20M Struggle

[flintstone mop]

Hello everyone,
I am trying to get my station up on 20M and with some hit and miss tactics, got my tuner to give me a 1.2:1 match to the dipole. I am only seeing 1/2 amp per leg of antenna current, 200 watt carrier. A little bit of heating on the clamps of the capacitor. I'm using the famous balanced tuner in the parallel config. Series gets me thinking that is for a short dipole.

20M is a quirky band and calling CQ didn't get any attention. After field day, conditions changed and it got really quiet around 8PM. I was troubleshooting distortion in my monitoring system and spinning wheels wasting time on that, when it turned out to be an overloaded receiver.

I sent an email to Tom Vu with this info, and I forgot he is recovering his station from possible tornado damage or a microburst.

My concern is about the RF current. It seems a little light. Dipole is 70 feet in the air and 180 feet long, fed by ladder line, approx 120 feet. On 160M, I see 3 amps of RF current with 300watt carrier.

There's a lot of things that come into play here. Frequency, height above ground, etc, etc, that determine antenna current.
Should I keep trying to call CQ and hope it's working?




I'll keep checking here for any guidance

Fred

[w4bfs]

170 ft for the radiator sounds in the neighborhood of 3 wavelengths on 20 mtr .... if radiating conditions remain balanced, the feedpoint is at high impedance .... 120 ft for the feedline sounds in the neighborhood of 2 wavelengths and will tend to not transform impedance .... your observed feedline currents tend to support this since high impedance suggests by Ohm's law high voltage and low current ... ignoring feedline velocity factor ( which is .94 or so for ladder line ), try adding or taking away 16 ft of feedline and see if your feedline current changes to a higher value ... 73   John

[K5UJ]

Hello everyone,
I am trying to get my station up on 20M and with some hit and miss tactics, got my tuner to give me a 1.2:1 match to the dipole. I am only seeing 1/2 amp per leg of antenna current, 200 watt carrier. A little bit of heating on the clamps of the capacitor. I'm using the famous balanced tuner in the parallel config. Series gets me thinking that is for a short dipole.

20M is a quirky band and calling CQ didn't get any attention. After field day, conditions changed and it got really quiet around 8PM. I was troubleshooting distortion in my monitoring system and spinning wheels wasting time on that, when it turned out to be an overloaded receiver.

I sent an email to Tom Vu with this info, and I forgot he is recovering his station from possible tornado damage or a microburst.

My concern is about the RF current. It seems a little light. Dipole is 70 feet in the air and 180 feet long, fed by ladder line, approx 120 feet. On 160M, I see 3 amps of RF current with 300watt carrier.

There's a lot of things that come into play here. Frequency, height above ground, etc, etc, that determine antenna current.
Should I keep trying to call CQ and hope it's working?




I'll keep checking here for any guidance

Fred

Hi Fred,

You're probably getting out, it's just that AM on 20 may not be all that common.  I know some of the guys are working AM on 20, like Jack but I don't know if there's a lot of activity there yet so that may be part of the reason for the low response.

I don't know what you mean by "famous balanced tuner" so I can't comment on that.  If you know ur power and current you can calculate you're Z at the point where the measurement is being taken.  I would think your E and I distribution along the line is whatever it is, and as long as it's balanced, ur vswr on the coax is okay and you don't detect any heating in the tuner, then I'd think you'd be good to go.  you say a little heating on the clamps of the cap.  How's the inductor or inductors doing?  They getting hot?  You could try a CQ for a minute or two, switch over to rx and put ur hand in there near the coils and see if you feel any heat.  I don't know how much current you expect to see--500 mils for 200 w. seems pretty good to me--on 75, with 250 w. I was seeing 300 ma.  I was using one of those MFJ balanced amp meters.  On 40, my feed is an awkward length.  I can tune the antenna, a 1 w/l loop on 80 to 1:1 on 40 but it's a tight swr null.  I'd get it tuned, and commence calling CQ.  a minute or so later I'd see a big increase in power coming back.  Get up and find the vswr had changed.  Finaly found out on 40 that MFJ amp meter was getting hot.  Something in there didn't like the Z and/or the carrier.   My tuner is synched 15 uH (or maybe they're 20 can't remember) rollers and a 1000 pF vac variable.  maybe on 20 you need to add some capacitance, or like the other fellow said, make a small change in the feedline so ur tuner can get you to 1:1. 

I have also chased monitor distortion to find it's either too high level on a mixer channel to my cans, or a rx up too high.  Oh I just saw where you get 3 A on 160.  That's all dependent on the Z on the line, where on the line you measuring, frequency, and so on.  if you want to be certain measure at the actual feedpoint which you can't do with a dipole up in the air, but that's the way to find what you're actually delivering to the load b/c if you have that, you can probably get the f/p Z with modeling, then you can figure out the power at the f/p that's getting out.  Let's see...to calculate the R at the measurment point of the current, I think you divide ur power by ur current squared.  200 / .5**2 should equal the resistance on the line.  800 ohms?  that's actually not too bad but I guess that doesn't give us the reactive part.  Well, I'm starting to feel like I've been in over my head so maybe a real antenna expert can take over 

73

rob K5UJ

[Steve - WB3HUZ]

You feedpoint impedance will be about 193 + j422 Ohms. At the end of the 120 feet of ladder-line, the impedance will be about 119 - j102. You should see about 1 Amp total under these conditions, so it seems like things are OK FB OM. 

[flintstone mop]

My inability to use math has haunted me all of my life. Thank you Steve for the calculations. I'll keep monitoring and calling the last couple of nights have been a little quiet.
Rob, Thanks for the input. It's not always a bed of roses using an untuned antenna system and ladder line. It seems once you get near a high impedance situation, there's high voltages that can be right at the tuner terminals.
The overload is from my R390A. Too much RF getting into the front end. Even though the receiver is isolated from the TX-RX relay with another antenna relay. 40M and down there's no problem. At 20M the leakage increases and overloads the receiver.
THanks to all.
Fred

[Tom WA3KLR]

Hi Fred,

I didn't have time to check this stuff last night.  I'm running the antenna and transmission line simulation programs, glad they exist, no math to do. 

I came very close to your results at 160 meters - 3.2 Amps.  I expected this to be close, not critical. 

For 20 meters though, the effective transmission line length and antenna impedance is much more critical.  As Steve said you should get 1 Amp ideally instead of the 1/2 amp you do.  This difference is due to tuner losses as you may be very close to an impedance peak; you may be just a few feet away from it. 

My thought - If you can, try adding 10 feet of line, this will move you farther away from the voltage node and the tuner losses will then drop.  This accounts for a couple dB.

You have a lot of lobing and nulls with this kind of antenna at 20 meters.  Another suggestion is to unfortunately add a separate 1/2 wave dipole at 35 feet for 20 meters to your antenna farm (or a vertical, got a pole barn roof?).  But this can be fed with some RG58 or RG8 and a 1:1 current balun at the feed point, no tuner needed.

[K1JJ]

Hi Fred,

You have a lot of lobing and nulls with this kind of antenna at 20 meters.  Another suggestion is to unfortunately add a separate 1/2 wave dipole at 35 feet for 20 meters to your antenna farm (or a vertical, got a pole barn roof?).  But this can be fed with some RG58 or RG8 and a 1:1 current balun at the feed point, no tuner needed.

I second that suggestion. A 1/2 wave, coax fed  20M dipole will have a clean, broad, figure-8 pattern.  The only nulls are off the sides.  Environmentally stable vs: open line - and no tuner needed.


Second choice would be to add a set of 20M legs (16' each) to the openwire fed dipole, but I wouldn't bother. Go with the coax fed dipole for 20M monoband operation. That antenna will work like a bomb at 35' high for USA contacts.  (1/2 wave high)

T

[K5UJ]

Tom, thanks for the suggestion to add some feedline.  I will have to try that on 40 m. on my antenna; it may make tuning on that band easier.

Fred, you could also try adding two more supports and turning your dipole into a loop which is what I did to get as much use out of one antenna as possible.  You'll still get all these lobes on 20, but they'll take off in more directions with a loop.

Rob K5UJ

[flintstone mop]

I think I'll try the "second choice" of Tom Vu.
Less work and expense for me.
I have lottsa good antenna wire and dacron rope.

Fred

[K3ZS]

I followed Tom's advice on adding a 10M length dipole under my 135 ft ladder line dipole.   This was for 10 meters and it greatly improved the operation.    Adding two 16 ft legs should do the same for you on 20M.     I got some PVC pipe and made some spreaders, much like you would when making homemade ladder line, to separate the second dipole.    Drilled a hole through one end of each spreader and strung the new elements through.    Cut a slit using a saw on the other side of each spreader and slipped them over the existing antenna wire.    Make the slit only far enough so that you can glue a pipe cap over each one to hold it in place.  It's quick to do, only had to lower the antenna, of course you will have to solder or bug the new elements to the existing dipole.

[Steve - WB3HUZ]

Tom Vu vely smawt man!

[K5UJ]

Hi Fred,

Adding another antenna may not be a bad idea but I'd consider a smaller lower version of what you already have, such as a center fed dipole with each side around 22 feet long and fed with parallel wire feed.  You could probably use it with the famous tuner, and it would probably give you a good match for 15 through 40 m so there would be a lot of versatility there.  This is an attempt at Cebik's 1/3 w. dipole for 40 m. on up, a nice all around general purpose antenna and it might even be okay on 10.

Rob K5UJ

[Steve - WB3HUZ]

3/8 WL at the lowest frequency is preferable.

[Steve - WB3HUZ]

The cleaner pattern when using a shorter dipole has been mentioned. Take a look at the patterns for your exsting 180 foot long antenna on 20-meters. The broadside elevation pattern (the first image below) doesn't look too bad - a nice low angle lobe at about 14 degrees and another higher angle lobe. But that's not the entire story.

Look at the azimuth pattern, just at the 14 degree elevation angle. Lots of stuff going on here, including lobes off the side or ends of the wire and more of a cloverleaf pattern over all (not necessarily a bad thing if those leafs are pointed in the right directions). But still, there are lots of nulls, so those directions will not be covered very well. Even here, the entire story is not told.

Take a look at the bottom image - the 3D pattern. Now you are seeing everything. Look at all those crazy lobes and nulls. This makes it clear why adding a 20-meter dipole (or similar) might be a good idea.

[K1JJ]

Yep, that 3-D pattern is what I call a "garbage" antenna pattern.

Good for nothin. The funny part is there is no meaningful gain anywhere but you're STILL giving up huge broadside nulls to boot.   The ideal pattern is like a flashlight. It puts the RF where you want and is dead in directions you don't. The gain is big on the front and the null is huge off the sides and rear where you don't mind. (Assuming it's rotatable or you have a second one to switch too.) 

Even a 1/2 wave dipole has nulls off the sides - that's why two dipoles at right angles, switchable, is a good way to go to cover all four directions.  When there's nulls, there's a chance of gain.  But the thing to remember is it's not an instinctive thing at all. Running  a centerfed 180' wire on 20M does not produce big gain in any particular direction cuz the wire is not phased properly. In contrast, only 100' of wire in the form of a 3el wire 20M Yagi will blow away the 180' dipole by over 10db - with a huge f-b to boot. The hearing ability is no contest.

[flintstone mop]

WOW that last image looked like RF hell. Tacking on a 3/8 WL for 20M onto the existing system sounds like a plan.
And Bob, thanks for the tips for separators to keep the two radiators from interaction and tangling. I just hope this extra stuff doesn't break on those high wind things we get around here in Winter.
But I guess that's the fun of Ham radio....................repairing antennas on a 5 degree day.


fred

[K1JJ]

WOW that last image looked like RF hell. Tacking on a 3/8 WL for 20M onto the existing system sounds like a plan.
fred

I could be wrong, but I think Huz was referring to the absolute minimum length (3/8 WL) before losses start to mount up in case you wanted to use the antenna on a somewhat lower freq too. But in your case, there is no freq until you hit 7mhz. (or 10mhz)  Normally, you'd be better off with  1/2 wave dipole (+- 16' legs for 20M) to give a reasonable impedance for the openwire to work with, keep losses reasonable and provide a broad figure-8 pattern.

T

[K5UJ]

Fred,

When I recommended the smaller doublet I meant a totally separate antenna with a separate feedline; not one co-located with the big dipole and fed at the same point with the same feedline--I don't know how that would work out.  I know this kind of thing is done with coax (i.e. "fan dipoles") but with balanced line things may get more complex with two antennas close to each other affecting the tuning and balance on the line.   You'd have to fix it so the feedpoint Z of the desired antenna on a band you want to work would always most closely match the 450 or 600 ohms line impedence so that antenna is the driven antenna by default.  I would force the situation by simply using a separate antenna and feed.

As one who is risk averse and who wants antennas to work with a minimum of fooling around, I'd put it up at right angle to the big dipole and shifted a bit one way so the feedline drop is not too close to the other feed and bring it to a switch point.  The simplest way would be to have a common feed ending with banana jacks and your two antenna feeds ending with plugs and you simply plug in whichever antenna you want to use.  If you want to get fancy and avoid getting out in the winter to change antennas you could bring them both in and have a big knife switch arrangement.

Rob

[Steve - WB3HUZ]

Yes, 3/8 WL at the lowest freq you want to use. In Fred's case, this is 20-meters. So make the thing 25 feet long. Yes, a half-wavelength will be slightly lower loss, but it will also produce much more difficult to tune impedances on 15 and 10 meter (possibly 17 and 12 too, but I haven't checked).

If you only want to use the antenna on 20 meters, just go with the half-wavelength dipole and be done with it. Feed it with coax and rock and roll.

WOW that last image looked like RF hell. Tacking on a 3/8 WL for 20M onto the existing system sounds like a plan.
fred

I could be wrong, but I think Huz was referring to the absolute minimum length (3/8 WL) before losses start to mount up in case you wanted to use the antenna on a somewhat lower freq too. But in your case, there is no freq until you hit 7mhz. (or 10mhz)  Normally, you'd be better off with  1/2 wave dipole (+- 16' legs for 20M) to give a reasonable impedance for the openwire to work with, keep losses reasonable and provide a broad figure-8 pattern.

T

[K3ZS]

Fred,

When I recommended the smaller doublet I meant a totally separate antenna with a separate feedline; not one co-located with the big dipole and fed at the same point with the same feedline--I don't know how that would work out.  I know this kind of thing is done with coax (i.e. "fan dipoles") but with balanced line things may get more complex with two antennas close to each other affecting the tuning and balance on the line.   You'd have to fix it so the feedpoint Z of the desired antenna on a band you want to work would always most closely match the 450 or 600 ohms line impedence so that antenna is the driven antenna by default.  I would force the situation by simply using a separate antenna and feed.

As one who is risk averse and who wants antennas to work with a minimum of fooling around, I'd put it up at right angle to the big dipole and shifted a bit one way so the feedline drop is not too close to the other feed and bring it to a switch point.  The simplest way would be to have a common feed ending with banana jacks and your two antenna feeds ending with plugs and you simply plug in whichever antenna you want to use.  If you want to get fancy and avoid getting out in the winter to change antennas you could bring them both in and have a big knife switch arrangement.

Rob

Rob,

The feed point impedance being matched to 400 - 600 ohm line has nothing to do with it.    Balanced line fed antennas hardly ever have an impedance matching the feed line.    It is the low loss of the feedline that allows you to run a high SWR between the line and the antenna.   That is why you can use these type of antennas on many bands.   Having a half wave dipole element in parallel with an otherwise higher impedance element, will take most of the RF current for the band it is cut for.    If you try paralleling a double extended zepp or other types, you won't get the same result.    Try modeling these this using the free version of EZNEC.    I did this after K1JJ set me on the path of looking at antenna patterns and you will see what happens with the different configurations.

[K1JJ]

Thanks for the kind comments, Bob.    BTW, you look very suave, debonaire and quite dapper in your tuxedo on QRZ.com...    It wud make a good avatar here.

T

[K5UJ]

Rob,

The feed point impedance being matched to 400 - 600 ohm line has nothing to do with it.    Balanced line fed antennas hardly ever have an impedance matching the feed line.    It is the low loss of the feedline that allows you to run a high SWR between the line and the antenna.   That is why you can use these type of antennas on many bands.   Having a half wave dipole element in parallel with an otherwise higher impedance element, will take most of the RF current for the band it is cut for.    If you try paralleling a double extended zepp or other types, you won't get the same result.    Try modeling these this using the free version of EZNEC.    I did this after K1JJ set me on the path of looking at antenna patterns and you will see what happens with the different configurations.

Hi,

Here's how I see it:   "It is the low loss of the feedline that allows you to run a high SWR between the line and the antenna."

That low loss is due to the line being balanced.  That's why vswr and a mismatch between a balanced antenna feedpoint and the characteristic Z of the line doesn't matter when it comes to line loss.  You can even operate with an antenna on a frequency below its lowest design frequency (up to a point). 
"The feed point impedance being matched to 400 - 600 ohm line has nothing to do with it."

I'm not convinced of that.  It _does_ matter in the unusual case of feeding two antennas from one balanced feedline.  If you have two antennas on the same feed, on a given frequency, the one with the f/p Z that comes closest to the Z of the line is going to get the most energy transferred to it whether that's the one you want to use or not.   Fan dipoles work on this simple principle because you have a bunch of 1/2 w. dipoles being fed by 75 ohm coax.  Whatever band you want to operate on, if you have a half wave dipole up in the fan that's cut for that band, that's going to be your radiator because its Z is gg to be near 75 ohms; none of the others will be (usually). 

My point, and I don't have the fancy modeling software so on that you have the advantage, is that if we have a big dipole and a 1/2 w. dipole for 14 MHz and you feed them both with some kind of parallel feedline and get on say, 12 meters, you're going to try to get your balanced tuner to find a match to 50 ohms unbalanced but both antennas will have some bizarre Z relative to the line, and, without doing some testing, or modeling, you might find that the unwanted antenna gets most of the RF because its Z is closest to the Z of the line.  If there's something I'm missing or something wrong with this reasoning please tell me.  Tnx,

Rob K5UJ

[K1JJ]

Feedline loss is directly related to its dielectric quality and type. Feedline can be balanced or unbalanced with no difference in loss figures.
The dielectric (insulation) separates and holds the two wires apart and in place. (or center conductor and shield) This wire placement is critical to maintain a certain characteristic impedance, like say 50 ohms.  *** Ideally, if the wires could be held in place with no dielectric at all, there would be almost no loss, like openwire.

The majority of the loss is generated from the secondary path (short circuit) from one wire to the other through the dielectric.... all along the feeder length. It will actually heat up the dielectric (and whole cable) if the swr is high with sufficient power.  Dielectric loss generally goes up as frequency increases.


For example, RG-213 uses a polyethylene  dielectric that has about a 0.5db loss per 100' on 40M when perfectly matched. (1:1).  As the swr (mismatch) goes up, so does this db loss, until little power arrives at the antenna. This is heat in the dielectric circulating "short circuit" path.  PolyFoam is better, but cannot handle the power (heat) as well.  Zip cord feedline, using rubber or plastic insulation (dielectric) is poor, especially at higher freqs. Hardline with thin poly disks fused every 3" or so is exellent due to the small amount of dielectric surface contact between wires. It's not the wire size so much as the insulation used that dictates the loss.  Though wire size becomes more important at higher HF and VHF freqs due to skin effect. Also low impedance circuits need larger conductors.


Now compare this to a feedline with little to no dielectric like wide-spaced, well made, homebrew openwire line.  The only dielectric for loss is the few widely spaced insulators (good quality) or AIR.  Air is a great insulator and little RF flows across the wires as loss. Even better, an idea I use is where the two feeder wires are stretched tight between the tower and the ground, about 18" apart using no spacers. This uses NO dielectric except air and the four end insulators. To get better than that would require thick wires in a vacuum. 

With high swr, bigger conductor size can have a beneficial effect.

Connecting two coaxes in parallel will NOT improve a lossy feedline situation cuz the path splits and sees the same amount of dielectric in the end. It will produce a double sized set of wire conductors, but the dielectric loss overshadows the conductor size improvement, so improvement is slight.  I believe when we get into air openwire, then conductor size can have a more meaningful percentage improvement - once the dielectric gremlin is gone, especially when dealing with high swr and very low impedances, thus high RF currents and, of course, skin effect.  When using coax at high swr below 14 mhz, we're already dead anyway due to dielectric loss, regardless of conductor size.

Hope this helps.

T

[Steve - WB3HUZ]

Here's how I see it:   "It is the low loss of the feedline that allows you to run a high SWR between the line and the antenna."

That low loss is due to the line being balanced.  That's why vswr and a mismatch between a balanced antenna feedpoint and the characteristic Z of the line doesn't matter when it comes to line loss. 

It has nothing to do with the line being balanced and everything with the loss of the line. You could do the same thing with air spaced coax.

Open-wire line has far less loss when connected to a matched load than do nearly all types of coax (most certainly those in common use by amateur radios ops). As the load is changed from the matched condition, loss increases (and not linearly) on both the open-wire line and coax. But since coax starts out with so much less loss, it exhibits far less loss due to mismatch or high SWR.

This is all explained in detail in the ARRL Handbook. There is even a graph and a formula you can use to determine the amount of loss at a given SWR, knowing the SWR, the matched loss of the line in question (matched loss implies a given frequency). Check it out and run a few examples to satisfy yourself.

To futher clarify JJ's post, yes, dielectric loss is the issue, most often. One case where an additional loss factor comes into play is when feeding a very low-Z antenna. A typical case is using a dipole well below it's half-wave resonant frequency. Here wire resistance loss will become a factor due to the very high currents involved. Dave, W2VW has made some interesting measurements on the loss with open-wire line using different conductor sizes when feeding a 75 meter dipole on 160 meters. I'll let him fill in the details, if he chooses, but short story is that larger conductors had much less loss (not surprising).

I'm not convinced of that.  It _does_ matter in the unusual case of feeding two antennas from one balanced feedline.  If you have two antennas on the same feed, on a given frequency, the one with the f/p Z that comes closest to the Z of the line is going to get the most energy transferred to it whether that's the one you want to use or not.   Fan dipoles work on this simple principle because you have a bunch of 1/2 w. dipoles being fed by 75 ohm coax.  Whatever band you want to operate on, if you have a half wave dipole up in the fan that's cut for that band, that's going to be your radiator because its Z is gg to be near 75 ohms; none of the others will be (usually). 

My point, and I don't have the fancy modeling software so on that you have the advantage, is that if we have a big dipole and a 1/2 w. dipole for 14 MHz and you feed them both with some kind of parallel feedline and get on say, 12 meters, you're going to try to get your balanced tuner to find a match to 50 ohms unbalanced but both antennas will have some bizarre Z relative to the line, and, without doing some testing, or modeling, you might find that the unwanted antenna gets most of the RF because its Z is closest to the Z of the line.  If there's something I'm missing or something wrong with this reasoning please tell me.  Tnx,

You are correct in the 12 meter case. The big (in this case 180 foot) wire will take most of the power and dominate the pattern (and produce an ugly one). The situation may be different on other bands. Use the force Luke and DL any one of a number of FREE antenna modeling programs. No need to work in the dark any more and make guess, educated guesses, or even good engineering estimate. Model it and know for sure within minutes.

[flintstone mop]

Hello again,
I think I'm interpreting what Rob is saying. I hope I'm close. Because there are two random lengths of antenna wire or 3/8 wave at 160M and 20M (eventually at my station) that there may be some frequencies that will go to both wires. I don't think so. I think Bob K3ZS has almost this same antenna system in operation at his station.

I can't explain the theory, but Tom and Steve and Bob, K3ZS have been trying to get us on track.

This may be a very easy way to keep the tuner happy and reduce losses and not have to hassle with resonating a 50 ohm system or even using lossy traps.

Fred