PDA

View Full Version : Dipole: calculate/measure impedance?



AE7F
12-21-2010, 04:45 PM
I need to figure out impedance for a dipole so that I can match the feed line.

The most likely scenario is a half-wave dipole cut for 3.750mhz in an inverted V configuration with the apex (starting) at 20ft and the ends are about 6ft off the ground. Antenna material is solid copper, I think 16 or 14 gauge.

Most figures I've seen state 73 ohms in free space. I don't know the soil condition at my QTH (other than alkaline clay) but I do know antenna height above ground. If I'm not mistaken, raising the antenna higher will increase impedence. Is that correct or do I have it backwards?

If the dipole was only 73 ohms, wouldn't you just use an SO-239 and then connect 75ohm cable between the SO-239 and the antenna tuner? The feedline length would be short (close to the shack) and frequency is low, so would this be an acceptable feed?

I'm guessing the impedance will be higher than 73ohms and I have 300ohm twin lead instead of coax, so I'm wondering what I can do to use it instead. To figure that out, I need to be able to calculate antenna impedance, right?

WB2WIK
12-21-2010, 04:53 PM
Nothing on 80m is in free space unless you have a couple of 1000 foot towers.

At 20' your 80m dipole should have a feedpoint impedance of about 23+j39 Ohms.

Not higher than 70 Ohms, lower, and quite a bit lower.

You'd have a closer match with 50 Ohm coax than with 75 Ohm.

K8JD
12-21-2010, 04:56 PM
An 80M dipole only 20 ft up will resonate a LOT lower in freq from your calculated length because of capacitive coupling to earth. you will need to trim it a foot at a time (equally lfrom lboth ends) and keep checking the SWR until it is reasonable on your preferred freq.
I recall my 160M dipole at 35 ft started with the formula length for 1.800 and I ended up cutting about 12 ft from each end to get it up to the bottom end of the band.
Generally 50 ohm Coax is the type to use.
Also , cutting it for the middle of the band will NOT give you an acceptable SWR on the high and lowest ends of 75/80 M, If you like CW , cut for the middle of the CW band etc...Good luck.

VK1OD
12-21-2010, 05:06 PM
I need to figure out impedance for a dipole so that I can match the feed line.

The most likely scenario is a half-wave dipole cut for 3.750mhz in an inverted V configuration with the apex (starting) at 20ft and the ends are about 6ft off the ground. Antenna material is solid copper, I think 16 or 14 gauge.

Most figures I've seen state 73 ohms in free space. I don't know the soil condition at my QTH (other than alkaline clay) but I do know antenna height above ground. If I'm not mistaken, raising the antenna higher will increase impedence. Is that correct or do I have it backwards?

If the dipole was only 73 ohms, wouldn't you just use an SO-239 and then connect 75ohm cable between the SO-239 and the antenna tuner? The feedline length would be short (close to the shack) and frequency is low, so would this be an acceptable feed?

I'm guessing the impedance will be higher than 73ohms and I have 300ohm twin lead instead of coax, so I'm wondering what I can do to use it instead. To figure that out, I need to be able to calculate antenna impedance, right?

The impedance of the antenna you describe is likely to be lower than the free space impedance, quite likely between 50Ω and 75Ω at the feedpoint at resonance.

If that is the case and you were to feed it with 50Ω coax and an effective balun, you would expect that VSWR(50) at its minimum would be between 1 and 1.5. Again, if that was the case, you could infer the R at resonance (very close to minimum VSWR) as approximately VSWR(50)*50Ω.

Have you tried feeding the antenna with 50Ω coax and an effective balun? What VSWR did you measure?

Owen

KA5S
12-21-2010, 05:17 PM
Take a look at figure 2 in W4JOQ's new book, Type S Series-Section Broad-banding of 75- to 80-meter Antennas, published by Antennex. 2009 PREVIEW (http://www.antennex.com/preview/Apr409/type_s_bb.pdf)

Note what has to be done to cover the whole 80/75 meter band.

Cortland
KA5S

AE7F
12-21-2010, 07:11 PM
Take a look at figure 2 in W4JOQ's new book, Type S Series-Section Broad-banding of 75- to 80-meter Antennas, published by Antennex. 2009 PREVIEW (http://www.antennex.com/preview/Apr409/type_s_bb.pdf)

Note what has to be done to cover the whole 80/75 meter band.

Cortland
KA5S

Wow. That is fantastic information. Thanks.

AE7F
12-21-2010, 07:12 PM
The impedance of the antenna you describe is likely to be lower than the free space impedance, quite likely between 50Ω and 75Ω at the feedpoint at resonance.

If that is the case and you were to feed it with 50Ω coax and an effective balun, you would expect that VSWR(50) at its minimum would be between 1 and 1.5. Again, if that was the case, you could infer the R at resonance (very close to minimum VSWR) as approximately VSWR(50)*50Ω.

Have you tried feeding the antenna with 50Ω coax and an effective balun? What VSWR did you measure?

Owen

Thanks Owen. I will see if I can find the best place to obtain some 50ohm cable and let you know how it goes.

AE7F
12-21-2010, 07:13 PM
Nothing on 80m is in free space unless you have a couple of 1000 foot towers.

At 20' your 80m dipole should have a feedpoint impedance of about 23+j39 Ohms.

Not higher than 70 Ohms, lower, and quite a bit lower.

You'd have a closer match with 50 Ohm coax than with 75 Ohm.

Thanks for the help.

AE7F
12-21-2010, 07:15 PM
An 80M dipole only 20 ft up will resonate a LOT lower in freq from your calculated length because of capacitive coupling to earth. you will need to trim it a foot at a time (equally lfrom lboth ends) and keep checking the SWR until it is reasonable on your preferred freq.
I recall my 160M dipole at 35 ft started with the formula length for 1.800 and I ended up cutting about 12 ft from each end to get it up to the bottom end of the band.
Generally 50 ohm Coax is the type to use.
Also , cutting it for the middle of the band will NOT give you an acceptable SWR on the high and lowest ends of 75/80 M, If you like CW , cut for the middle of the CW band etc...Good luck.

Good advice. What do you mean by "acceptable SWR?" Do you mean an SWR that cannot be corrected with an MFJ-949e or an SWR that cannot be used without a tuner?

WB2WIK
12-21-2010, 07:20 PM
Good advice. What do you mean by "acceptable SWR?" Do you mean an SWR that cannot be corrected with an MFJ-949e or an SWR that cannot be used without a tuner?

Typical 80m dipole SWR curve (no special techniques, just a regular 1/2-wave dipole). In this example, the dipole is cut for 3.8 MHz.

Your tuner should be able to make it work across the whole band, but it will match best over a narrow frequency range, about 150 kHz or so. The 80m band is 500 kHz wide...

AE7F
12-21-2010, 08:06 PM
Typical 80m dipole SWR curve (no special techniques, just a regular 1/2-wave dipole). In this example, the dipole is cut for 3.8 MHz.

Your tuner should be able to make it work across the whole band, but it will match best over a narrow frequency range, about 150 kHz or so. The 80m band is 500 kHz wide...

Okay, that's what I was thinking also.

After looking at alot of different options, I have isolated some good possibilities:
------------------------------------------------
-If I am going to use 300ohm twin lead feed line:

A. Use a folded dipole design instead, which would much more closely match the impedance of the feed line (there are many sources I found, but here are a couple: http://www.k7mem.150m.com/Electronic_Notebook/antennas/folded_dipole.html, and http://forums.qrz.com/showthread.php?t=137238).

A Pros:
Close match to existing twin lead feed
Wide bandwidth?

A Cons:
Have to buy more wire
Have to buy spacers / other stuff
Less aesthetic / stealthy antenna design

B. As a special case, use a 1/2 wavelength section of the twin lead feeder so that the feed line is transformed to ~50ohms (http://forums.qrz.com/showthread.php?t=276969 post #3).

B Pros:
Closer impedance match

B Cons:
Might have to buy more twin lead
Too much feed line - what to do with it?

------------------------------------------------
-If I am going to use coax cable:
A. Eliminate the 300ohm feed line; install so-239 adapter; connect antenna to tuner via either 35ohm or 50ohm coax cable

A Pros:
Closer impedance match
Easy coupling (coax couplers)
Coax more durable to weather than twin lead
Maybe use twin lead for a separate folded dipole antenna design later

A Cons:
Have to buy so-239
Have to modify existing antenna
Have to buy coax cable

What IS evident right now is the fact that my current dipole is not matched well at all with the 300ohm twin lead feed line!

VK1OD
12-21-2010, 08:15 PM
...
What IS evident right now is the fact that my current dipole is not matched well at all with the 300ohm twin lead feed line!

That little snippet thrown in at the end is highly relevant.

Yes, your dipole will cause high standing waves on the 300 ohm line, and the impedance at the transmitter end is very dependent on length and frequency.

You could transform it to deliver a 50 ohm load to the transmitter using an ATU.

Another option is to use an integral number of electrical half waves which will deliver an impedance to the tx close to the feed point impedance, and probably good enough for the tx, but keep in mind that his 'tuned feeder' configuration narrows bandwidth... possibly an issue for you on 80m.

You could feed the antenna with coax, and again the load presented to the tx should be good enough, but could be improved with an ATU. More importantly, the ATU would give you better frequency coverage over the band. In this case, the coax VSWR will increase towards band edges, but for practical coax types and length, the loss should be acceptable.

Owen

K9FV
12-21-2010, 08:53 PM
"IF" you are looking for 80 meters only, then just get some coax, either 50 ohm or 75 (72?) ohm - doesn't really matter much which it is, what ever you can snag cheap. Connect the coax to the dipole by twisting the shield to one side, center conductor to other side, run it back to your 949 tuner, then to rig. You're good to test.

That should allow you to cover whole 80 meter band - try that a couple of days and if it works pretty good, go back and make good water tight connections.

Decisions based on YOUR experience at YOUR qth combined with knowledge gained from suggestions and info here will teach you more than reading alone. Book learning is good, but it needs to be seasoned with OJT also.

73 de Ken H>

AI3V
12-22-2010, 02:41 PM
A really, really, REALLY good book written just about this subject used to be avalible fron the ARRL:

Antenna Impedance Matching, by W. Carron.

The basic drill:

Install antenna
MEASURE antenna impedance
Plot impedance vs frequency on Mr Smith's chart
Refer to book for suggestions on what matching circuit will match what impedance
Build matching circuit
Install matching circuit
MEASURE the resulting VSWR

Adding and removing antenna SYSTEM components randomly, will lead to random results.

Rege

KL7AJ
12-22-2010, 02:47 PM
There's a handy-dandy little graph that's been in the Antenna Book since the dark ages that plots dipole impedance as a function of height over ground. If your ground is really lousy (as it is here) your antenna impedance can actually INCREASE as you get close to the ground, but this is mostly due to increased losses. Over perfect ground, the impedance approaches 0 as the height approaches 0


Eric

AE7F
12-22-2010, 05:23 PM
Thanks for the tips. I probably need more knowledge of electronics to understand things the same way many of you do. I did find some useful links while exploring the Smith Chart ideas:
http://www.karinya.net/g3txq/ant_impedance/
http://fermi.la.asu.edu/w9cf/tran/

Sometimes you just need to figure out a good search term to use. For those links, I searched "how to measure antenna impedance."

WB2WIK
12-22-2010, 07:18 PM
Height above ground vs. dipole feedpoint impedance: 40m dipole plotted below.

For 80m, just double all the heights shown.

AE7F
12-22-2010, 07:24 PM
Another interesting link:
http://www.antenna-theory.com/measurements/impedance.php

AE7F
12-22-2010, 07:31 PM
Height above ground vs. dipole feedpoint impedance: 40m dipole plotted below.

For 80m, just double all the heights shown.

Really nice. How was this generated? I would like to see this for different configurations.

Taking your advice and doubling the X axis height scale on the impedance graph on the right above while assuming "average" ground condition would put an 80m dipole at 20ft at between 50-60ohms impedance. However, poor ground would plot impedance around 70. Is there an existing chart for an 80m dipole?

Naturally, this leads me to start looking into my ground conditions.... wonder how I can estimate that...

If I can even estimate pretty closely, I can choose either 75ohm or 50ohm coax and give it a try.

WB2WIK
12-22-2010, 07:40 PM
Really nice. How was this generated? Can it be done for an 80m dipole?

Naturally, this leads me to start looking into my ground conditions.... wonder how I can estimate that...

You don't need to do a separate one for an 80m dipole. Change all the "heights in feet" to double that shown, and the same graph applies to an 80m dipole. It's all based on height in wavelengths and applies to a broad range of frequencies.

I don't know what my soil conductivity is, either, so I just use the center (average) line as a guideline: The variation from "perfect" soil to "very poor" soil, once you get a quarter wavelength high, is very small and eventually becomes zero as you get a couple of wavelengths above ground.

WB2WIK
12-22-2010, 07:45 PM
Naturally, this leads me to start looking into my ground conditions.... wonder how I can estimate that...



It's very site specific; however a "general trend" map of soil conductivity for all of the U.S. is on the FCC website, here: http://www.fcc.gov/mb/audio/m3/index.html

Your area in southwest Utah is at the "very good" conductivity end of the range, showing between 15 and 30 millimhos/meter. That's really high, and above average but certainly nowhere near "perfect." Sea water is probably close to the "perfect" end of the scale.

K1DNR
12-22-2010, 08:44 PM
never mind...

VK1OD
12-22-2010, 10:05 PM
...
Naturally, this leads me to start looking into my ground conditions.... wonder how I can estimate that...


I think you have been talking about a half wave dipole for 80m only, ie not a multiband antenna.

There are many ways to estimate ground parameters, but remember that ground parameters vary seasonally (eg with rainfall).

Ground parameters are not your primary objective, your primary objective to to understand the feed point impedance of your dipole as rigged, ground, nearby structures, vegetation etc for your own scenario... then to design a matching scheme (if needed).

Now that you know that it is highly likely that the R component of the dipole is likely to be in the range of 50 to about 80 ohms, if an exact match is your target, you could achieve that with a very simple scheme.

It is a characteristic of a half wave dipole that near resonance, R varies slowly with f, and X quickly with f.

If you have a dipole with an R between about 50Ω and 110Ω, you can transform it to 50+j0Ω by:


connecting a length of 75Ω coax;
adjust the length the dipole (high or low) until the VSWR on the 75Ω line is 1.5;
cut the 75 ohm line at the length were the impedance looking into the line is 50+j0.

For example, if the dipole happened to be 60+j0 at resonance (though you don't need to know this), and you length it until VSWR(75) is 1.5, Z at the feedpoint will be 60-j24Ω (though you don't need to know that), and the length looking into 35 of the 75Ω line will be very close to 50+j0Ω.

Owen

VK1OD
12-22-2010, 10:46 PM
...
For example, if the dipole happened to be 60+j0 at resonance (though you don't need to know this), and you length it until VSWR(75) is 1.5, Z at the feedpoint will be 60-j24Ω (though you don't need to know that), and the length looking into 35 of the 75Ω line will be very close to 50+j0Ω.


That should read ...and you adjust the length by shortening it until VSWR(75) is 1.5...

Apologies.

Owen

AE7F
12-22-2010, 11:59 PM
I think you have been talking about a half wave dipole for 80m only, ie not a multiband antenna.

There are many ways to estimate ground parameters, but remember that ground parameters vary seasonally (eg with rainfall).

Ground parameters are not your primary objective, your primary objective to to understand the feed point impedance of your dipole as rigged, ground, nearby structures, vegetation etc for your own scenario... then to design a matching scheme (if needed).

Now that you know that it is highly likely that the R component of the dipole is likely to be in the range of 50 to about 80 ohms, if an exact match is your target, you could achieve that with a very simple scheme.

It is a characteristic of a half wave dipole that near resonance, R varies slowly with f, and X quickly with f.

If you have a dipole with an R between about 50Ω and 110Ω, you can transform it to 50+j0Ω by:


connecting a length of 75Ω coax;
adjust the length (high or low) the dipole until the VSWR on the 75Ω line is 1.5;
cut the 75 ohm line at the length were the impedance looking into the line is 50+j0.

For example, if the dipole happened to be 60+j0 at resonance (though you don't need to know this), and you length it until VSWR(75) is 1.5, Z at the feedpoint will be 60-j24Ω (though you don't need to know that), and the length looking into 35 of the 75Ω line will be very close to 50+j0Ω.

Owen

Very interesting, and I appreciate your help. I'm trying to understand your comments. Would you care to correct me as I restate/question you:

connect a length of 75Ω coax (let's say 100ft, common pre-cut length);
shorten the length of the dipole until the VSWR on the 75Ω line is 1.5 (on your antenna tuner/meter with the switch set in Bypass mode);
cut the 75 ohm line at the length were the impedance looking into the line is 50+j0.


Not sure I understand step 3 other than the fact that we will be adjusting the feed coax length. How do I know when impedence is 50+j0?

VK1OD
12-23-2010, 12:13 AM
Very interesting, and I appreciate your help. I'm trying to understand your comments. Would you care to correct me as I restate/question you:

connect a length of 75Ω coax (let's say 100ft, common pre-cut length);
shorten the length of the dipole until the VSWR on the 75Ω line is 1.5 (on your antenna tuner/meter with the switch set in Bypass mode);
cut the 75 ohm line at the length were the impedance looking into the line is 50+j0.


Not sure I understand step 3 other than the fact that we will be adjusting the feed coax length. How do I know when impedence is 50+j0?

For step 2 you use a 75Ω VSWR meter. They used to be common, there were also models with switchable terminations to cal for 50Ω and 75Ω.

For step 3 you use a 50Ω VSWR meter. You will probably chop a bit of coax up doing this, but then RG6 is pretty cheap.

Once you find the 50Ω point, connect that point to the tx, or extend it with any length of 50Ω coax.

Bowral 7MHz dipole for local contacts (http://www.vk1od.net/antenna/7MDipole/7MDipole02.htm) describes one form of the technique, An antenna for 7MHz local contacts (http://vk1od.net/antenna/7MDipole/7MDipole.htm) describes another. There is differing level of explanation of how it works in the two articles.

Owen

W4PG
12-23-2010, 12:36 AM
The most likely scenario is a half-wave dipole cut for 3.750mhz in an inverted V configuration with the apex (starting) at 20ft and the ends are about 6ft off the ground.

Your antenna is going to be seriously detuned this low to the ground. I'd really try to find some way to get it higher. It's not going to work well regardless of what you feed it with. Of course, "well" is relative.

I've had no problem feeding 1/2 wave dipoles with 50 ohm coax and achieving a near 1:1 match, with them well up in the air! There is no such thing in the real world as a dipole in free space, which theoretically has an impedance of 72 ohms. Putting the dipole on Planet Earth above ground reduces the impedance such that 50 ohm coax make a nice feedline.

However, I'm now a fan of folded dipoles and currently have one for 80 meters made from 450 ohm twinlead and fed with 300 ohm twinlead to a 6:1 balun with a short run of coax.

It has about 300 kHz 2:1 SWR bandwith. The extra bandwith is a nice characteristic of the folded dipole. My SWR meter reads 1:1 for about 50 kHz. Really nice!

.............Bob

AE7F
12-24-2010, 03:10 AM
In short order, I will have a length of both 50ohm coax and 450ohm twin lead cable. I already have a length of 300ohm twin lead. I should be able to get something working out of this.

Last night I checked and my 80m dipole plus twin lead feeder at around 3.8mhz was like 5:1 swr. I'm going to try using the 50ohm coax and see what happens.

Should I implement an so-239 connector at the antenna or just splice right to the coax?

AE7F
01-01-2011, 04:05 AM
In the course of maybe having either a radio or antenna tuner issue, I gave up and went with the G5RV spec of 51x51 with 28ft of 300ohm twin lead and then coax to the radio. Initial test on 80 was very good. Now that I know the antenna design is good, I'll look into other things. I can always clip on extra length for 80m.