Broadening a 160 meter dipole

[steve_qix]

I've got a 160 meter dipole up - with the center at 90 feet and the ends around 60 feet off the ground.

It is fed in the center with coax (no, I can't change it to be open wire line    )

Anyway, the antenna is very sharp.  It's resonant at around 1915 or so, and at 1945 the SWR climbs to a little over 2:1, and about the same at 1885.

Is there anything I can do to reduce the Q of this antenna.  I am thinking of adding 2 or possibly 3 more wires to each side and fanning them out a bit.  That will be difficult from a physical standpoint, so I'm wondering if just using something like 3 conductors on each leg spaced about a foot apart will be sufficinet.

Any ideas?

I'm trying to keep the line flat.  I could tune things at the transmitter end, but it doesn't really solve the problem.

[KD6VXI]

How about using a stub to broadband it?  I did just that in my mobile installation pre-screwdriver antenna, and it worked fairly well.....

I'm guestimating you don't want to reinvent it, either.....  There are a couple broadbanding designs you can do with coax in the legs of it.....  Seems a waste to completely re-engineer it, though.

--Shane
KD6VXI

[K1JJ]

Yes, the fan idea is good. Pick your favorite freqs on 160M and add dipole legs to cover them.  For example, for 1885 add an 1850 leg and for 1945 add a 1960 leg.  That should work well across the desired band.

Of course, you will need to prune them to find the best swr points due to their mutual coupling effects and heights above ground in inverted V config.


T

[WD8BIL]

Hi Steve,
Nice Christmas card BTW!

According to 468/f there's 8 foot difference 'tween 1885kHz and 1945kHz.

I wonder ifn you put one dipole at 1885 (@248ft) and one at 1945 (@240) ifn 1915kHz swr wouldn't be acceptable? I've done this with 75/80 meter dipoles off one feedline before and it worked great. In fact, the 7 year old fan dipole at the cabin is setup this way. 80M is 126ft and 75M is 120ft.

[Steve - K4HX]

Why are you concerned with a 2:1 SWR at such a low frequency?

[WD8BIL]

Why are you concerned with a 2:1 SWR at such a low frequency?

Class E??? 3 legged fuses ?

[Steve - K4HX]

Could be but I zeroed in on these lines in Steve's post.

I'm trying to keep the line flat.  I could tune things at the transmitter end, but it doesn't really solve the problem.

Why are you concerned with a 2:1 SWR at such a low frequency?

Class E??? 3 legged fuses ?

[K1JJ]

It might have to do with the class E control shutdown circuitry. Once optimized to handle a good 1:1 swr antenna, a somewhat higher SWR can shut the rig down if set close. Most of us have the various shutdown adj trimmers inside the rig, so a flat antenna is an advantage to have.

Or, maybe Steve is an SWR perfectionist and likes every last .01db  pushing up that S-meter...


My other tube pi-network rigs with no shutdown circuitry are fine with 2:1 or more SWR on 160M with coax-fed ants, so I don't worry about it.



T

[steve_qix]

Why are you concerned with a 2:1 SWR at such a low frequency?

Class E??? 3 legged fuses ?

Nothing like that.  The transmitter can work into anything - and this one has (the antenna has fallen down when I was on the air - the overload simply shuts things down).

I just want to have as lossless a situation as possible, and also I don't want any high voltages developing anywhere (not that the voltages would be all THAT high).

[steve_qix]

Or, maybe Steve is an SWR perfectionist and likes every last .01db  pushing up that S-meter...

That is really more the point, truth be known.  It just bugs me to have loss, albiet small ones, anywhere in the system.

[WD8BIL]

Do you have the dipole setup where you can easily let the ends down for trimming, Steve.
The fan dipole approach may require a few cut and measure cycles because of capacitive loading. They will end up somewhat shorter than the math produced lengths.

[WA1GFZ]

My Cousin Joe had the same problem at W1AW on 80 meters. he built a nice cage dipole between two towers. It took a couple extra support ropes to the ground to keep it from twisting in the wind. 2 or 3 elements should work for you to cover 75 KHz. Might be worth simulating.

[Steve - K4HX]

Then it becomes a tradeoff between the loss you have now and the loss created by adding wires, stubs, etc. If you are using hardline, the loss will be no worse than 0.02 dB per 100 feet.

Unless you have a very long run, it hardly seems worth worrying about. You could easily be losing that much in connectors and jumpers.

Or, maybe Steve is an SWR perfectionist and likes every last .01db  pushing up that S-meter...

That is really more the point, truth be known.  It just bugs me to have loss, albiet small ones, anywhere in the system.

[k4kyv]

Nothing like that.  The transmitter can work into anything - and this one has (the antenna has fallen down when I was on the air - the overload simply shuts things down).

I just want to have as lossless a situation as possible, and also I don't want any high voltages developing anywhere (not that the voltages would be all THAT high).

Then why is open wire line and a tuner not an option?

I presently run a buried coax link between the shack and the tuners at the base of the tower. When I first put it I up, I used RG-214, doubly shielded and silver plated throughout, and widely used for big-dish satellite receivers.  With a 140 ft. run, on 160m at 100w measured at the transmitter I got only 93 watts at the antenna with a perfectly flat 1:1 SWR working into a temporary dummy load.  After a few years, I ran the test again, and the efficiency had dwindled to about 80 watts at the far end.  I found that critters had eaten holes in the coax jacket allowing water to contaminate the coax.  So I replaced the line with some brand new direct-burial RG-213.  Once again, the efficiency came back up to 93%.  But this past summer I re-checked it, and found that the power at the other end was 75 watts.  I am now losing 25% of my power in that coax link!

Plans are to temporarily replace the buried coax with some more RG-213 from the same roll, but instead of burying it, it will be carried to the tower in an overhead run supported by 8' high metal T-posts.  As a permanent solution I am working on using a flat 440Ω open wire line for the link between shack and tuning unit, and then using the present open wire tuned feeders up the tower to feed the dipole, and designing and constructing a new balanced-to-unbalanced set of tuners to match the flat open wire line to the vertical on 160, and as an added feature, a half-wave vertical for 75/80.

I would not tolerate a spider web of separate dipoles coming off the tower to cover every part of every band that I work.  I prefer to put up one good, solid, efficient dipole and use tuned feeders to make it work on all the bands that I operate.

Regarding the current tuner re-build project, I now have the five bread-slicers ganged together and running with the reversible DC motor, along with the indicator in the shack. Several of the coils are permanently mounted into position. Next comes the task of completing the wiring of all 5 of the separate tuners and the selector switch.

Once the tuners are all running, I'll replace the present coax link with new cable, and begin testing the design of the revised tuners to match the flat open wire line to the tuned feeders.

The only glitch so far is that the supposedly hermetically sealed precision indicator pot seems to have developed several noisy spots that cause the meter to flicker when the motor is turning, after it sat unused out in the unheated dawg house for a couple of months. I plan to apply about 75 volts a.c. across the pot when not in use to keep it warm hoping to avoid moisture condensation, which must be what happened, since it worked perfectly on the bench before I mounted it in the dawg house.

In the meantime, I am still using JS tuners mounted in the old dawg house for 80 and 40, and this has me temporarily off 160.

[W2VW]

Almost anything you do to remedy the loss will add it's own loss and/or complication.

Double Bazooka = cool name of an antenna that wastes power in the name of low SWR bandwidth.

Tuner in the shack fed into coax will normalize the load for your 50 ohm meters but will likely lose a few percent by just being there. 

Multi wires will increase wind load and ice load. They will however get you in good shape while pruning.


If you can get the ends up higher the bandwidth will increase as you probably already know.

It might be interesting to calculate/measure the complex impedance in the shack and come up with a bandaid network simply to satisfy the meters. It will not make any improvement to the signal even if made with very heavy components.

Clicking your feedline ruby slippers together can get the same antenna back to Kansas in style while allowing you to have possibly similar gain as doubling your FETS while in Oz. 

[KM1H]

I use dipoles cut for 1850 and 1950 to cover the full band. I could care less about SWR at those frequencies just as long as the amp can load into the antenna.

For 80 its similar, one is cut for 3575 and the other for 3750 which loads OK at 3885.

Both bands are fed with the same coax so a lot of trimming was needed as nothing conformed to any books.

Actual VSWR doesnt exceed 2.5:1 over 1.8-2 MHz and 3.5 to 3.9 MHz.

I have no use for antenna tuners except the built in ones in the TS-940 and 950SDX when I decide to run barefoot. Nor do I have any use for open wire line.

The LK-500 and Alpha 76 amps load well as do any of the pi network BA rigs and amps.

Carl

 

[Bill, KD0HG]

Steve, on 160 I use a conventional center fed half wave dipole configured as a sloper, one end tied to the top of the stick, the other end some 6' off the ground maybe 100' out. Coax feed with a current balun near the feedpoint.

The dipole is configured as a fan design with the ends spread 5' using fiberglass fence posts as spreaders. I can cover about 1/3 of the band with less than a 2:1 VSWR. I don't want to put full fire in the coax beyond that.

I suppose you could also use an unbalanced tuner, if your coax line can take QRO with a high VSWR.

The effective diameter of a dipole made of two spread wires is the mean distance between the two. Adding a third wire would have less broadening effect than using a 2-wire design unless you also increased the spread of the wires at the ends.

If you know the effective diameter of a fan dipole from the rule, then you can plug a wire diameter value of X feet into EZNEC and calculate the antenna bandwidth and the length of the wires you would need. Saves a lot of time.

[k4kyv]

I think the obsession many have with SWR is a little Hammy Hambonish. For a good treatise on the subject, I would highly recommend Walt's book Reflections. Pay particular attention to the sections on SWR and "conjugate matching".

A (nearly) loss-less antenna that covers a wide band of frequencies is about equally achievable to a perpetual motion machine.

About the closest one could come to a 100% efficient antenna would be a folded dipole fed with a 300Ω "Q" section (1/4" copper tubing spaced 1 1/2" apart). Drive it with a push-pull final using a balanced PI network, with split tank coil, and split stator plate and loading capacitors.  (A push-pull class E  circuit should also be feasible). The efficiency (excluding plate losses in the tube or losses in the solid state output devices) would be nearly 100% at the resonant frequency (but there still would be some loss in the tank circuit). It wouldn't cover an entire band, but should work over a substantial portion of a band.  But I would say go ahead and use it across the entire band anyway.  There would be some SWR at the band edges and the reactance and mismatch could be tuned out with the PI network, but I would challenge anyone to detect any variation in signal strength across the entire band, particularly once you move beyond the near-field of the antenna.

Now that the FeeCee defines transmitter power in terms of "output", that could be a distinct advantage for those who wish to follow the letter of the law to a "T" while running the absolute maximum legal limit.  Place your power meter right at the feed point of the antenna.  For a dipole, this would be at the point where the feed line connects to the flat-top.  For a vertical, it would be right at the base of the antenna.  Then, regardless of what kind of final you are using or how crappy the feedline is, turn up the power until the meter reads exactly the legal limit.  So what if you are running an extra 25-50% DC input power to achieve that output reading? You already had measurable losses in the power transformer and other power supply components, as well as the filament power consumed by any tubes in the transmitter, plus all the power consumed by the rf exciter and audio stages. Furthermore, I suspect many who take pride in how "efficient" they think their set-up is, would be surprised find that their DC input to the final was running at 1kw or more.

[K1JJ]

Don't worry about a little swr. Go for the gusto.  

Use hardline or openwire feeders in the first place. Then add a reflector behind the driven element  (or phase a second antenna to the first) Match it 1:1 at your favorite freq. Double the tubes in the final.

That's an easy 8db.  


Ding! Lice all done.

T

[K5UJ]

Steve QIX, what type coax feedline are you using now?    How is the feed point getting supported?   You basically run into some physical and electrical barriers with a coax fed dipole and getting it to show a low and flat vswr curve but still be efficient (for a low dipole on 160).   If you are using some sort of average coax like 213, I'd swap that for LDF4-50 if your support can hold it and consider this if you have room:  your dipole is pretty much non-directional at 90 feet on 160.  how about adding a second dipole cut for the upper end of the range you want and making it at a right angle to the first one.   You could use a robust remoted coax switch up near the feed points so you'd have one run of Andrew hardline to the switch, then 10 foot jumpers from it to the dipoles.  You wouldn't want to feed them both simultaneously because I think you might get into some weird phasing issue problems (but that's just my intuition) with them at 90 degrees to each other.  Why this and not the "fan" dipole scenario--not sure--there's something about two discrete dipoles at right angles to cover the band I like but even if you go with only one, I think the biggest loss prevention measure you can take is using hardline.   If you go to one of those on-line loss calculator websites and do some analyzing you'll find that hardline even on 1.8 MHz makes a relatively big difference even for runs of 200 feet at 1:1 vswr. 
Another good reason for using it is that the stuff lasts forever.

Rob     

[steve_qix]

Lots of great suggestions and much food for thought!  I am using old 3/4 inch (or thereabouts) helical hard line.  It's been up for almost 20 years.  The center is supported by a 90 foot tower.  The ends are in trees.

On the same feedline, and at a 90 degree angle to the 160 meter antenna, is my 75 meter diople.  That antenna is broadside S/W, and there is also a lot of, as Brent calls it, "topographical gain" in the S/W direction (the hill drops off steeply to the West and keeps dropping off for a long distance (thousands of feet)).

Anyway, that's the setup.  On 75, the antenna is sufficiently broad, but on 160, because the ends are not electrically high off the ground, the antenna is a lot sharper than I'd like.

An extra set of wires (one on each end) might be the trick, but more work in keeping it in the air, so it may not be worth the effort..... I am pondering what to do!

I really appreciate the suggestions and info so far.

[WA1GFZ]

Steve, I wonder if you attach a vertical wire say 20 feet from the end and tune it you might make it a bit flatter. You wouldn't need to climb the tower just drop the ends.

[ke7trp]

I understand his obsession with flat SWR.   My 160 verical has about 80 to 90 KC of band width and this is in the CW/phone portion.  On 1885 its 2 to 1. The GK500 does not care with its output network but it always bugs me to see that 15 to 20 watts reflected on the meter. 

The fan design seems kind of silly to me. Adding the wires really is just making the transmitter happy and its an extreme amount of work to lower the antenna and work on it.  At this point, build or buy a simple T tuner and throw it in line to make the rig happy.  The loss wont be that much.  Otherwise, I think its time you ditch that coax and build a balanced antenna so actualy have more performance when you move around.  Not to mention at 160 length you will then have a killer 80 meter antenna up.


 

[w3jn]

"Better" is often the enemy of "good enough".

[k4kyv]

I have two antennas for 160.  I mainly use  the 127' base-insulated vertical, which is actually a vertical tee, since the feedline to the dipole is fed up through the interior of the tower to the 119' level, and the proximity of the feedline to tower couples the dipole to the tower tightly enough to add substantial top loading. The base impedance, instead of being on the order of 36 ohms, varies from about 150 to 350 ohms across the band, with a similar range of +j.  I use a simple L-network at the base of the tower to match it to the coax  link back to the shack, and tuned it to 1:1 at 1900.  I work it from 1800-2000 without adjusting the L-network, with a SWR of slightly less than 3:1 at the upper and  lower band edges, but I can't tell any difference in performance between the band edges and the middle.  Frankly, I was surprised that it has as much bandwidth as it does. With 120 quarter-wave radials in the ground, and everything brazed together with silver alloy hard solder, I don't think I can attribute the bandwidth to resistive losses in the system, even though I had expected  maybe +/- 40 kc/s without having to re-adjust the L-network for even a 3:1 match before I first tried it out. I use another matching network in the shack to couple the transmitter into the coax link.

The 80m dipole is fed with 440 ohm open wire tuned feeders (#10 copperweld spaced 2" apart).  I can load it on 160 as a quarter-wave dipole; in fact that's what I used before I got my  radial system installed.  I didn't bother with trying to run the vertical without a ground system.  I was able to get good reports from all over N. America with the dipole, but the vertical usually works much better.  The short dipole is extremely sharp tuning. I cannot move even 5 kc/s without having to re-adjust the tuner.  It's almost like being crystal controlled.  So I installed the reversible DC motor and worm drive so I could tune it remotely and not have to make a trip to the tower to QSY 5 kc/s.  Adjusting the tuner, I can get close to 1:1 in the coax link across the entire band.

One observation is that the short dipole is far superior to the vertical for close-in work.  In Nashville, about 50 miles away as the crow flies, I get about 30 dB stronger signal with the dipole, enough to make the difference between audible electrical noise in the background and full quieting carrier.  But out beyond about 100 miles, the vertical begins to take over as the superior antenna.

I use the vertical most of the time because of a limited number of AM stations within a radius of 100 miles of here.  There is a group of semi-locals up at the top of the band, so I have been running the dipole tuned to 1985, but using the vertical for all other work.  The SWR is barely above 1:1 at 1885.

Right now I am off 160 while I rework the tuner system.  That same reversible DC motor now rotates all 5 tuning condensers ganged together to tune all 5 tuners.  I am now in the process of wiring all the tuner circuits back together in the new dawg house and hope to have it up and running in a matter of days, not weeks. For the moment, I'll still have to make the trip to the tower to turn the selector switch, but will be able to adjust the frequency of whichever tuner I am using remotely from the shack.

I was recently given a high-torque stepping motor that rotates @ 1.8 degrees per step. Eventually I plan to use it to rotate the selector switch.  The rotary switch has 10 positions, or 36 degrees per step, so 20 pulses should accurately turn the switch from one position to the next.