The AM Forum
December 05, 2024, 05:20:48 PM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: 1 [2] 3 ... 5   Go Down
  Print  
Author Topic: vertical antennas  (Read 99430 times)
0 Members and 1 Guest are viewing this topic.
k4kyv
Contributing Member
Don
Member

Offline Offline

Posts: 10037



« Reply #25 on: May 30, 2011, 01:40:45 PM »

There is no guarantee that the beverage will be noise free. It, also, responds predominantly to vertically polarised signals.  It kills the noise by virtue of its horizontal selectivity - only selected sectors of the 360° pie are received, cutting out a lot of the noise that would be picked up by an omnidirectional vertical.  If the noise source happens to be in the same direction as the bore-sight of the beverage, you are SOL. I have that problem here. A road about a half-mile from here has noisy power lines on both sides, and a major intersection with another road with noisy power lines is right in line with New England and Europe from here, where my terminated beverage is aimed. Whenever any one of those power lines acts up, the beverage is useless.

The same is true for the K9AY, flag, pennant, EWE and similar "low noise" antennas.  They are essentially phased short verticals with a cardioid pattern. The trick is to get the null in the cardioid to point towards the noise source. A rotatable arrangement is most effective.

I carefully built a K6STI loop, but found it worthless. It was no more low-noise than a piece of wire tossed out the window and connected to the antenna terminal of the receiver.
Logged

Don, K4KYV                                       AMI#5
Licensed since 1959 and not happy to be back on AM...    Never got off AM in the first place.

- - -
This message was typed using the DVORAK keyboard layout.
http://www.mwbrooks.com/dvorak
K5UJ
Contributing
Member
*
Offline Offline

Posts: 2814



WWW
« Reply #26 on: May 30, 2011, 10:04:19 PM »

I wondered about that K6STI antenna and had planned to make one.  I got some of the parts to build the tuning network for it but never got around to it.   Now I have in the back of my mind a plan to make a pair of ferrite loopstick antennas for 160 that can be rotated.  I'd put them inside some cover like a PVC pipe and mount them a few feet off the ground, separated by 40 or 50 feet.  Each would have a preamp and they would feed my phase shifting nulling noise network.  First they would be rotated to null one point source then the noise canceling network would shift noise picked up on one, 90 degrees to null it with the signal from the other loopstick.  Two noises eliminated.   I can get the ferrite rod from Amidon.  I know a guy a few miles away from me who did something similar with a single loop stick to aid him in operating 160 CW.  I have his plans.  He built a single one in a PVC pipe with the preamp in the pipe getting DC via the coax feedine.  He had to null out some kind of line noise and he simply laid his on the ground and oriented it to null the noise and he said even on the ground it worked as a rx antenna (but he doesn't chase DX).   Anyway, the problems for me are that the costs of the materials and rotators start to add up but worse is finding the free time to set it all up. 
Logged

"Not taking crap or giving it is a pretty good lifestyle."--Frank
G3UUR
Member

Offline Offline

Posts: 141


« Reply #27 on: May 31, 2011, 09:26:33 AM »

Guys,

I’ve been watching this thread from the beginning and since Tom appeared to be doing a great job of advising AE1CT, I didn’t jump in and help. It was a great shame that it later turned into a pissing contest between a few ‘Big Guns’ because talk of 100ft high dipoles, elevated 80m ground-plane antennas and the like would have been of no help to AE1CT. In fact, it may have put him off amateur radio all together. I hope not, though, but the digression certainly didn’t helped.

He’s already stated that he has limited real estate and no trees higher than 30ft. His desire is to operate 40, 80 and possibly 160m. It sounds as if he’s in a typical suburban situation, and is probably surrounded by other houses all with a multitude of electrical appliances emitting digital broadband rubbish. It’s a hell of a situation to be in, but is typical of many amateur stations around the world. I operated from a similar situation for 7 years after getting my license. You’re all the time reacting to new noise sources and experimenting with different receive antennas. It complicates things enormously because in the early years after getting your license you’re normally trying to find the best transmitting antenna for your local situation anyway.

My antenna experience is limited to amateur radio, but having been in some difficult situations over the years I have experimented with antennas that are near to the ground and checked out all sorts of different configurations. I’ve also tried to learn what I can from others by reading books on EM theory and antennas. I don’t read books written by other amateurs, even if they are professionally involved in antenna work, but restrict my reading to the classic texts which I know I can trust. There’s too much mythology in amateur radio to trust books written by amateurs, though these books are often good for getting ideas about something out of the ordinary, which might just work well in your situation and enable you to put out a better signal.

So, let me tell you what I’ve found out over the years. If you want to operate 160m from a small plot, you’ll have to try both a short doublet and a bent quarter-wave inverted-L to see which works best in your situation. It very much depends on who you want to work and how far away they are. It’s too complicated to model and you could never do it realistically anyway, even if you could model the ground accurately enough, because of other obstacles, such a trees, shrubs and buildings with electrical wiring. The inverted-L antenna requires a good ground system to be effective and you might want to try the short doublet first because that’s easier to install. A major source of ground loss on 80 and 160 is E-field coupling to the ground beneath the antenna. Short doublets operate with a much higher E field than full-sized ones, so this ground loss can be a killer for low, short dipoles or doublets. This is one reason why the ends of low dipoles or doublets should not be bent downwards. It’s fine to bend the last 20 or 30ft down towards the ground if your antenna is at 100ft, but if it’s at 30ft high and you bend the last 20ft down towards the ground the extra loss reflected into the antenna can lose you more than 6dB. The exact amount depends on both the dielectric constant and conductivity of the ground beneath your antenna. The radiation resistance also drops a bit when you bend an antenna, so this doesn’t help the efficiency either. If you have to bend an antenna to fit it into a small plot, make the bends in the horizontal plane and keep the ends up as high as you can.   

The end-loading of short dipoles doesn’t reduce interaction with the ground much but it does raise the radiation resistance by up to 4 times. This improves the radiated signal by up to 6dB. Linear end-loading is preferable to inductive loading because it doesn’t reduce the bandwidth as much. Linear end-loading can very often be invisible in trees, whereas end-loading coils add weight to a run of wire and are also more visible. If you have any handy trees where you can terminate a dipole leg, you can run the linear loading wires up and down between branches to use up the extra length.

Reinartz loops are a great antenna for small plots, and although the radiation resistance is only about 15 ohms at the feed point, the fact that the ends are close together reduces ground loss when they are very low in height. A Reinartz loop is just a half-wave dipole bent into a square with the ends close together, but not connected. This not only reduces ground loss, but also reduces electrical pick-up to some extent. Both ends of a Reinartz loop on 80m could be loaded for 160m using linear loading sections for each end in the same tree, but you might want to add 80m traps so you don’t spoil the performance on 80m when you add the 160m linear loading section.

If you can manage to fit in an 80m full-wave loop fed with open-wire feeders, you can load this up on all bands, including 160m, though you would probably get better performance on 14MHz and above from a dipole for one of the bands at 30ft. I’ve never found multi-band wires, doublets or loops to work as well as theory predicts on the higher bands and have always used a separate antenna for 20 thru 10m. One word of warning about loops though. Generally, balanced antennas do not pick up interference as badly as unbalanced antennas, and balanced full-wave loops pick up less than balanced doublets or dipoles because they are closed loops. However, if in the process of putting up a full-wave loop, the extra run of wire takes your antenna closer to a dominant noise source, then it could be worse than a balanced doublet or dipole.

If you try a balanced antenna first, you can assess your local noise situation and see whether you can get away with using the antenna successfully for both receive and transmit. If you can’t and have to use a separate receive antenna, you might want to leave the balanced antenna up for a season and see how it performs over a period of time. Next year you might consider trying an inverted-L for transmit and a small magnetic loop, EWE or K9AY on receive. The inverted-L will need a good earth system to work well and that will require a lot of effort. However, it has a big advantage over a straight vertical because it produces high-angle radiation as well as low angle, and you’ll find that a great advantage for working stations closer than 250 to 300 miles on skywave. You’ll already have read about some of the noise problems associated with ground-mounted verticals, but besides that restricting all your radiation to ground wave and low angle in a suburban situation can be disastrous if you’re hemmed in by other buildings and a lot of your transmitted signal is absorbed locally. Elevated verticals with a separate ground plane are a different story, but in your situation they appear not to be a possibility.

Sorry about the length of my posting, AE1CT, but I hope some of this helps you decide what you need to do.

Dave.
Logged

Dave,G3UUR
Vintage AM from the East of England
flintstone mop
Contributing
Member
*
Offline Offline

Posts: 5048


« Reply #28 on: May 31, 2011, 10:56:54 AM »

What is the alleged magic about the 43' vertical on 160? Huh  I hear a lot of people, even those with plenty of space to erect something taller, report that's what they are using. Some operate with a good radial system and some without.  Obviously, performance will be much improved with the good ground system, but so would increasing the height to at least 1/8λ, approximately  65'.  Even with top loading and a good radial system, as the vertical length is decreased below 1/8λ, radiation efficiency drops off rapidly.

43' on 160 would be equivalent to 10'9" on 40m, barely more than an 8' mobile whip.
Hey Don
There is no free ride with a 43 footer. It needs a "loading wire" like an L and the unun included usually will not work either. The 'manufacturers' even say to use a certain type of coax with a certain velocity factor and 150 feet....so the magic is gone. You pay $300 for a 43 foot mast to really make an L antenna.
For the urban guy and small lot a dipole crammed in the space allotted is all he can do. Open  ladder line and a JJ tuner gets the dipole to other bands. "The SLAB Bacon" KB2AHE,  has had wonderful success with his shortened dipole even on 160M....Somewhere on this site Frank 'The Slab Bacon" posted info on his shortened dipole and how he made it work so well. If you follow his method it will work for you too.
Fred
Logged

Fred KC4MOP
Sam KS2AM
Contributing
Member
*
Offline Offline

Posts: 716



WWW
« Reply #29 on: May 31, 2011, 11:19:21 AM »

i dont have very tall trees in my yard (30 ft) high and the most room i have is probably 40 meters and i want to get on 80. i want to operate mosty sw from my location and there are no trees that will give me that direction. i want to get on the air but im in a position were i dont have a big yard or the tree hight that i need. WHATS a future AMer to do. any ideas would be greatly helpful.

Hello Gary.  Theres a new book, "ARRL's Small Antennas for Small Spaces". I have not read it but the description sounds encouraging and it may help you with ideas that will fit your property and your goals.

http://www.arrl.org/shop/ARRL-s-Small-Antennas-for-Small-Spaces/

Logged

--- Post No Bills ---
Steve - K4HX
Guest
« Reply #30 on: May 31, 2011, 03:39:31 PM »

http://amfone.net/Amforum/index.php?topic=24242.msg179317#msg179317
Logged
k4kyv
Contributing Member
Don
Member

Offline Offline

Posts: 10037



« Reply #31 on: May 31, 2011, 05:08:58 PM »

Linear end-loading is preferable to inductive loading because it doesn’t reduce the bandwidth as much.

Also true for inductive or capacitive feed line loading to electrically lengthen or shorten the feed line so that the feed point at the transmitter end presents a voltage or current loop at the tuner. That's why I use the "tennis net" (60' of feed line folded back on itself, switched in between the tuner and the feed line going up the tower, when I use the 80m dipole on 160, instead of simply inserting a couple of capacitors or coils in series with the feeders.

I once used an inverted-L, 130 ft. long, 65 ft. vertical and 65 more feet horizontal.  The ground plane consisted of 30 quarter-wave radials. When a storm took down both masts, I was able to salvage enough material to re-erect one mast, 65' tall, without the horizontal section. The resulting 1/8λ vertical performed about the same as the 1/4λ inverted-L.  I don't recall the difference in bandwidth, but at the time, we were limited here to only a couple of 25 kc/s slots of band, 1800-1825 and 1875-1900, so bandwidth wasn't so much an issue.
Logged

Don, K4KYV                                       AMI#5
Licensed since 1959 and not happy to be back on AM...    Never got off AM in the first place.

- - -
This message was typed using the DVORAK keyboard layout.
http://www.mwbrooks.com/dvorak
G3UUR
Member

Offline Offline

Posts: 141


« Reply #32 on: June 01, 2011, 06:07:09 AM »

Don,

I don’t know the height of the 80m dipole you use on 160m, but I suspect it’s quite a lot higher than 30ft. Ground loss reflected into the antenna at 60ft, or more, is probably low enough that you don’t notice any difference, but at lower heights as the radiation resistance drops and the ground loss increases the situation changes dramatically. Switching in extra feed line doesn’t alter the antenna at all, and at low heights improving the efficiency of the antenna is crucial to getting reasonably good results. My previous point was that you need some sort of end loading to increase the radiation resistance to improve the efficiency, and linear loading is the better option because the characteristic impedance of the extra section is lower and it introduces much less inductance into the antenna. This lowers the Q of the antenna and improves the bandwidth because the radiation resistance is higher and also the effective inductance of the series-resonant circuit formed by the antenna is much lower than it would be for inductive loading. Raising the radiation resistance is the primary goal and the bandwidth improvement is a secondary issue, though pretty useful to have.

Regarding your comments about the comparison between a 65ft high, 130ft long inverted-L and a 65ft vertical; are you suggesting AE1CT should use a vertical on 160m?  The purpose of this last part of your posting is not clear. Are you trying to convince him not to use the inverted-L type? I’ve done tests on 160m between a 32ft base-loaded vertical and 132ft inverted-L with 32ft of vertical using the same modest ground system from a small plot and the inverted-L wins by nearly 10dB on ground wave alone. When it comes to high-angle stuff there is just no comparison. Noise pick-up is also a problem with verticals, and I found that using a balanced doublet which could also be used with the feeders strapped for local ground-wave working on 160m was the best all-round compromise antenna in a small plot. It could be used balanced for receive even when transmitting with the feeders strapped, if needed.

Given the fact that any vertical installation needs to be out in the clear, have a reasonably extensive ground system and be at least one-twelfth of a wavelength high to work moderately well, I’d have thought that it would be a poor choice for AE1CT in his restricted circumstances because even if he put lots of effort into the ground system it wouldn’t be very extensive and he’d still be limited by other factors outside his control. Also, he’d have a skip zone between the limit of his ground-wave range and where his low-angle signals return to earth. There may be lots of stations he wants to work in this zone around New England. I may be wrong, but I’d have thought he needed some high-angle radiation as well.

Dave.
Logged

Dave,G3UUR
Vintage AM from the East of England
AJ1G
Contributing
Member
*
Offline Offline

Posts: 1292


« Reply #33 on: June 01, 2011, 07:44:13 AM »

Chris, AJ1G, was talking in another thread about yesterday driving his car mobile out on Stonington Point, CT (next to the ocean) and using 5 watts  20M CW to work Europeans easily. It wud be an interesting experiment to see the difference when he drove inland 10 miles and tried the same thing......

Tom  - you can up that to New Zealand (ZL2AGY)with 5 watts on 40 CW !  Did that the other morning bout 0630 EDT with the K1 and the hamstick clone on the Volvo...

I dont really need to go more than a half mile up the point inland before the signals drop off a lot relative to being on the Point.  A lot of that probably has to do with local absorbtion by buildings though.


* 100_2094.JPG (1057.51 KB, 3056x2292 - viewed 1120 times.)
Logged

Chris, AJ1G
Stonington, CT
flintstone mop
Contributing
Member
*
Offline Offline

Posts: 5048


« Reply #34 on: June 01, 2011, 08:38:16 AM »

Chris, AJ1G, was talking in another thread about yesterday driving his car mobile out on Stonington Point, CT (next to the ocean) and using 5 watts  20M CW to work Europeans easily. It wud be an interesting experiment to see the difference when he drove inland 10 miles and tried the same thing......

Tom  - you can up that to New Zealand (ZL2AGY)with 5 watts on 40 CW !  Did that the other morning bout 0630 EDT with the K1 and the hamstick clone on the Volvo...

I dont really need to go more than a half mile up the point inland before the signals drop off a lot relative to being on the Point.  A lot of that probably has to do with local absorbtion by buildings though.

Definitely makes sense when using 20M. Everything gets almost cookie cutter simple from about 40M on up.
80M on down there are certain criteria that cannot be avoided. Antenna efficiency, wavelengths, dipole height above ground, ground losses.
I hope the OP is not totally cornfused............but antennas are our favorite subject....hi
Keep the dipole at the top of your plans. As high as you can get it. As long as you can fit it in your property, even if it zigs or zags a little and feed it with ladderline and build a good tuner.

Fred
Logged

Fred KC4MOP
k4kyv
Contributing Member
Don
Member

Offline Offline

Posts: 10037



« Reply #35 on: June 01, 2011, 01:13:05 PM »

I don’t know the height of the 80m dipole you use on 160m, but I suspect it’s quite a lot higher than 30ft. Ground loss reflected into the antenna at 60ft, or more, is probably low enough that you don’t notice any difference, but at lower heights as the radiation resistance drops and the ground loss increases the situation changes dramatically. Switching in extra feed line doesn’t alter the antenna at all, and at low heights improving the efficiency of the antenna is crucial to getting reasonably good results.

You are correct.  My 135' long 80m dipole is close to 1/4λ high, at 119' at the apex, drooping to about 100' at each end. That was my only top band antenna after I first put up the tower, before I installed the radials.  I didn't even bother to try feeding the tower as a vertical without a radial system.  I think the height of the antenna makes up for the shortness of it; I can get fair results all over N America with it although the vertical tends to be superior at greater distances, and it is far better than the vertical within the skip zone.  For instance, in Nashville, which is about 50 miles away as the crow flies, the 80m dipole comes in up to 30 dB stronger than does the vertical, sometimes making the difference between a full quieting carrier, and a signal buried in local urban electrical noise although still readable.

Since each leg of the dipole is 1/8λ on 160, and the length of the open wire tuned feeders up the tower is 1/4λ, the total length of feed line plus dipole leg is 3/8λ - making the feed point at the bottom end of the feeders precisely midway between a voltage loop and a current loop. My first attempt to feed it was using a balanced link  coupled tuner at the base of the tower, parallel tuned, with the feeders tapped down on the coil.  I was able to get a 1:1 match with the transmission line from the shack, but using a split-stator capacitor with 3/16" spacing (rated at 7 kv), I couldn't modulate 100% at more than 100 watts before the capacitor arced over. Apparently, the tuner couldn't handle the highly reactive load very well. So I tried adding an additional 60' to the feeders, making the total feeders + dipole leg = 1/2λ, I was able to feed the open wire line directly with parallel tuning, get a perfect match, and modulate as much RF as I could generate well beyond 100% on positive peaks, with no sign of arc-over.

Not surprisingly, this set-up is extremely sharp tuning.  I cannot move much more than ± 5 kc/s without having to re-adjust the tuner. Originally, it was like being crystal controlled, since the tuner is at the base of the tower, about 140' away from the shack, so I ended up driving the tuning capacitor remotely from  the shack using a reversible motor and worm drive.  My point about bandwidth was that I could have tuned out the reactance at the OWL feed point by using a lumped value reactance (coil or capacitor) in series with each feeder, but (according to my intuitive reasoning, which may not be correct) I was afraid the tuning would have become even sharper, and not only would the split stator capacitor in the tuned circuit have to be re-adjusted with changes in frequency, so would the series capacitors or inductors. The equivalent series loading inductance in the form of the additional length of feed line, automatically varies to some extent as the frequency varies, but a lumped value inductance using a set of fixed coils would have to be manually adjusted with changes in frequency to maintain the best efficiency with the tuner, whie minimising the amount of reactance the tuner must compensate for.

Maybe my seat-of-the-trousers intuition is wrong, so I would be interested in other opinions: would a pair of fixed inductors each inserted in series with one of the balanced feeders work as well, with as wide a frequency range and as wide a bandwidth per setting, without incremental adjustment of the series L as the frequency is changed, as does my solution of merely inserting the additional 60' of feed line leaving the resonating capacitor in the ATU as the only variable adjustment?

Quote
My previous point was that you need some sort of end loading to increase the radiation resistance to improve the efficiency, and linear loading is the better option because the characteristic impedance of the extra section is lower and it introduces much less inductance into the antenna. This lowers the Q of the antenna and improves the bandwidth because the radiation resistance is higher and also the effective inductance of the series-resonant circuit formed by the antenna is much lower than it would be for inductive loading. Raising the radiation resistance is the primary goal and the bandwidth improvement is a secondary issue, though pretty useful to have.

Agreed. My additional section of resonant feeder would have no effect on radiation resistance or efficiency of the antenna. But as I thought up the idea while designing the antenna configuration way back in 1981, a nearly identical train of logic occurred to me as what you stated (above, in bold), although I never bothered to go through all the tedious calculations with various possible examples taking a set of frequencies across the entire band using both approaches, and comparing the results. I just tried out the added feeder section, it worked well, so I configured it to make the additional feed line length automatically switch in with the coil, and have thought little more of it since.

Quote
Regarding your comments about the comparison between a 65ft high, 130ft long inverted-L and a 65ft vertical; are you suggesting AE1CT should use a vertical on 160m?  The purpose of this last part of your posting is not clear. Are you trying to convince him not to use the inverted-L type?

No, I suppose I didn't make myself perfectly clear on that point. I was merely pointing out my own experience with the above described antenna. I think in AE1CT's case, the inverted-L would be the better choice, since the radiation resistance of a 32 ft vertical (approximately 1/16λ) would be very low and the antenna would be extremely inefficient without an extensive radial ground plane, and even with that, tuner loss would still be substantial. In my case, the vertical section was 1/8λ, so the radiation resistance should have been something on the order of 12Ω, which my radial system and tuning unit appeared to handle well. I wasn't able to make A-B comparisons of the L vs the vertical of course, but signal  reports always seemed about the same. Still, for 160m, 32 ft. is not much height for a horizontal radiator, but the additional top loading should make the vertical section radiate better than it would as a mere vertical standing alone.  At 32' on 160, this would be the exact equivalent of using the standard 8' mobile whip on 40m.


Quote
I’ve done tests on 160m between a 32ft base-loaded vertical and 132ft inverted-L with 32ft of vertical using the same modest ground system from a small plot and the inverted-L wins by nearly 10dB on ground wave alone. When it comes to high-angle stuff there is just no comparison. Noise pick-up is also a problem with verticals, and I found that using a balanced doublet which could also be used with the feeders strapped for local ground-wave working on 160m was the best all-round compromise antenna in a small plot. It could be used balanced for receive even when transmitting with the feeders strapped, if needed.

Given the fact that any vertical installation needs to be out in the clear, have a reasonably extensive ground system and be at least one-twelfth of a wavelength high to work moderately well, I’d have thought that it would be a poor choice for AE1CT in his restricted circumstances because even if he put lots of effort into the ground system it wouldn’t be very extensive and he’d still be limited by other factors outside his control. Also, he’d have a skip zone between the limit of his ground-wave range and where his low-angle signals return to earth. There may be lots of stations he wants to work in this zone around New England. I may be wrong, but I’d have thought he needed some high-angle radiation as well.

Too right. The horizontal section would radiate some high-angle, making it work fairly well in the local area once skywave propagation opens up in the evening. I found it interesting in the series of articles published a year or two ago in QEX, that both using modelling and experimental results, the author found that with a smaller number of  radials, for example 8 or less, the efficiency was actually better when the radials were kept short (less than 1/8λ) than when they were extended out to a full 1/4λ.  He attributes that phenomenon to resonance effects of the quarter-wave length. OTOH, once a large number of radials is installed, the efficiency improves with additional length.

73,

Don


Logged

Don, K4KYV                                       AMI#5
Licensed since 1959 and not happy to be back on AM...    Never got off AM in the first place.

- - -
This message was typed using the DVORAK keyboard layout.
http://www.mwbrooks.com/dvorak
G3UUR
Member

Offline Offline

Posts: 141


« Reply #36 on: June 01, 2011, 03:38:49 PM »

Don,

I’m glad we’ve got that sorted out because I wouldn’t want Gary (AE1CT) to be left in any doubt about what the consequences of choosing one antenna type over another are going to be. To my mind there is one obvious one to try initially, and that’s the doublet with linear end-loading, but in restricted locations you’ve just got to try everything to see what works best in that situation. Sometimes what seems to work best defies logic and explanation; but that’s the great thing about amateur radio – the serendipity factor!

I don’t subscribe to QEX, so I missed the article you mentioned. However, I’m not surprised that fewer radials work better if they’re shorter. The dielectric constant of the various types of soil and rock varies by nearly a factor of 10, but if you take an average value you’d expect a radial buried deep in soil to be about one-third the length that it is in free space for the same resonant frequency. The dielectric medium doesn’t normally completely surround the radial for a significant distance in all directions, of course, because it usually isn’t buried that deep, so it might need to be half the normal length for resonance in free space instead of one third. 

 You certainly have some set-up for 160 and 80m there. I don’t know what power you run on 160m AM, but a tuning unit designed for use with anything up to a half-wave end-fed on 160 would probably be designed for a Q of 10 at 5kohm max. Your half-sized dipole on 160 would look more like about 40kohm resistive at the bottom of the 119ft feeder with an extra 60ft switched in circuit. That’s still a pretty high impedance to match and any parallel tuner would be working overtime unless the inductance was incredibly large (335uH?). I’d be inclined to ditch the extra 60ft loading section and try to match the rest with the simplest possible series arrangement and keep the Q as low as I can. It’s going to be around a 50, anyway, because the 12 ohms radiation resistance is going to drop to 6 halfway down the feeder, and then be transformed back up to 12 in series with +j600 at the bottom. You should be able to tune that with a pair of 350pF series variable capacitors in the feeder legs and match to 50 ohms with a shunt 2.35uH coil. That should give you an improvement in bandwidth.

Dave.
Logged

Dave,G3UUR
Vintage AM from the East of England
KM1H
Contributing
Member
*
Offline Offline

Posts: 3514



« Reply #37 on: June 02, 2011, 09:46:30 AM »

Quote
There is no guarantee that the beverage will be noise free. It, also, responds predominantly to vertically polarised signals.  It kills the noise by virtue of its horizontal selectivity - only selected sectors of the 360° pie are received, cutting out a lot of the noise that would be picked up by an omnidirectional vertical.  If the noise source happens to be in the same direction as the bore-sight of the beverage, you are SOL.

While a Beverage responds to vertically polarized signals they are predominantly skywave and not surface wave which is how local noise travels. If its aimed right at a close noise source then its a given it will pick up a lot.  Ground conductivity is a factor and here less is better; my RF ground is extremely poor, about 200 Ohms which seriously reduces groundwave pickup; even local BCB stations are relatively weak and dont bother 160.  Noise pickup here is gone in about a 1/4 mile, a recent local power outage gave me a chance to run the HRO-500 on a 12V battery and silence was complete even tho the houses over about 1/4 mile away still had power.

On 6M its completely different, I can pinpoint line noise 1-3 miles away.

Carl
Logged
Steve - K4HX
Guest
« Reply #38 on: June 02, 2011, 09:59:51 AM »

Here’s an interesting article on radials and getting the most out of a given amount of wire. It predates the QEX articles.




I don’t subscribe to QEX, so I missed the article you mentioned. However, I’m not surprised that fewer radials work better if they’re shorter. The dielectric constant of the various types of soil and rock varies by nearly a factor of 10, but if you take an average value you’d expect a radial buried deep in soil to be about one-third the length that it is in free space for the same resonant frequency. The dielectric medium doesn’t normally completely surround the radial for a significant distance in all directions, of course, because it usually isn’t buried that deep, so it might need to be half the normal length for resonance in free space instead of one third.  


* k3lcmaxgainradials.pdf (674.1 KB - downloaded 888 times.)
Logged
k4kyv
Contributing Member
Don
Member

Offline Offline

Posts: 10037



« Reply #39 on: June 02, 2011, 12:47:36 PM »

You certainly have some set-up for 160 and 80m there. I don’t know what power you run on 160m AM, but a tuning unit designed for use with anything up to a half-wave end-fed on 160 would probably be designed for a Q of 10 at 5kohm max. Your half-sized dipole on 160 would look more like about 40kohm resistive at the bottom of the 119ft feeder with an extra 60ft switched in circuit. That’s still a pretty high impedance to match and any parallel tuner would be working overtime unless the inductance was incredibly large (335uH?). I’d be inclined to ditch the extra 60ft loading section and try to match the rest with the simplest possible series arrangement and keep the Q as low as I can. It’s going to be around a 50, anyway, because the 12 ohms radiation resistance is going to drop to 6 halfway down the feeder, and then be transformed back up to 12 in series with +j600 at the bottom. You should be able to tune that with a pair of 350pF series variable capacitors in the feeder legs and match to 50 ohms with a shunt 2.35uH coil. That should give you an improvement in bandwidth.

I didn't take the trouble to calculate the inductance of the coil; it is the balanced tank coil I had on hand that came out of an old AM broadcast transmitter that used a push-pull final. It is resonated on 160m with a total of about 180 pf across the coil (300 pf/section split stator with 50 pf fixed vacuum in parallel with the whole thing). The link has 4 or 5 turns directly fed by the 50Ω transmission line, with no series or parallel capacitor. Matching is achieved using clips to trim the number of turns on the main coil at resonance; about 10% of each half of the coil is left unused at each end. The 440Ω OWL is parallel fed directly across the parallel tuned circuit. This set-up has been run up to about 800 watts carrier at 130+% positive peak modulation, and the variable capacitor, with 7 kv spacing, never arced over, and the rf voltages don't seem excessive. I might try your suggestion, and compare the bandwidths and efficiencies (DC input to rf amps in the feed line). I always figured that the sharp tuning indicated low loss in the system; at the low feed impedance and high SWR on the OWL, resistive losses in the system would lower the Q and broaden the tuning. Since this was originally a stop-gap measure for getting on 160m until I could get the ground radials laid, and it works well, I never analysed the circuit any further.

Looks like your suggestion uses series capacitors to electrically shorten the 3/8 λ total dipole leg+OWL length to a non-reactive current loop, excited by series-feed at the ATU.  Since the feedline loading capacitors and ATU series tuning capacitors are themselves in series, your 350 pf variables could be thought of as the composite of the series loading capacitances and the ATU series resonating capacitances, as the equivalent of two capacitances in series, kind of like a PI-network can be thought of as one L-network from the PA tube working into a second L-network that feeds the  load.

Steve, that is a good article on ground radials.  Thanks for posting the link. I wasn't aware that there were technical articles in the contesting rag.
Logged

Don, K4KYV                                       AMI#5
Licensed since 1959 and not happy to be back on AM...    Never got off AM in the first place.

- - -
This message was typed using the DVORAK keyboard layout.
http://www.mwbrooks.com/dvorak
K5UJ
Contributing
Member
*
Offline Offline

Posts: 2814



WWW
« Reply #40 on: June 02, 2011, 02:06:07 PM »


<<Here’s an interesting article on radials and getting the most out of a given amount of wire. It predates the QEX articles.>>

At the risk of being considered an irritant once again by not seeming to offer encouragment, I read that article not long after it came out and found its premise flawed.   The idea of having a finite amount of wire and planning an entire ground system with that as a limiting factor is completely wrong, unless all wire and cable manufacturing ceases and the cost of remaining supplies goes up by an order of magnitude or more.

It is much wiser to go ahead with the best most complete ground system possible within the limits of one's property lines, and upon running out of wire, halt the installation until more can be purchased and continue in increments until finished, rather than install a compromise ground system based on N feet of wire on hand, considering the entire project finished, once the current wire supply is exhausted.

Where this method might make sense is for a temporary ground system for a dxpedition antenna or Field Day.


 
Logged

"Not taking crap or giving it is a pretty good lifestyle."--Frank
flintstone mop
Contributing
Member
*
Offline Offline

Posts: 5048


« Reply #41 on: June 02, 2011, 02:40:19 PM »

Here’s an interesting article on radials and getting the most out of a given amount of wire. It predates the QEX articles.




I don’t subscribe to QEX, so I missed the article you mentioned. However, I’m not surprised that fewer radials work better if they’re shorter. The dielectric constant of the various types of soil and rock varies by nearly a factor of 10, but if you take an average value you’d expect a radial buried deep in soil to be about one-third the length that it is in free space for the same resonant frequency. The dielectric medium doesn’t normally completely surround the radial for a significant distance in all directions, of course, because it usually isn’t buried that deep, so it might need to be half the normal length for resonance in free space instead of one third.  



And read on that the article is talking about a 1/4 wave vertical. So us folks messing with miracle 27 foot high masts and top loading and capacity hats or even more magical 43 foot 'antennas' are not going to benefit from the findings.
Logged

Fred KC4MOP
W3GMS
Contributing
Member
*
Offline Offline

Posts: 3048



« Reply #42 on: June 02, 2011, 04:53:50 PM »


"This set-up has been run up to about 800 watts carrier at 130+% positive peak modulation"

Hi Don,
Now I know why your so strong up this way Smiley
Joe, W3GMS   
Logged

Simplicity is the Elegance of Design---W3GMS
G3UUR
Member

Offline Offline

Posts: 141


« Reply #43 on: June 02, 2011, 05:22:10 PM »

Although I can see why the radials might need to be resonant if you use very few of them, I still believe that saturating the ground at the base of the vertical with copper wire is the best way to go, especially if the length of the radials is very limited. I favor putting down whatever I can fit in until I no longer measure any improvement in field strength at a distance.

I’m suspicious of these NEC4 simulations where the effect of ground is critical to the results and the authors have not done any controlled experiments to verify even a tiny part of their simulated data. I’ve heard that even NEC4D, the professional version, does not simulate ground effects very well. Because of this, and also because it’s sensible, they ought to be using simulations in conjunction with experimental results to get a better understanding of how things work and gain more insight, not blindly using them without checking.

Don, I think the sharp tuning you observe on 160m with your current set-up is more to do with the tremendous working Q of the tuning unit than the low loss of your system as a whole. You’re probably working with a Q of over 100 at the moment. The capacitor can take it because the voltage rating of variable capacitors in the old days was usually based on the DC supply voltage and not the RF, so your capacitors should be able to stand 14kV peak with some margin. 

 I should warn you that my suggestion for matching your 80m dipole on 160m with just the quarter-wavelength of open-wire feeder is based on free space values of radiation resistance and reactance for the dipole, so you might need to vary the shunt inductor and series capacitors a bit to get the best match to 50 ohms. You’re right, each of the series capacitors can be considered to be a composite of two in series, one canceling the reactance looking into the feed line and the other providing the matching to 50 ohms with the shunt inductor. In this case the capacitor in each leg of the feeder is set at a lower capacitance than that required just for reactance cancellation.

There is another possibility, which you might like to consider, and that is to set the series capacitors at a slightly higher capacitance so that the inductive reactance is not fully canceled and there is some residual inductance that can be used with a shunt capacitor across the feeders at the 50-ohm input end. This will require a capacitor of around 3000pF plus or minus some, depending on the exact value of the radiation resistance of the 80m dipole on 160m. It might be worth comparing the two versions to see which one gives the broadest bandwidth.

Dave. 
Logged

Dave,G3UUR
Vintage AM from the East of England
k4kyv
Contributing Member
Don
Member

Offline Offline

Posts: 10037



« Reply #44 on: June 02, 2011, 06:10:54 PM »

At the risk of being considered an irritant once again by not seeming to offer encouragement, I read that article not long after it came out and found its premise flawed.   The idea of having a finite amount of wire and planning an entire ground system with that as a limiting factor is completely wrong, unless all wire and cable manufacturing ceases and the cost of remaining supplies goes up by an order of magnitude or more.

That is true.  Starting out designing a limited radial system based only on a roll of wire on hand and letting that be it seems too Hammy Hambone-ish to me, but the articles are interesting regarding the most efficient use of a given amount of wire. Perhaps for the first round one could lay a limited number of short radials for the most efficient use of the short roll, and later on plan to inter-lay longer radials in between. Some broadcast installations double up the number near the base using shorter radial wires, with the normal count of 120 on out to a quarter wavelength. This has proved to be a more efficient use of copper than using a mesh screen near the base of the tower. Of course, we must remember that regardless of length, radials are not "resonant" when buried in the earth as they are when elevated above-ground.

I planned way ahead with mine, with the idea of some day having a place to lay a full commercial grade radial system. I found a source of affordable wire, bought a 16,000 ft. roll of #12 bare soft-drawn, and stored it away in the event I would ever be able to use it.  About 6 years later, moved back here and a couple of years later the wire was buried. It had been in storage for 8 years by the time I installed the radials.

In retrospect, I might have been better off just burying 60 quarter-wave radials, and saving the rest of the wire for additional radial systems, but at the time I never suspected that copper was going to turn into a semi-precious metal. From numerous charts I have seen, the critical point of diminishing returns occurs at about 60 quarter-wave radials, and the difference between 60 and 120 is a tiny fraction of a dB. Or I might have used more than 60 but fewer than 120, making them all longer than 1/4λ, although that might have taken up more field space than I would have wanted.



Dave,

I want to keep the system simple, so that only one capacitor, the split stator, is varied with frequency, now that I am using 5 separate capacitors in 5 separate tuners ganged together to the reversible motor in the antenna tuner. With the series arrangement, two capacitors would have to be ganged together with insulated shafts and frames. I don't think the idea of splitting the coil and feeding the resonant feeders at the gap would work, given the high amount of reactance that would be inserted into the inductor at the gap. Feeding at the gap would work with a lo-Z non-reactive load, and I could still use the split stator, but in this case the load is highly reactive. It would take a lot of mechanical re-building to use two isolated capacitors instead of the one big split stator.

But I still might throw together an experimental set-up just to see how it works. In any case, I plan to sooner or later replace the 50Ω transmission line with 440Ω OWL, with the tuner used to convert balanced flat OWL line from the shack to the resonant feeders that run from the tuner at the base, up the tower, to the dipole. The long total length of OWL would complicate using resonant feeders all the way from the dipole, down the tower,  to the transmitter in the shack, because the cumulative error when QSYing across certain bands, resulting from the large count of quarter-wavelengths in the resonant line, would make it difficult to maintain series or parallel tuning with one simple split stator variable capacitor all the way across each entire band. Plus, the 140 ft. run from the shack to the tower would have less loss if it ran as a flat line  than if it ran as tuned feeders.

Logged

Don, K4KYV                                       AMI#5
Licensed since 1959 and not happy to be back on AM...    Never got off AM in the first place.

- - -
This message was typed using the DVORAK keyboard layout.
http://www.mwbrooks.com/dvorak
KM1H
Contributing
Member
*
Offline Offline

Posts: 3514



« Reply #45 on: June 02, 2011, 08:53:32 PM »

Quote
Steve, that is a good article on ground radials.  Thanks for posting the link. I wasn't aware that there were technical articles in the contesting rag.

The NCJ is about the only place you will find quality articles published. Contesters take performance seriously.

As far as 43', W1FV has been using that length on 80/160 since the 80's in 4 squares and has published a few times in the NCJ and maybe QEX. John is a one of those MIT guys that knows his stuff.

Carl
Logged
Steve - K4HX
Guest
« Reply #46 on: June 02, 2011, 09:55:42 PM »

Don:

With 119 feet of feedline on your 80 M dipole, the impedance at the end of the feedline on 160 meters is like 8 - j436. Adding another 60 feet yields 7 + j70. Seems the latter would be easier to tune.

An Inverted-L, 60 feet up and 60 feet out will have little high angle radiation. Over ground with average conductivity and with 30 quarter-wave radials, the peak is at 26 degrees. At 60 degrees elevation the L is down about 8 dB compared to a dipole at 60 feet. At 70 and 80 degrees respectively, the L is down about 10 and 12 dB compared to a dipole.

The original poster doesn't have space for a full-sized dipole, so the  L might be his only option. But for local work, even a short dipole would probably be better.
Logged
k4kyv
Contributing Member
Don
Member

Offline Offline

Posts: 10037



« Reply #47 on: June 02, 2011, 11:26:02 PM »


The NCJ is about the only place you will find quality articles published. Contesters take performance seriously.


Looks like further fragmentation of technical information published by the League.  Once upon a time all the technical information was published in QST, the Handbook plus, for those interested, the speciality handbooks (antennas, SSB, Hints & Kinks, mobile etc). Now you have to subscribe to three periodicals in addition to purchasing the speciality publications to get all the technical articles... and the least fruitful of these is QST, the default membership publication.  However, lately I have seen some improvement in the technical content in QST.
Logged

Don, K4KYV                                       AMI#5
Licensed since 1959 and not happy to be back on AM...    Never got off AM in the first place.

- - -
This message was typed using the DVORAK keyboard layout.
http://www.mwbrooks.com/dvorak
G3UUR
Member

Offline Offline

Posts: 141


« Reply #48 on: June 03, 2011, 04:17:57 AM »

Steve, while I agree with you on two points, I strongly disagree with you on the third. Taking the one where we disagree first; a short dipole on 80m at 120ft will have a radiation resistance between 12 and 16 ohms and look capacitive on 160m. Roughly halfway down the 119ft OWL section it will look purely resistive at 6 to 8 ohms. If the OWL is 500 to 600 ohms the low resistance at the halfway point will be transformed to an inductive reactance in series with a low resistance at the bottom because it’s a further one-eighth of a wavelength to that point.

I agree that the amount of power radiated from the top section of an inverted-L drops as the height of a quarter-wave inverted-L is increased, and pointed out to Don in a previous posting that he was not comparing like with like since Gary (AE1CT) would be using only about 30ft of vertical and a 100ft top.

I agree that Gary’s best approach might be an inverted-L in the long run, but the folded horizontal doublet would be easier to implement in the first place given that he will have to put in a lot of work to lay down a satisfactory ground system for the inverted-L. Using a doublet for a season would also allow him to assess his noise situation with a balanced antenna.

Don, if you only need to shift about 5 kHz before you notice a significant increase in SWR, then your system -3dB bandwidth could be as low as 15 kHz. This corresponds to a system working Q of 127!!!  The Q of your antenna and feeder system alone ought to be 40 to 50, so you should to be able to gain an increase in bandwidth of well over 2 if you use the simplest matching arrangement possible. That’s what I’m suggesting.

You only need two ganged, isolated variable capacitors and one small inductor (or possibly another large fixed capacitor) to implement the tuning system that I’m advocating. This is the simplest possible arrangement you can use to match your antenna system with an 80m dipole on 160m. There is no simpler way, unless you do the matching at the top of the tower, and it doesn’t increase the overall working Q, because that is already defined by the reactance and radiation resistance and with my simple matching and tuning arrangement you’re making it no worse. The Q of the 50-ohm matching circuit is under 2 and hardly reduces the bandwidth at all.

You add to the overall Q of the system by using a matching unit that has a further tuned circuit, especially if it doesn’t have a large enough inductance for the high impedance to be matched. That’s what you’re currently doing. You may also be losing about 2dB in the tuner if the Q of your coil is 350 or so, but it could be worse if the Q is lower. Simple is best!

Carl, many 160m DXers in the UK used 40ft verticals, both top loaded and folded monopole, from small suburban plots to work all over the world using very low power in the ‘60s and ‘70s. There are probably some using the same arrangement these days but we're allowed higher power now. I’ve worked W1BB in mid summer with 2.5 watts input and a 40ft folded monopole on 160m, so you can have great fun and do quite well with a modest set-up, but you can’t compete in a pile up with the Big Guns. You’ve got to know your propagation and be ahead of everyone else to snag the DX before the gang turn up. That’s harder to do now because everyone is a propagation expert with all these computer aids.

Dave.
Logged

Dave,G3UUR
Vintage AM from the East of England
KM1H
Contributing
Member
*
Offline Offline

Posts: 3514



« Reply #49 on: June 03, 2011, 08:15:30 AM »

Dave, the 40-43' has been popular for a long time and was really popularized when the US was given the  full band and 1500W in the late 70's. Then 160 was included on DXpeditions an the 43' gained popularity. W1FV's contribution was to produce a high performing 4 Square with shortened radiators and only 1/8 wave spacing on 160 using the 80M array he had been already using.

Yes, 160 propagation is amazing at times, Ive worked 13 countries with 100mw on nights when every EU with 100W or less to a wet noodle was at least 20 over S9.

I was looking at a full 1/4 wave L for my 500 KHz antenna and didnt like the modeling results with only 170' being vertical. Efficiency and bandwidth were better with 4 140' sloping top hat wires than the roughly single 325' top wire.

Carl
Logged
Pages: 1 [2] 3 ... 5   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone © 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.058 seconds with 19 queries.