Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Anyone recommend a source of 36:50 ohm ununs?

147 views
Skip to first unread message

Robert Smits

unread,
Jul 25, 2016, 6:18:26 PM7/25/16
to
Hi. Can anyone recommend a source of 36:50 ohm ununs?

I'm adding an inverted L for 160 meters and would like to add an unun of 36
ohm to 50 ohms to reduce my SWR. EZNEC shows it would lower it
substantially and when you're running barefoot on 160 you need all the help
you can get.

Thanks, Bob VE7HS
--
bobrsmits.ca

Dave Platt

unread,
Jul 25, 2016, 7:26:56 PM7/25/16
to
>Hi. Can anyone recommend a source of 36:50 ohm ununs?
>
>I'm adding an inverted L for 160 meters and would like to add an unun of 36
>ohm to 50 ohms to reduce my SWR. EZNEC shows it would lower it
>substantially and when you're running barefoot on 160 you need all the help
>you can get.

One gotcha on this: before you add such an unun, make sure that your
feedpoint impedance is actually close to 36 ohms. It may not be,
depending on your actual soil and ground-plane impedance and losses,
and an unun of this sort might actually make matters worse.

One of the local ham clubs has a regular Field Day practice of sending
up a quarter-wave vertical wire, attached to a helium balloon, to
operate on 160 meters at night. They had planned to use an unun of
around that ratio to match it to the feedline... but when they tried,
it made the SWR at the transmitter rather worse.

I suggested that since they'd only run four radials (along the ground
surface), they had a rather poor ground plane and probably had quite a
few ohms of ground-loss resistance at the feedpoint. The simple
assumption of "quarter-wave vertical over a perfect ground plane" just
didn't apply very well. I think they just chose to drive the antenna
directly and live with the SWR on the line. It's possible that they
chose to hook up the unun "backwards".

Depending on the soil, and radials installed below your Inverted L,
you might encounter the same issue. And, depending on your soil type
and weather, it's possible that your ground losses might vary quite
a bit over the course of a year.

Adding an adjustable transmatch right at the base of the antenna would
have been a more flexible solution for them - not as broadband as an
unun, but more adaptable to different ground-loss conditions. Might
work for you, too (although I grant it's likely to need a rather large
inductor!)

Also... I'd suggest calculating the actual "excess loss" in your coax,
if you accept the mismatch at the feedpoint and match to 50 ohms at
the rig. At 2 MHz I'd guess that the excess loss from even a 2:1 SWR
would be quite low.

(My copy of "Reflections" is at home, and I can't recall whether
Walt Maxwell W2DU provides a winding formula for a 36:50, if
that's actually what you do turn out to need. I'll try to remember
to check. My guess is that you'd probably end up having to wind your
own.)




Robert Smits

unread,
Jul 25, 2016, 8:02:28 PM7/25/16
to
Dave Platt wrote:

>>Hi. Can anyone recommend a source of 36:50 ohm ununs?
>>
>>I'm adding an inverted L for 160 meters and would like to add an unun of
>>36 ohm to 50 ohms to reduce my SWR. EZNEC shows it would lower it
>>substantially and when you're running barefoot on 160 you need all the
>>help you can get.
>
> One gotcha on this: before you add such an unun, make sure that your
> feedpoint impedance is actually close to 36 ohms. It may not be,
> depending on your actual soil and ground-plane impedance and losses,
> and an unun of this sort might actually make matters worse.

Yes, I'm considering that, and figure if it occurs I'll add more radials.
>
> One of the local ham clubs has a regular Field Day practice of sending
> up a quarter-wave vertical wire, attached to a helium balloon, to
> operate on 160 meters at night. They had planned to use an unun of
> around that ratio to match it to the feedline... but when they tried,
> it made the SWR at the transmitter rather worse.
>
> I suggested that since they'd only run four radials (along the ground
> surface), they had a rather poor ground plane and probably had quite a
> few ohms of ground-loss resistance at the feedpoint. The simple
> assumption of "quarter-wave vertical over a perfect ground plane" just
> didn't apply very well. I think they just chose to drive the antenna
> directly and live with the SWR on the line. It's possible that they
> chose to hook up the unun "backwards".
>
> Depending on the soil, and radials installed below your Inverted L,
> you might encounter the same issue. And, depending on your soil type
> and weather, it's possible that your ground losses might vary quite
> a bit over the course of a year.
>
> Adding an adjustable transmatch right at the base of the antenna would
> have been a more flexible solution for them - not as broadband as an
> unun, but more adaptable to different ground-loss conditions. Might
> work for you, too (although I grant it's likely to need a rather large
> inductor!)

Large Inductors aren't a problem - keeping a tuner weathertight on BC's wet
coast is, however.
>
> Also... I'd suggest calculating the actual "excess loss" in your coax,
> if you accept the mismatch at the feedpoint and match to 50 ohms at
> the rig. At 2 MHz I'd guess that the excess loss from even a 2:1 SWR
> would be quite low.

How do I calculate "excess loss?

>
> (My copy of "Reflections" is at home, and I can't recall whether
> Walt Maxwell W2DU provides a winding formula for a 36:50, if
> that's actually what you do turn out to need. I'll try to remember
> to check. My guess is that you'd probably end up having to wind your
> own.)

Thanks - I'll see if I can find a copy.
--
bobrsmits.ca

Dave Platt

unread,
Jul 25, 2016, 8:25:55 PM7/25/16
to
In article <XO6dnQyjDrkSNQvK...@giganews.com>,
Robert Smits <b...@rsmits.ca> wrote:

>Large Inductors aren't a problem - keeping a tuner weathertight on BC's wet
>coast is, however.

Understood :-)

>> Also... I'd suggest calculating the actual "excess loss" in your coax,
>> if you accept the mismatch at the feedpoint and match to 50 ohms at
>> the rig. At 2 MHz I'd guess that the excess loss from even a 2:1 SWR
>> would be quite low.
>
>How do I calculate "excess loss?

http://www.qsl.net/co8tw/Coax_Calculator.htm seems to work nicely.

Plug in the coax-cable type and length, the frequency, the load SWR,
and your input power. It's an approximation (the actual losses depend
not just on the SWR, but the actual load) but it should give you a
good idea of the tradeoffs.

It'll calculate:

- Matched loss (what you'd lose in the coax even if you have a
perfect 1:1 match at the antenna) in dB
- SWR loss (what I referred to as excess loss - additional power
lost in the coax due to the higher currents caused by a higher
SWR) in dB
- Total loss (the sum of the two)
- Actual power into the antenna

As an example: Belden 8237 (RG-8 type), 200 feet, at 2 MHz, into a
3:1 load. The program calculates a matched loss of .487 dB, and an
SWR loss of .281 dB, for a total of .768 dB.

Not a whole lot. If you had a perfect lossless match at the antenna
base, you'd cut your coax loss by about 40%, and gain .281 dB of
signal.

Then, compare that to ground losses. Based on a cursory glance at
some figures on the web, it looks as if a quarter-wave vertical
monopole installation can easily lose 2-3 dB in the ground, even with
as many as 24 radials. Might need up to 60 radials to get down to 1
dB of ground loss. Matching or not-matching the antenna to the
feedline won't affect these losses significantly.

So, it may be that SWR loss in your coax isn't the best place to spend
your efforts... at least, not at first. Adding more radials, or
elevating the radials above ground may have a greater payback.


amdx

unread,
Jul 25, 2016, 10:33:04 PM7/25/16
to
I don't have a source, but I have Jerry Sevick's book,
"Transmission Line Transformers"fourth edition.
On pages 7-4 thru 7-15 are descriptions of how to wind
32 ohm to 50 ohm UNUNs.
I found this, >
http://documents.mx/documents/amidon-transmissionlinetransformershandbook.html
Go to page 39 of the book (pg 61 of the docslide) and see the 5 winding
configuration to give you 32 ohm to 50 ohm.


If you search W2FMI-2:1-HU50 using Google it is the first hit
and show several approaches. I think this is the same as above.

Mikek


Ralph Mowery

unread,
Jul 25, 2016, 11:16:48 PM7/25/16
to
In article <d9ni6d-...@coop.radagast.org>, dpl...@coop.radagast.org
says...
>
> Not a whole lot. If you had a perfect lossless match at the antenna
> base, you'd cut your coax loss by about 40%, and gain .281 dB of
> signal.
>
> Then, compare that to ground losses. Based on a cursory glance at
> some figures on the web, it looks as if a quarter-wave vertical
> monopole installation can easily lose 2-3 dB in the ground, even with
> as many as 24 radials. Might need up to 60 radials to get down to 1
> dB of ground loss. Matching or not-matching the antenna to the
> feedline won't affect these losses significantly.
>
> So, it may be that SWR loss in your coax isn't the best place to spend
> your efforts... at least, not at first. Adding more radials, or
> elevating the radials above ground may have a greater payback.

Unless something unusual is going on he will probably burn up way more
in the unun than if it is not used. When below around 8 MHz coax has
very low aditional loss for a SWR of 3 or 4 to 1.

Not sure of the antenna being used, but if it needs a radial system,
that would be the place to spend time and money.




---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

Rob

unread,
Jul 26, 2016, 3:54:52 AM7/26/16
to
Robert Smits <b...@rsmits.ca> wrote:
> Dave Platt wrote:
>
>>>Hi. Can anyone recommend a source of 36:50 ohm ununs?
>>>
>>>I'm adding an inverted L for 160 meters and would like to add an unun of
>>>36 ohm to 50 ohms to reduce my SWR. EZNEC shows it would lower it
>>>substantially and when you're running barefoot on 160 you need all the
>>>help you can get.
>>
>> One gotcha on this: before you add such an unun, make sure that your
>> feedpoint impedance is actually close to 36 ohms. It may not be,
>> depending on your actual soil and ground-plane impedance and losses,
>> and an unun of this sort might actually make matters worse.
>
> Yes, I'm considering that, and figure if it occurs I'll add more radials.

Wouldn't it be easier and better to first measure the impedance?

Allodoxaphobia

unread,
Jul 26, 2016, 10:13:54 AM7/26/16
to
On Mon, 25 Jul 2016 15:18:14 -0700, Robert Smits wrote:
> Hi. Can anyone recommend a source of 36:50 ohm ununs?
>
> I'm adding an inverted L for 160 meters and would like to add an unun of 36
> ohm to 50 ohms to reduce my SWR. EZNEC shows it would lower it
> substantially and when you're running barefoot on 160 you need all the help
> you can get.

Just run two sections of 1/4 wavelength 72 ohm coax in parallel as
a matching section. KISS.

73
Jonesy
--
Marvin L Jones | W3DHJ | W3DHJ | http://W3DHJ.net/
Pueblo, Colorado | @ | Jonesy | __
38.238N 104.547W | jonz.net | DM78rf | 73 SK
* Killfiling google & XXXXbanter.com: jonz.net/ng.htm

John S

unread,
Jul 26, 2016, 11:16:36 AM7/26/16
to
On 7/26/2016 9:13 AM, Allodoxaphobia wrote:
> On Mon, 25 Jul 2016 15:18:14 -0700, Robert Smits wrote:
>> Hi. Can anyone recommend a source of 36:50 ohm ununs?
>>
>> I'm adding an inverted L for 160 meters and would like to add an unun of 36
>> ohm to 50 ohms to reduce my SWR. EZNEC shows it would lower it
>> substantially and when you're running barefoot on 160 you need all the help
>> you can get.
>
> Just run two sections of 1/4 wavelength 72 ohm coax in parallel as
> a matching section. KISS.
>
> 73
> Jonesy

Where does the transformation to 50 ohms occur? If you have a 36 ohm
antenna and you build a 36 ohm transmission line with the paralleled 72
ohm lines (doesn't have to be 1/4 wavelength) you still have 36 ohms at
the transmitter end to deal with.



highlandham

unread,
Jul 26, 2016, 11:27:27 AM7/26/16
to
================================================
Ref:Building and using Baluns and Ununs
(Practical designs for the experimenter ) by Jerry Sewick , W2FMI SK

Publisher : CQ Communications .Inc in 1994 ISBN 0-943016-09-6

On page 71 (Chapter 8) a dual-ratio Unun 32 and 18 Ohm to 50 Ohm.

A very useful book possibly available from a library or perhaps via eBay

I bought it in the 1990s from Ham Radio Outlet (HRO) in the USA


Frank , GM0CSZ / KN6WH in IO87AT

Allodoxaphobia

unread,
Jul 26, 2016, 9:21:01 PM7/26/16
to
In rec.radio.amateur.antenna, you wrote:
> On 7/26/2016 9:13 AM, Allodoxaphobia wrote:
>> On Mon, 25 Jul 2016 15:18:14 -0700, Robert Smits wrote:
>>> Hi. Can anyone recommend a source of 36:50 ohm ununs?
>>>
>>> I'm adding an inverted L for 160 meters and would like to add an
unun of 36
>>> ohm to 50 ohms to reduce my SWR. EZNEC shows it would lower it
>>> substantially and when you're running barefoot on 160 you need all
the help
>>> you can get.
>>
>> Just run two sections of 1/4 wavelength 72 ohm coax in parallel as
>> a matching section. KISS.
>
> Where does the transformation to 50 ohms occur? If you have a 36 ohm
> antenna and you build a 36 ohm transmission line with the paralleled 72
> ohm lines (doesn't have to be 1/4 wavelength) you still have 36 ohms at
> the transmitter end to deal with.

Ahhh... You are so correct!
I read the OP "intently" and then went off half-cocked.
Apologies!!

So, for a revised "suggestion" I would try parallel sections
of 93 ohm coax -- RG-62 -- and see how that plays out.

73
Jonesy W3DHJ

John S

unread,
Jul 26, 2016, 11:05:25 PM7/26/16
to
Errr... how does that help? Paralleled 93 ohm coax results in a 46.5 ohm
transmission line. It might as well be a single 50 ohm line instead.

Rob

unread,
Jul 27, 2016, 3:41:11 AM7/27/16
to
The line needs to be sqrt(Zi*Zo) = sqrt(50*36) = 42.4 ohm.

Paralleled 93 ohm is probably the closest you can get using off-the-shelf
cable.

Roger Hayter

unread,
Jul 27, 2016, 4:19:24 AM7/27/16
to
How about 75 ohm in parallel with 93 ohm giving a theoretical 41.5 ohm?
Or doesn't this work?

--

Roger Hayter

Allodoxaphobia

unread,
Jul 27, 2016, 10:35:20 AM7/27/16
to
On Tue, 26 Jul 2016 22:06:12 -0500, John S wrote:
> On 7/26/2016 8:20 PM, Allodoxaphobia wrote:
>> In rec.radio.amateur.antenna, you wrote:
>>> On 7/26/2016 9:13 AM, Allodoxaphobia wrote:
>>>> On Mon, 25 Jul 2016 15:18:14 -0700, Robert Smits wrote:
>>>>> Hi. Can anyone recommend a source of 36:50 ohm ununs?
>>>>>
>>>>> I'm adding an inverted L for 160 meters and would like to add an
>> unun of 36
>>>>> ohm to 50 ohms to reduce my SWR. EZNEC shows it would lower it
>>>>> substantially and when you're running barefoot on 160 you need all
>> the help
>>>>> you can get.
>>>>
>>>> Just run two sections of 1/4 wavelength 72 ohm coax in parallel as
>>>> a matching section. KISS.
>>>
>>> Where does the transformation to 50 ohms occur? If you have a 36 ohm
>>> antenna and you build a 36 ohm transmission line with the paralleled 72
>>> ohm lines (doesn't have to be 1/4 wavelength) you still have 36 ohms at
>>> the transmitter end to deal with.
>>
>> Ahhh... You are so correct!
>> I read the OP "intently" and then went off half-cocked.
>> Apologies!!
>>
>> So, for a revised "suggestion" I would try parallel sections
>> of 93 ohm coax -- RG-62 -- and see how that plays out.
>
> Errr... how does that help? Paralleled 93 ohm coax results in a 46.5 ohm
> transmission line. It might as well be a single 50 ohm line instead.

We only have a "guess" in the OP that the "inverted L for 160 meters"
has a feedpoint impedance of 36 ohms -- probably from a book relying on
theoretical perfect conditions. This is AMATEUR RADIO! Try it.
See it it helps or hurts. Learn something from the experience.

Jonesy

John S

unread,
Jul 27, 2016, 1:28:18 PM7/27/16
to
There is no harm in using the tools available to get guidance. If the
results do not agree, then okay, you learned something about
simulations. Trying it and simulating it are both learning experiences.
It is nice when they connect even a little bit.

Robert Smits

unread,
Jul 27, 2016, 4:40:24 PM7/27/16
to
Thanks for all the suggestions.

I'll see what it's like with 50 ohm coax first, and then make other changes.

--
bobrsmits.ca

Ian Jackson

unread,
Jul 27, 2016, 5:43:16 PM7/27/16
to
In message <ccCdna7ngPqvgQTK...@giganews.com>, Robert Smits
<b...@rsmits.ca> writes
>Thanks for all the suggestions.
>
>I'll see what it's like with 50 ohm coax first, and then make other changes.
>
What about making the antenna somewhat longer than a quarterwave (which
presumably is what it is)? This will make the feedpoint impedance higher
resistive, and inductive. Make it long enough to get the resistance
close to 50 ohms, and then add series C to cancel the inductance.
--
Ian

John S

unread,
Jul 28, 2016, 3:19:06 AM7/28/16
to
Or, rather than fussing with the antenna, just use a 50 ohm .051
wavelength open stub located 0.113 wavelengths from the antenna. Use a
T-connector. Wavelengths are electrical lengths. Good for 1.8Mhz.

Feel free to check my numbers.

Ian Jackson

unread,
Jul 28, 2016, 4:03:59 AM7/28/16
to
In message <nncbl9$hmg$4...@dont-email.me>, John S <Sop...@invalid.org>
writes
I'll believe you, as my Smith chart skills are rusty at the moment!

However, has it been determined why the OP wants to try and achieve a
perfect match?

If it is to reduce losses in the coax, unless the coax is really long,
on 160m the losses due to having a 2:1 SWR are probably negligible.

If it is to present the TX with a perfect 50 ohm load, this can be done
at the shack end with the appropriate simple matcher.
--
Ian

John S

unread,
Jul 28, 2016, 10:37:12 AM7/28/16
to
On 7/28/2016 3:03 AM, Ian Jackson wrote:
> In message <nncbl9$hmg$4...@dont-email.me>, John S <Sop...@invalid.org>
> writes
>> On 7/27/2016 4:43 PM, Ian Jackson wrote:
>>> In message <ccCdna7ngPqvgQTK...@giganews.com>, Robert Smits
>>> <b...@rsmits.ca> writes
>>>> Thanks for all the suggestions.
>>>>
>>>> I'll see what it's like with 50 ohm coax first, and then make other
>>>> changes.
>>>>
>>> What about making the antenna somewhat longer than a quarterwave (which
>>> presumably is what it is)? This will make the feedpoint impedance higher
>>> resistive, and inductive. Make it long enough to get the resistance
>>> close to 50 ohms, and then add series C to cancel the inductance.
>>
>> Or, rather than fussing with the antenna, just use a 50 ohm .051
>> wavelength open stub located 0.113 wavelengths from the antenna. Use a
>> T-connector. Wavelengths are electrical lengths. Good for 1.8Mhz.
>>
>> Feel free to check my numbers.
>
> I'll believe you, as my Smith chart skills are rusty at the moment!
>
> However, has it been determined why the OP wants to try and achieve a
> perfect match?

Not entirely. Something about running "barefoot".

> If it is to reduce losses in the coax, unless the coax is really long,
> on 160m the losses due to having a 2:1 SWR are probably negligible.

Very true. And, actually, would be less than 1.4:1 SWR if his numbers
are correct.

> If it is to present the TX with a perfect 50 ohm load, this can be done
> at the shack end with the appropriate simple matcher.

Also very true. Much easier to adjust and maintain as well.
0 new messages