J pole antenna for FM, impedances other than 50 ohm

[wa2ise]

Oh, there's tons of instructions on how to make a J pole antenna fed by 50 ohm coax. But there's absolutely nothing I can find if you want to use feedline of some other impedance, say 75 ohm coax, 300 ohm twinlead, 450 ohm ladderline, or even a twisted pair in ethernet cat 5 cable (100 ohms balanced, I ran some to extend a POTS phone line, and had a leftover twisted pair, and hey, Cat 5 runs in the VHF spectrum anyway, but not on this cat 5 cable or else the FM radio would suffer too much leakage). This must be classified or something...

As a scientific wild ass guess (SWAG) it looks like you can locate the feedpoints for these other feedline impedances by taking the square root of the ratio of (desired impedance/50 ohm), and taking that resulting number and multiply it by the length from the bottom (shorted end of the J pole) the instructions tell you to place the 50 ohm coax connections. It seems to work on a J pole I made for my FM broadcast receiver, though I realize there's a ton of variables, like receiver mismatches, stray conductors near the antenna and so on.

___x__________
|__x___________________________________________________

<--300 ohm ---><--------- 18-22 AWG wire-------------------->
twinlead

X's are where the feedline to the tuner connect.

The length of twinlead is 0.66 meters, the length of wire is 1.4 meters, and the feedline connects, measured from the left,
for 50 ohm coax, 5.8cm
if my SWAG is correct:
for 100 ohm twisted pair, like that in ethernet cat 5 cable, 8.2cm
for 300 ohm twinlead feedline, 14cm
for 75 ohm, 7.1cm

Both wires of the antenna's 300 ohm twinlead are shorted at the left end.

Yes, I know that all ham rigs are 50 ohm, and thus why noone ever says anything about any other impedance location. But if I wanted to make a J pole to receive say an FM broadcast station at 101.1 MHz and couple it to the 75 or 300 ohm FM tuner input, I'd want to know where these impedances exist on a J pole antenna.

 

[rcs16]

Hello,

I do not profess to be anything of an expect but I have an article about practical antenna construction/design. It is to big to load but I found it on the net.
practical_antenna_design.pdf
Published by LEDF Media. COPYRIGHT, 2000 by Elpidio Latorilla
Look at page 33, j-fed 1/2 wave.
You obviously have to tune the feedpoint to match your transmission line and load Z. A swr meter or better yet a vector/network analyzer which I myself have never done, would be used. HP/Agilent have app notes about the 8753b,c,d Network anaylzers, which i believe one would use for this application.
For un-balanced to balanced a balun can be used for this. Most receivers contain a balun, if they have a 300 ohm input. I have never used cat5 cable as a transmission line, but have used it for 10/100/1000 ethernet.
Would like to track how you resolve your issue.
Rick

 

[rcs16]

http://home.windstream.net/tanders/files/copperjpole.pdf
http://www.tgc-ares.org/DBJ-2 Roll up.pdf

 

[wa2ise]

As a scientific wild ass guess (SWAG) it looks like you can locate the feedpoints for these other feedline impedances by taking the square root of the ratio of (desired impedance/50 ohm), and taking that resulting number and multiply it by the length from the bottom (shorted end of the J pole) the instructions tell you to place the 50 ohm coax connections. ...

The length of twinlead is 0.66 meters, the length of wire is 1.4 meters, and the feedline connects, measured from the left,
for 50 ohm coax, 5.8cm
if my SWAG is correct:
for 100 ohm twisted pair, like that in ethernet cat 5 cable, 8.2cm
for 300 ohm twinlead feedline, 14cm
for 75 ohm, 7.1cm

Looks like my SWAG is off, that it's linear and not square root. According to

The impedance at the bottom of the stub is approx 0 ohms, and very high at the top end where it is an open circuit. In the middle the stub is equal to its characteristic impedance. (paraphase here) which for twinlead is 300 ohms.

http://home.windstream.net/tanders/files/copperjpole.pdf

So now it looks like you can locate the feedpoints for these other feedline impedances by taking the ratio of (desired impedance/50 ohm), and taking that resulting number and multiply it by the length from the bottom (shorted end of the J pole) the instructions tell you to place the 50 ohm coax connections. NO square rooting.

So now:

The length of twinlead is 0.66 meters for the stub, the length of wire is 1.4 meters, and the feedline connects, measured from the left,
for 50 ohm coax, 5.8cm
if my NEW SWAG is correct (for a stub made out of 300 ohm twinlead):
for 100 ohm twisted pair, like that in ethernet cat 5 cable, 11.6cm
for 300 ohm twinlead feedline, 33cm
for 75 ohm, 8.7cm

 

[JBL GUY]

I was going to ask you if you had measured the return loss based on your original calculations.

 

[for_p1]

The proper solution is using impedance transformer right near antenna itself and then using the rest of connection to tuner with coax cable of whatever impedance your tuner has (usually either 50 or 75 ohm). Impedance transformer can be made with losses low enough not to have any influence on reception quality.

 

[JBL GUY]

The proper solution is using impedance transformer right near antenna itself and then using the rest of connection to tuner with coax cable of whatever impedance your tuner has (usually either 50 or 75 ohm). Impedance transformer can be made with losses low enough not to have any influence on reception quality.

I am not sure I follow your reasoning.

If the feed line impedance is 75 Ohms, 100 Ohms or 300 Ohms, that would be the impedance of one side of the transformer, what would be the impedance of the other side of the transformer?

If one can obtain return losses of greater than 20 dB, why is additional impedance transformation needed?

J-Pole antennae may benefit from additional RF decoupling of the feed line.

 

[for_p1]

I am not sure I follow your reasoning.

If the feed line impedance is 75 Ohms, 100 Ohms or 300 Ohms, that would be the impedance of one side of the transformer, what would be the impedance of the other side of the transformer?

If one can obtain return losses of greater than 20 dB, why is additional impedance transformation needed?

J-Pole antennae may benefit from additional RF decoupling of the feed line.

Most tuners have either 50 or 75 (more common) ohm impedance. Then you get the best low loss coax cable (again 75 or 50 ohm is cheapest). Any transformers are best use where signal/noise ratio is highest - that means on the pole near antenna itself. There are almost no antenna designs with exact own impedance at 50 or 75 ohms, so you need transformer anyway.

 

[JBL GUY]

Most tuners have either 50 or 75 (more common) ohm impedance. Then you get the best low loss coax cable (again 75 or 50 ohm is cheapest). Any transformers are best use where signal/noise ration is highest - that means on the pole new antenna itself. There are almost no antenna designs with exact own impedance at 50 or 75 ohms, so you need transformer anyway.

You still did not answer my question, what is the impedance of the transformer on the antenna side?

One of the advantages of a J-Pole antenna is the fact that it is possible to find a feed point with a return loss of 20 dB or greater.

How can adding a matching transformer improve on this?

Again, you did not answer my question, what is the impedance of the transformer on the antenna side?

BTW, I am not talking about feed line loss, I am talking about return loss, two entirely different things.

 

[for_p1]

Proper transformer has very low insertion loss (like 10-15%). The goal is to have low loss, low reflection (impedance of transformer matches impedance of antenna as close as possible), and lowest noise level added by transformer.

 

[JBL GUY]

Proper transformer has very low insertion loss (like 10-15%). The goal is to have low loss, low reflection (impedance of transformer matches impedance of antenna as close as possible), and lowest noise level added by transformer.

Again you did not answer my question.

If one can pick a feed point on the J-Pole antenna that achieves a return loss of 20 dB or greater, how is a matching transformer going to improve on this?

Do you understand the term "return loss"?

A 20.8 dB return loss is a mismatch loss of about 0.8 percent which is less than 0.1 dB loss, considerably less than your 10 to 15 percent transformer insertion loss.

If one has an antenna with an odd ball feed point impedance, and no other matching solution is available then a matching transformer might be considered.

But, again one would have to know this impedance and it doubtful that an off the shelf transformer of the correct impedance would be available.

In general in our hobby we will be looking at antennae with feed point impedances of 75 Ohms or 300 Ohms and a slight mismatch on either side of these impedances will not result in a large loss of signal.

 

[wa2ise]

I was going to ask you if you had measured the return loss based on your original calculations.

I'm not sufficiently knowledgeable in antenna theory and equipment capable to do that, though I'd imagine that there is significant loss in my run of feedline (40 feet of an ethernet cat 5e twisted pair, 100 ohm impedance, and that connected to 20 feet of RG59 to the receiver, with the ethernet twisted pair looped thru a ferrite bead several times to act as a balum at that connection to the coax. Yes, awful, but I'd dealing with hard to change out cable runs inside walls of the house.). Even with this crappy setup, I still am getting pretty good reception of all the local NYC FM stations, and most of the 2nd adjacents between the NYC stations. Jpoles are supposed to have fairly narrow bandwidths, something like 1 1/2 MHz at 100MHz, but my system is sufficiently bad to make that cover most of the FM b'cast band.

 

[WobblySam]

With no disrespect to the other posters, a 75Ohm to 50Ohm mismatch for a receiving antenna is of little concern. Your use of cat5 connected to rg-59 is far worse. You have a couple of serious impedance bumps that are effectively unknown. Connecting some good RG-6 directly to your J-pole (even with a nominal Z of 50OHMs) and running that directly to your tuner will be a much better solution. Return loss is not much of a concern with an omni receive antenna - given that you'll most likely experience more severe effects from multipath.

From your description of what you have to work with (cat5 situation), I realize the use of proper feedline might not be possible. With your current feedline situation, you really don't need to worry much about the 50Ohm Z of your antenna.

Good luck

 

[JBL GUY]

Yes, as I pointed out.

A 20.8 dB return loss is a mismatch loss of about 0.8 percent which is less than 0.1 dB loss, considerably less than a 10 to 15 percent transformer insertion loss.

For those that are interested.

I'm not sufficiently knowledgeable in antenna theory and equipment capable to do that, though I'd imagine that there is significant loss in my run of feed line (40 feet of an Ethernet cat 5e twisted pair, 100 ohm impedance, and that connected to 20 feet of RG59 to the receiver, with the Ethernet twisted pair looped thru a ferrite bead several times to act as a balm at that connection to the coax. Yes, awful, but I'd dealing with hard to change out cable runs inside walls of the house.). Even with this crappy setup, I still am getting pretty good reception of all the local NYC FM stations, and most of the 2nd adjacents between the NYC stations. Jpoles are supposed to have fairly narrow bandwidths, something like 1 1/2 MHz at 100MHz, but my system is sufficiently bad to make that cover most of the FM b'cast band.

Return loss is the loss of signal power resulting from the reflection caused at a discontinuity in a transmission line.

This discontinuity can be a mismatch with the terminating load or with a device inserted in the line. It is usually expressed as a ratio in decibels (dB).

Two lines or devices are well matched if the return loss is high. A high return loss is therefore desirable as it results in a lower insertion loss.

It is another method of expressing VSWR.

I was curious as to how the OP's originally calculated impedance tap point for a J-POLE antenna agreed with his measured results, it could be in terms of return loss or VSWR.

J-POLE antennae can be good performing antennae, but there is a fair amount of misinformation regarding their construction.

Of course this is true with the antennae that we as amateur radio operators use.

They can be somewhat hit or miss, especially when built for reception only and if the person does not have a method of verifying their performance.

I designed and built a number of J-POLE antennae for use with my 2 meter repeater, including multiple element collinear types. They were not necessarily better than other collinear gain type vertical antennae, but they could be built very ruggedly and having the radiating element at DC ground was something that I liked. I also paid attention to feed line decoupling.

I am a big fan of making whatever antenna I am working with resonant at the frequency of interest and then working to achieve a good match to the feed line.

In terms of our hobby of FM broadcast reception and given the fact that most people live in strong signal areas, attention to these small details may not make a noticeable difference.

There is another currently running thread on AK discussing antennae for over the air digital TV reception. This is an area where attention to small details may be worth the effort.

Due to the nature of digital TV reception it works or it does not and a small increase in delivered signal strength may make the difference of a watchable picture or no picture at all (the digital TV cliff as it is called).

The OP was just trying to give a method of finding the correct feed point tap location for a J-POLE antenna for a given feed point impedance and I was curious about his results.

BTW, in general, connecting a 75 Ohm feed line to an impedance of 50 Ohms would result in a signal loss of 4 percent or about 0.2 dB and the return loss would be about 14 dB or in terms of VSWR it would be 1.5:1.

Not much signal loss at all.

 

[rcs16]

Hi JBL,

Can you help me/us construct a j-pole antenna for my testing.
Maybe both methods, std copper pipe for plumbing and one using 300ohm balanced, std steel copper clad.
I have designed a FM radio using the Si4735D60 part.
It works really well ,when I connect it to either.
1) a deep fringe log-periodic with a amp up on the tower, 300 ohm feed (~50ft) , through another amp, the to RG-6/U straight into the Si4735D60. I can receive stations 75-100 miles away.
2) A standard folded dipole that they sell with a old stereo. This is only good for local stations. Have not mounted it up high.
I can start a new thread since I do not want to interfere with the original OP.
I have a HP 8556B, but do not have a return loss bridge. I also have a Hp 436B as well and a 8484A sensor, 50-75 matching pad ...
I will look for the thread on the HDTV

Thanks
Rick

 

[wa2ise]

With no disrespect to the other posters, a 75Ohm to 50Ohm mismatch for a receiving antenna is of little concern. Your use of cat5 connected to rg-59 is far worse. ...
From your description of what you have to work with (cat5 situation), I realize the use of proper feedline might not be possible. With your current feedline situation, you really don't need to worry much about the 50Ohm Z of your antenna.

Good luck

I though I could "get away with" the mismatch of 75 ohm coax to 100 ohm balanced pair, with a common mode choke coil, the twisted pair threaded thru a ferrite bead several times, located at the junction of that coax to the twisted pair. As the mismatch, once the balanced to unbalanced issue taken care of, I'd have a SWR of 1.333 to 1. Which is a little better than 50 ohm to 75 ohm, that swr 1.5 to 1.

BTW, in general, connecting a 75 Ohm feed line to an impedance of 50 Ohms would result in a signal loss of 4 percent or about 0.2 dB and the return loss would be about 14 dB or in terms of VSWR it would be 1.5:1.

If so, then my horrible setup is doing reasonably well.

Wondering how to determine how to attach a feedline of some other impedance to the typical Jpole was a question I could not find anywhere. Out of seemingly hundreds of sets of instructions on how to build one. I'm fairly ignorant of antenna theory, and figured I'd get an education from my posts... Though I'm still not sure I learned what I was after. :scratch2:

 

[ConradH]

JBL Guy, (if you have it, and you probably do) Jerry Sevick's book on baluns, ununs and transmission line transformers seems to show how to match most anything to most anything, but it seems mainly aimed at transmitting. Are those techniques just as good for receiving?

 

[rcs16]

Wondering how to determine how to attach a feedline of some other impedance to the typical Jpole was a question I could not find anywhere

I agree, lots of construction details, except for calculating the position of the feedline point to match to the tline/load. It would be nice to have reference to some text containing that info.
Even have me reading my old communications systems book from my college days, getting back into smith charts again.:scratch2:
The problem with the internet = close but no cigar.
I was looking around yesterday, maybe the ARRL antenna handbook or the reference that ConradH mentions.
I will keep looking and let you know what I come up with.
Could try asking folks on diyaudio as there seems to be more EE types on that forum or some amateur radio forums.
How about trying edison_fong at hotmail dot com.? the author of the DBJ2 article that I referenced.
If not just keep experimenting. As it is even though the math will give you the location, you still have to tune it manually to get the best match.

Are those techniques just as good for receiving?

My "Electronics Communications Systems" book by Kennedy says that for antenna's "Apart from their different functions,transmitting and receiving antennas behave identical. That their behavior is reciprocal"

Good luck
Rick

 

[wa2ise]

I agree, lots of construction details, except for calculating the position of the feedline point to match to the tline/load. It would be nice to have reference to some text containing that info.
...
The problem with the internet = close but no cigar.
...

My "Electronics Communications Systems" book by Kennedy says that for antenna's "Apart from their different functions,transmitting and receiving antennas behave identical. That their behavior is reciprocal"

Reminds me when I was a kid reading about astronomy, solar system and such. All the books I could find at the local library said that the further out a planet was from the Sun the longer its year. But not one book showed the equation you could use to calculate this. This ran contrary to teachers saying that you could find out anything you wanted to know by going to the library... :thumbsdn:

As for antennas receiving as well as transmitting, a VHF dipole and a UHF dipole will transmit equally well, but the UHF dipole will grab less signal than the VHF dipole will, as the UHF antenna is smaller. A reason why UHF TV stations used more transmit power than VHF stations.

 

[Ed in Tx]

:scratch2: Could just hook it up and adjust the feed points for maximum signal strength on a weak station. That'll get it pretty close for receiving purposes. I do something like that when I initially tune up my ham rig on one of the HF bands especially on 160m where my antenna has a very narrow bandwidth and requires re-adjusting the tuner if I go much off where it's tuned. I start by adjusting my antenna tuner for maximum peak receiver noise, then when I transmit a few Watts test signal for final adjustment it's usually a very close match just a little more tweaking to near 1:1 VSWR. Of course that's too easy!