere around 1.7 (I haven't bothered "tuning" it). The elevation pattern shows RF at high angles as expected from a vertical full wave dipole. I've ignored feedline and am making an assumption that a good common mode choke would be in place at the feedpoint, and that it's not connected to any nearby ground or whatever.
<snip>

Performance (modeled) though is crap compared with a centre fed half wave vertical dipole at the same height, especially considering all the effort that goes into such an antenna. And with them being so small at VHF, I can't help but wonder why!

73

Pete

Pete,

The problem is when you fold the antenna that bottom area becomes a transmission line. The bottom section is no longer a half wave long, it is a 1/4 wave long transmission line.

The easiest way to visualize this is to picture a 1/2 wave antenna on one side, and a 1/4 wave antenna on the other side of an open wire line.

Now take the feedline and fold it to the short OCF dipole side. When you fold it that way, the one conductor can just be merged with the 1/4 wave OCF leg.

Now you see what you really have. The stub doesn't just feed the top element. It can't do that because any end-fed antenna always has to work against a second terminal. We can't force current into one terminal with no other connection.

So what we have is an OCF 3/4 wave long dipole, fed 1/4 wave up from the bottom or down 1/2 wave from the top. The "stub" is both a feedline and an antenna.

If we had a perfectly ground independent source like a really good current balun on v coax, we could attach it to the stub at a 50 ohm point and we would have the OCF dipole. The coax would not interact. We would also have to NOT parallel the stub with the coax, and we would have to not ground the bottom of the stub to a mast.

This is because the stub is also the 1/4 wave leg of the 3/4 wave OCF, so it is hot with RF.

If we break those rules then we start modifying the pattern and SWR from the 3/4 wave OCF dipole.

Now I agree with you about the end result. The wave angle is raised, the antenna becomes sensitive to what it is mounted on and how the coax is treated. This doesn't mean it won't "work", just that the real results depend on things we have no control over.

That's why they are not favored commercially even though they are easy to build. The ease and fun of building them makes them popular, the unpredictable nature keeps them out of commercial circles.

73 Tom

KC9CLM
12-02-2008, 09:19 PM
I too have built and had great success with the copper pipe j-pole. I also could not understand how the j-pole did not create a dead short across the coax, until it was explained in great detail to me. Here it is: Just remember that RF electricity is much smarter than regular electricity and therefore knows it should not act the same as regular electricity. All I can tell you is, it works!!

KC9CLM
Dave

K0CMH
12-02-2008, 09:53 PM
M3KXZ

I guess it is not the same on this side of the pond.

I travel extensively in the USA and have yet to see a wire center fed dipole in civilian, commercial use. They may be there, but I don't see them. Maybe others here will reply and "educate" me if they do commonly exist here.

Yes, the US Army and other US military branches do still have some HF center fed wire dipoles in the inventory, but most of the US military radio com is via space.

KA0GKT
12-03-2008, 02:27 AM
Actually, it is NOT a balun. A balun is a BALANCED to UN-BALANCED device. On the J-pole, you have the coaxial cable and the J-pole....both un-balanced. It is a COMMON-MODE choke. Prevents RF on the outer conductor of the coax. Best if you read up on the type of antenna in the ARRL antenna book. All the ones that I have built ...worked fine. In fact, I have some copper foil ribbon and plan to attach it to a window. (It has adhesive on one side.) You can "tap up" on the antenna (about two (2) inches from the "U" to achieve the lowest SWR at your design frequency. Would be nice to use an MFJ 259B antenna analyzer ....if you have one. 73 Joe W4CBJ

Actually, teh J pole is a balanced feed antenna just as a Zepp is a balanced feed antenna. The choke suggested for the coaxial transmission line is meant to act as a 1:1 choke balun. A person could do just the same thing by using a 4:1 coaxial balun and just tapping the "U" shaped portion at a different point...or more correctly, tapping it at the point where the short side of the "J" is 1/4-wave at the operating frequency (or an odd multiple thereof) and adjusting the shorted stub for a proper match.

Then again, you still have a zero dBd antenna which could have been made from a pair of 1/4-wave pieces of copper pipe and a plastic coupler...no soldering needed. Same gain, same radiation pattern, less cost.

All of this discussion for an antenna which, at VHF is mediocre at best...SHEESH!

M3KXZ
12-03-2008, 08:20 PM
<snip>
All of this discussion for an antenna which, at VHF is mediocre at best...SHEESH!

You may mock, but it's all worthwhile and educational!

Pete

KL7AJ
12-03-2008, 08:51 PM
M3KXZ

I guess it is not the same on this side of the pond.

I travel extensively in the USA and have yet to see a wire center fed dipole in civilian, commercial use. They may be there, but I don't see them. Maybe others here will reply and "educate" me if they do commonly exist here.

Yes, the US Army and other US military branches do still have some HF center fed wire dipoles in the inventory, but most of the US military radio com is via space.


How does 360 center fed wire dipoles grab ya?

http://www.haarp.alaska.edu/haarp/cam1.html

M3KXZ
12-03-2008, 09:17 PM
Pete,

The problem is when you fold the antenna that bottom area becomes a transmission line. The bottom section is no longer a half wave long, it is a 1/4 wave long transmission line.

Tom...sorry to say, but I've been thinking about all this again today while clearing leaves on the job! My thoughts are that it's wrong to treat this section as a transmission line. In EZNEC (and I assume other NEC models) a transmission line is incapable of radiating. But we know that this section of the antenna DOES radiate because it carries unequal currents. It carries currents for the 1/4 element connected one side, and for the 3/4 wave (or 1/4+1/2 wave) element connected to the other side. The lines are relatively close together and are parallel. Because the currents are not equal, theyu do not cancel. The sum of the currents results in this being a radiating section - something that the model can't show if this is modeled as transmission line.



The easiest way to visualize this is to picture a 1/2 wave antenna on one side, and a 1/4 wave antenna on the other side of an open wire line.

Now take the feedline and fold it to the short OCF dipole side. When you fold it that way, the one conductor can just be merged with the 1/4 wave OCF leg.

Now you see what you really have. The stub doesn't just feed the top element. It can't do that because any end-fed antenna always has to work against a second terminal. We can't force current into one terminal with no other connection.


This is where I have a problem with this section being modeled as transmission line. If it's modeled instead as 2 parallel wires (with segments lining up of course) with the source at the bottom, then it is shown to radiate. The source is able to put currents into both sides of the antenna - the 3/4 length side is driven against the 1/4 length side in the same way as a long element being driven against a short counterpoise (or conversely it could be seen as a short element being driven against a long counterpoise - either way doesn't matter). The fact that the currents in the section consisting of parallel 1/4 wavelength wires don't sum to zero is what gives rise to radiation in this section, and this is shown clearly by the model when modeled this way. This way, the issue of trying to force currents into a 1/2 wavelength section from the end of a transmission line, and then having to deal with where the opposing currents are, can be explained. To quote from your website "A feedpoint simply cannot force current into something without equal current flowing into something else (to push against) at the feedpoint." - well, by modeling the antenna as two separate wires (as opposed to a transmission line and a 1/2 wavelength wire) we can see that the short section and the long section push against each other.




So what we have is an OCF 3/4 wave long dipole, fed 1/4 wave up from the bottom or down 1/2 wave from the top. The "stub" is both a feedline and an antenna.

If we had a perfectly ground independent source like a really good current balun on v coax, we could attach it to the stub at a 50 ohm point and we would have the OCF dipole. The coax would not interact. We would also have to NOT parallel the stub with the coax, and we would have to not ground the bottom of the stub to a mast.

This is because the stub is also the 1/4 wave leg of the 3/4 wave OCF, so it is hot with RF.

If we break those rules then we start modifying the pattern and SWR from the 3/4 wave OCF dipole.


Agreed, the stub is both feedline and antenna. I think (although I may be wrong) that it's better to model it as antenna, knowing as we do that it radiates (so can't be modeled as a feedline that is unable to radiate). By keeping segments nicely aligned, and using adequate number of segments, then the calculations in EZNEC for closely spaced wires should be pretty good.

Also, it is fairly easy to do something daft like attached a supporting pole to one side of the antenna (like many j-pole installations) to see how it might screw things up. We can also continue the twin section downward and add a short to model the shorted tuning section, and again this will give an idea as to how this might (possibly in some minor way) affect the antenna.

QUOTE=W8JI;1412421]
Now I agree with you about the end result. The wave angle is raised, the antenna becomes sensitive to what it is mounted on and how the coax is treated. This doesn't mean it won't "work", just that the real results depend on things we have no control over.

That's why they are not favored commercially even though they are easy to build. The ease and fun of building them makes them popular, the unpredictable nature keeps them out of commercial circles.

73 Tom[/QUOTE]

EXACTLY!

Thanks again Tom and Steve - great discussion (which is keeping me away from things I should be doing!)

Pete

K0CMH
12-03-2008, 10:16 PM
7aj:

That's no fair.:D

G3TXQ
12-03-2008, 10:32 PM
This is where I have a problem with this section being modeled as transmission line.
Pete, I don't think Tom or I ever said it should be modelled as a transmission line. Certainly I have always modelled it as two parallel wires.

In the models I have run the currents in the bottom half of the stub have been very close to equal and antiphase. About halfway up the stub the common-mode current begins to become significant, and by the top of the stub it is dominant.

I still tend to view it as a transmission line stub unbalanced by some common-mode current - probably because the stub is primarily performing a matching function. But as I said earlier, we may all have different ways of looking at the same thing.

73,
Steve