Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: [1] 2 Next   Go Down
  Print  
Author Topic: Ununs - Why use them with non-resonant verticals?  (Read 7688 times)
N4UM
Member

Posts: 476




Ignore
« on: June 05, 2011, 04:47:05 PM »

I've noticed that a number of antenna manufactures recommend the use of 4:1 UNUNs for use with muiltiband non-resonant verticals (typically 43 feet for 80-10, or 24 feet for 40-10M).  They also recommend the use of fairly long coax feedlines in the range to 100 to 150 feet (apparently even though the antenna might be  considerably closer to the transmitter.)  Vague claims of improving "efficiency" and reducing SWR on the coax are made but the basis for such claims never seems to be made explicit.

Perhaps someone out there who knows what they're talking about can enlighten me as to whether or not there is any real benefit to be had from using a 4:1 UNUN with a multiband non-resonant vertical - aside from lining the pockets of the purveyors of such devices.  I find the recommendation of using excessive lengths of feedline particularly suspicious.
Logged
N1UK
Member

Posts: 1492




Ignore
« Reply #1 on: June 05, 2011, 05:13:58 PM »

I would have to dig back and I don't mean to be vague but there were some articles on changing that UNUN to improve the efficiency of those 42 foot verticals.


Mark N1UK
Logged
N3OX
Member

Posts: 8847


WWW

Ignore
« Reply #2 on: June 05, 2011, 06:57:35 PM »

Perhaps someone out there who knows what they're talking about can enlighten me as to whether or not there is any real benefit to be had from using a 4:1 UNUN with a multiband non-resonant vertical -

Yes, in many/most cases, if you can achieve a very low loss 4:1 impedance transformation, it will improve the loss in the feedline and tuner.  It does almost always lower the SWR seen by the feedline.  And I think it's not too hard to build a fairly low loss UNUN for this situation, otherwise people with just 30 feet of good coax, a good tuner, and a kW amplifier would routinely blow up or overheat even a beefy UNUN.   I'm not saying it's an optimum approach, but it does HELP if you're going to run a coax fed antenna of this type.

Minimum coax length requirements are, of course, suspicious, but I think in some cases the main intention may be an attempt  to avoid very low resistance impedances at the tuner on 80m and/or 160m.

Just for one concrete example, let's look at the base impedance of a uniform diameter 1.5 inch 43 foot vertical on 3.5MHz.  The impedance there is about 11-j222.  Divide that by four with an excellent, low loss UNUN, and you'll get 2.75-j55.5.  Run that through 30 feet of LMR-400 (using http://vk1od.net/calc/tl/tllc.php) and you get a horrible impedance of 1.8-j2.5 and about 1.7dB extra line loss.  But this impedance is really bad.  W9CF's tuner simulator (http://fermi.la.asu.edu/w9cf/tuner/tuner.html) on default settings shows 3.7dB tuner loss, and then the inductor burns or melts on a long kW+ transmission.

Lengthen the coax to 100 feet, and you end up with a much more matchable 53-j211, with 3.3db total line loss, 1.6dB more loss in the line than at 30 feet.  But then the TUNER loss is just 0.7dB, an improvement of 3dB!  There's actually less loss TOTAL if you assume W9CF's default settings for your tuner.  100 feet of coax and that particular tuner configuration has 1.4dB MORE radiated power than the same tuner with 30 feet of coax.  A more capable higher power tuner, especially one that could be configured into an L-network, would be better for this, but it still might be a wash in terms of which had more loss.  And the "more line loss, easier impedance" makes it less likely to run into an "unmatchable" impedance at the shack end of the coax.

=======================

The main thing you need to do if you make this kind of thing for a living is to make sure nothing burns up at 1.5kW full duty cycle and that your users with their big expensive tuners don't get a "not matchable" impedance.  The radiated power is secondary or tertiary to that, but at the same time it's not such a simple case of intentionally adding loss to help the matching.

The loss tradeoffs of these antennas get really complicated if you take the whole system into account.  A good example is the radial field.  If you put in really good radials, it makes it more likely to get low impedances that dissipate a lot of power in the feedline and tuner.  It's possible that making the radial field worse will increase ground loss but reduce system loss for a net gain.  It gives me a headache to ponder these antennas too much anymore.

But you know what?  Antennas give lots of people headaches, and I think these fill a pretty important niche.  Buy some stuff, plug it in, run 1500W into it, and make lots of contacts.  Done.  These antennas accept full legal limit into the shack end of the system without blowing up (maybe), don't dissipate a shockingly obvious amount of the applied power like some of the antennas with intentional resistors and junk like that, and don't require any tuning.  With full legal limit applied, you probably end up getting 100W+ ERP on 160m, 500+ W ERP on 80m and 750W-1kW ERP on the higher bands if you install one of these things with the appropriate "minimum length" of coax and a UNUN.

That works for a lot of folks.
« Last Edit: June 05, 2011, 07:01:20 PM by N3OX » Logged

73,
Dan
http://www.n3ox.net

Monkey/silicon cyborg, beeping at rocks since 1995.
AA4PB
Member

Posts: 12898




Ignore
« Reply #3 on: June 06, 2011, 09:03:09 AM »

Look at it this way: If your antenna feed impedance were 200 Ohms then the SWR with directly connected 50 Ohm coax would be 4:1. If you put in a 4:1 UNUN then it converts the coax 50 Ohms to 200 Ohms and you have a perfect match (1:1 SWR). The problem is that an all-band non-resonant antenna is not going to have a 200 Ohm feed impedance on all frequencies - maybe not on any. However, the impedance is probably high enough on most frequencies that the 200 Ohm output of the UNUN provides a better match than the 50 Ohm coax.

I think the idea of using longer lengths of coax is to introduce enough loss to provide a load that your tuner is more likely to be able to match.

I'm not a very big fan of these antennas unless you use an automatic tuner outside at the base of the antenna. When you try to use one with a tuner in the shack connected to the antenna base via coax you are going to be operating the coax at a very high SWR on many bands. High SWR means increased losses. People like them because they are all band, simple, and fairly easy to install.  They make contacts because even if your are disipating most of your 100W in losses you still radiate a few watts. Personally, I think there are better (and certainly more efficient) antenna designs available.
Logged
N4UM
Member

Posts: 476




Ignore
« Reply #4 on: June 06, 2011, 04:57:35 PM »

Thanks very much to those of you who replied.  I found the example using W9CF's tuner simulator to be quite instructive and confess I had never even considered losses in the tuner.  DUH!  W9CF's tuner simulator appears to be  a very useful program, right up there with VK1OD's transmission line loss program.  It also now lives on my "Favorites" bar.

I had been feeding my 23 foot flagpole with a remote autotuner on my porch thru 15 feet of Davis BuryFlex and living with the losses on the short run of coax which appeared to be acceptable.  The tuner would balk every now and then and I'd have to go out on the porch and  play with it to get it back on track.  I wound up moving the tuner into the shack because I also wanted to use it to tune an attic antenna and now have 90 feet of coax between the tuner and the flagpole  - hence my concern with SWR on the longer piece of coax.

I can see where the UNUN might be beneficial on 20 thru 10 meters but suspect it might make the SWR on the coax worse on 30 and 40 meters.  Of course high SWR on 30 and 40 should be more tolerable than on the higher bands so perhaps a UNUN might be worth trying.   Sometimes convenience is a worthwhile exchange for efficiency!
Logged
WD4ELG
Member

Posts: 875




Ignore
« Reply #5 on: June 07, 2011, 11:20:39 AM »

DISCLAIMER: I am no expert, but I have followed some good guidance from VK1OD and some other experts, and have tinkered with the 43 foot antenna for a couple of years.

N4UM, great thread.  Who says eHam is of no value.  You're getting some awesome guidance here.   BTW, lots of folks (me 2) have often failed to consider the entire system for losses.

I won't comment on the feedline length recommendations for chaning those up, except to say I have never understood the logic of that recommendation.  But I am not an antenna manufacturer.

I agree with AA4PB's comments, although I have seen my installation work quite well on 40 and 30 meters when stealth and multiband operation was required.  (Separate note on the 43 footer, higher angle of radiation on 15/12/10 means it won't work as well for DX when other antenna solutions are available...like a shorter vertical!). 

As VK1OD points out on his website (and he has schooled me on this and many other topics, for which I am thankful), we need to define "good."

Good means maximum power radiated from the antenna (at least it does for me).  The trick is how to accomplish that.

Good does not necessarily mean lowest SWR at the shack.
Good does not necessarily mean lowest cost.
Good does not necessarily mean simplest.

In my case, after much trial and error, I have found that maximum power radiated from the antenna happens when I do two things:

1. Use a remote tuner at the base of the 43 foot vertical, along with as many radials as I can put down (poor soil).  This eliminates my concerns about SWR losses in coax, and consistently works around my issues with the unun.

2. Don't use the 43 footer on 160 meters, because it is not radiating much RF at all.  Instead I use a 132 foot inverted L.  (Surprisingly, the 43 footer in my case does work OK on 80 for 100+ DXCC entities at 100 watts).

3. Don't use the 43 footer on 15/12/10 unless it is short skip and/or there is no other alternative (unless DX is not the objective, and you want stateside only!)


http://vk1od.net/balun/application/Z5.html

http://vk1od.net/antenna/multibandunloadedvertical/index.htm

Keep us posted on your results, OM

Mark Lunday
WD4ELG
Logged
N4UM
Member

Posts: 476




Ignore
« Reply #6 on: June 17, 2011, 10:01:10 AM »

UPDATE:
I just installed a 4:1 UNUN from Balun Designs on my 23 foot flagpole and it seems to have improved the SWR on the higher bands (20 thru 10) as I had suspected it might.  My autotuner located 90 feet away at the rig has no difficulty getting a decent match on 40 thru 10.  SWR on the line is still relatively high on 30 and 40 - again, as expected.  The antenna should  still be useable on 30 and 40 since the coax losses at these frequencies are less than on the higher bands. 

All in all, I think the addition of the 4:1 UNUN  was beneficial.
Logged
W6RMK
Member

Posts: 656




Ignore
« Reply #7 on: June 17, 2011, 10:39:22 AM »

The other thing to remember is that a 4:1 UnUn isn't necessarily the same across all frequencies.  As the frequency goes up, the parasitic capacitance between the windings makes it more like a 1:1 than a 4:1..

So if you have a 'short' antenna, with a small feedpoint impedance on the lower bands, the 1:4 helps get it up closer to 50 ohms.  As the frequency goes up, the "leakage" around the transformer makes it more like a 1:1 so it stays with a decent match even though the feedpoint impedance of the antenna is now higher.

A good example of using the non-ideal behavior of the device to your advantage.
Here's some measured data on the UnUn at the base of a screwdriver (it's a 2:1 turns ratio: 4:1 impedance ratio).
http://home.earthlink.net/~w6rmk/antenna/screwdriver.htm
about half way down.. "Matching box"
Logged
N1ZHE
Member

Posts: 68




Ignore
« Reply #8 on: June 19, 2011, 09:53:45 AM »

I have a 30 foot vertical that was made for 6 meters through 40 meters. The manufacturer suggested a 4:1 unun to make the antenna easier to tune. This ended up being a bit of overkill as my Omni VII will tune a 10:1 swr. The tuner in it was originally designed or based on tuners built by LDG.

I started off with a 4:1 unun and I found the antenna with my tuner would easily load on 80 meters as well and work with fair to good results. But I had a real problem with RF in the shack so I replaced the unun with a choke or isolation 1:1 balun.

I can no longer tune 80 meters but that no real loss as it wasn't supposed to work there anyway. Besides, I'd like to have something more efficient for 80/160.

David, N1ZHE
Logged
G3TXQ
Member

Posts: 1524




Ignore
« Reply #9 on: June 19, 2011, 10:56:31 AM »

So if you have a 'short' antenna, with a small feedpoint impedance on the lower bands, the 1:4 helps get it up closer to 50 ohms.

That may be the case when you are loading the vertical to resonance; the opposite is more likely to be the case with an untuned vertical.

Take a 43ft vertical used on 80m; according to EZNEC its feedpoint impedance is about 12-j230 assuming an excellent ground system. Although the resistive part of the feedpoint impedance is low, there is a large capacitive reactance and we would want a UnUn which stepped the impedance down in order to improve the VSWR(50). Here are the VSWR(50) figures for that feedpoint impedance:

No UnUn: 92:1
4:1 UnUn: 38:1
9:1 UnUn: 47:1

My own observation on the UnUn is that they are often implemented using iron powder cores in order to reduce losses; as a result they rarely have adequate winding reactance. However, this doesn't seem to matter much in the "untuned vertical" application - the non-ideal impedance characteristics and the imperfect feedpoint impedance transformation simply get "mopped up" in the tuner.

73,
Steve G3TXQ

Logged
N3OX
Member

Posts: 8847


WWW

Ignore
« Reply #10 on: June 19, 2011, 11:11:49 AM »

My own observation on the UnUn is that they are often implemented using iron powder cores in order to reduce losses; as a result they rarely have adequate winding reactance. However, this doesn't seem to matter much in the "untuned vertical" application - the non-ideal impedance characteristics and the imperfect feedpoint impedance transformation simply get "mopped up" in the tuner.

An important question in the overall efficiency calculations is whether they actually are capable of performing an appropriate transformation to reduce the reactance presented to the coax.

I wonder how innocuous the departures from ideal 4:1 transformation are, even when the loss is low.
Logged

73,
Dan
http://www.n3ox.net

Monkey/silicon cyborg, beeping at rocks since 1995.
KI4SDY
Member

Posts: 1452




Ignore
« Reply #11 on: June 19, 2011, 12:11:25 PM »

After reading the highly informative and well qualified responses to your post from good hams that provide positive Elmering on eHam.net, here is my cheap vs. effective response. It was a move after my own cheap heart to add the 4:1 unun at the base of the vertical, but not the overall best solution.  Wink

If it was me, I would haunt eBay for a good working older manual tuner for inside the shack, the attic antenna and all future antennas. I have never paid more than $38.00 for a used tuner and they all worked great! Then return your auto tuner to the base of the outside vertical. If you saved the receipt for the 4/1 unun, return it! This will get you broader tuning range using both tuners, if you need it, but you shouldn't and your outside antenna will be much more efficient!  Grin  
« Last Edit: June 19, 2011, 12:37:15 PM by KI4SDY » Logged
G3TXQ
Member

Posts: 1524




Ignore
« Reply #12 on: June 19, 2011, 12:18:26 PM »

Dan,

That's an excellent question, and one which I want to try to answer given some "free" time.

I did some preliminary measurements a few months ago on an iron powder 4:1 UnUn which indicated that the departure from ideal actually improved the VSWR(50). At the time I thought it might be due to the inductive reactance of the UnUn cancelling to some extent the capacitive reactance at the feedpoint, but I need to investigate that apparent improvement a lot more.

What is clear is that for an untuned, electrically-short vertical, the impedance transformation needs to be "downwards" (or possibly 1:1 as it approaches quarter-wave resonance), not upwards as suggested by W6RMK

73,
Steve G3TXQ
Logged
G3TXQ
Member

Posts: 1524




Ignore
« Reply #13 on: June 19, 2011, 01:20:57 PM »

Dan,

I just located the results of those rough measurements in my lab notebook - they were even more startling than I remembered!

I had put a 12 Ohm resistor and a 180pF capacitor in series; they measured 10.6-j234 on my AIM analyser at 3.7MHz - similar to the impedance you might expect of a 43ft vertical. That would be a VSWR(50) of 108:1 with no impedance transformation.

If we had an ideal 4:1 UnUn the impedance would drop to 2.65-j58.5, improving the  VSWR(50) to 48:1.

However, I measured a VSWR(50) of 20:1 at the input to the UnUn, and an impedance of 737+j437 !!!! How come?

The UnUn was 13 bifilar turns on a T200-2 iron powder toroid - typical of what I see of Web designs. Its secondary reactance at 3.7MHz is around 190 Ohms which is getting close to resonating with the capacitive reactance of the load. If you look at the equivalent circuit of the UnUn with the load, you will see how that arrangement can transform the resistive component of the load up to a very high value.

In fact, "working backwards", a UnUn secondary impedance of 3+j226 would transform the load to 2948+j1748 across the secondary, and 737+j437 at the input.

73,
Steve G3TXQ
Logged
N4UM
Member

Posts: 476




Ignore
« Reply #14 on: June 19, 2011, 04:15:55 PM »

After reading the highly informative and well qualified responses to your post from good hams that provide positive Elmering on eHam.net, here is my cheap vs. effective response. It was a move after my own cheap heart to add the 4:1 unun at the base of the vertical, but not the overall best solution.  Wink

If it was me, I would haunt eBay for a good working older manual tuner for inside the shack, the attic antenna and all future antennas. I have never paid more than $38.00 for a used tuner and they all worked great! Then return your auto tuner to the base of the outside vertical. If you saved the receipt for the 4/1 unun, return it! This will get you broader tuning range using both tuners, if you need it, but you shouldn't and your outside antenna will be much more efficient!  Grin  

I'm using the MFJ 998 1500 Watt autotuner and it is definitely NOT waterproof.  I know of no one who manufactures a high power autotuner that is weatherproof.  Wish I did! I have several high power manual tuners but have gotten used to quick band changes with an autotuner.
 
BTW, when is MFJ going to use decent turns counters on their variable inductor tuners?  I have several of their tuners (both high and low powered) and the turns counter loses calibration very quickly on all of them.  The tuners work fairly well but are useless for quick band changes since the settings are always getting messed up.
Logged
Pages: [1] 2 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!