Re: Backward balun

[dw-4]

If you have an mfj analyzer and a 4:1 balun, you should be able to test
it.
Just connect up a 50 ohm resistor to the unbalanced side.
Set the analyzer for the frequency of interest and look at the "R" that
the analyzer measures.
Keep the leads going from the analyzer to the balun as short as
possible.

I wind my own baluns and use this to check them to see how broad banded
they are.

N1BBR
--
 [hidden email]

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[W8JI]

> If you have an mfj analyzer and a 4:1 balun, you should be able to test
> it.
> Just connect up a 50 ohm resistor to the unbalanced side.
> Set the analyzer for the frequency of interest and look at the "R" that
> the analyzer measures.
> Keep the leads going from the analyzer to the balun as short as
> possible.
>
> I wind my own baluns and use this to check them to see how broad banded
> they are.

Since he is stepping down, not up, he would need to measure the balanced
terminals with the analyzer while terminating the normally unbalanced 50 ohm
port with 12.5 ohms.

It won't be good, especially at higher frequencies, but some baluns with
short internal leads will work to some extent mismatched. For example the
DXE baluns only have about six electrical feet of 100 ohm line, and with a
4:1 only the upper transmission line is critical for impedance when balance
is satisfied. In a perfectly balanced load condition the upper core has no
flux, and the lower core has all the flux behaving as a traditional 4:1
voltage balun. Only when terminal voltages are unbalanced does the upper
core have flux.

http://www.w8ji.com/balun_single_core_41_analysis.htm

As for testing, if you do not test with a center tapped load (in the example
of a 1:1 balun, two 25 ohm resistors in series), and with the split load
grounded in the center, it is not really a balun test. You much force the
balanced terminals to be balanced to test how it acts as a balun. With a
single small resistor you are testing the system only as a transmission
line, and not really testing any balanced characteristics of the system.

73 Tom

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[alorona]

Thanks for all of the excellent advice.

To sum up, today I measured a 4:1 current balun backwards and Tom was pretty
right on... the high frequency cutoff is so severely lowered that it is unusable
for my application. I did not test a 4:1 voltage balun.

I think one approach to take is to use a unun to transform down to 12 ohms and
then a conventional 1:1 balun following that to transform to a balanced feed.
I'll try this and report back if it doesn't work.

I know that there were several who suggested Sevick's book-- and I hope I'm not
commiting sacrilege by saying that I've gotten very little understanding of
balun theory from that book. It's a great cookbook for constructing baluns (and
ununs), but not so great to learn why one is doing what he's doing. But that's
just me.

It is also very surprising that I found almost no reference to a 50-to-12.5 ohm
balun on the web. You can certainly find one for purchase, but you can't find
the instructions on how to wind one. You have to buy the book, I guess!

This is surprising to me because I can think of many applications where one
needs to transform 50 ohms to a lower (balanced) impedance, including the one I
am currently battling.

Thanks again to everybody.

Good weekend,

Al  W6LX
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[W8JI]

> I think one approach to take is to use a unun to transform down to 12 ohms
> and
> then a conventional 1:1 balun following that to transform to a balanced
> feed.
> I'll try this and report back if it doesn't work.

You will have trouble. Lead lengths will kill SWR (as will lengths through
connectors and connectors) because it is almost impossible to build 12.5 ohm
line and connections without special cables (very thin dielectrics and thick
center conductors).

Look at baluns for solid state amplifiers.

You can series connect two 25 ohm lines (each small 50 ohm cables) on the 50
ohm end, and parallel them on the 12.5 ohm end. Use an RG Teflon cable about
1/8th inch or so diameter.

Don't try the silly idea of a single core, build two isolated lines.

I've done this with solid state amps with bandwidths of 1-100 MHz.

> I know that there were several who suggested Sevick's book-- and I hope
> I'm not
> commiting sacrilege by saying that I've gotten very little understanding
> of
> balun theory from that book. It's a great cookbook for constructing baluns
> (and
> ununs), but not so great to learn why one is doing what he's doing. But
> that's
> just me.

You should look at other sources anyway. A few errors in books aren't
uncommon, and that's why we should never trust a single source for learning
how things work unless it has been extensively critically reviewed.

You might want to look at broadband matching applications for solid state
amplifiers. Many of them go from several ohms balanced to 50 ohms
unbalanced. Dig through Motorola application notes on transformers by Helge
Granberg and look at Walt Maxwell's books for theory. Walt is much more in
tune with how antenna systems work, and Granberg spent a lot of time working
with transformers. DeMaw might have some stuff on baluns also in the form of
matching solid state amps to 50 ohm filters if you can borrow a copy of
Solid State Design for the Radio Amateur.

You'll have few core worries because balanced impedance is so low, and since
you can use coax balun windings will be easy. I think your project will be
easy if you watch lead or connection lengths on the 12.5 ohm end.

73 Tom

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[Cookie]

Al, your comments about Jerry's book makes me wonder even more about what you
are trying to do with a backward 4:1 balun.  With a BS in Physics and 30 years
as an EE I found the book anything but a cook book.  To begin to understand what
I thought previously to be a simple device I had to drag up every bit of
transformer theory I ever knew and found the book to be at the PHD level with
basics assumed to be common knowledge.  I had to read the chapters several times
to feel that I understood what Jerry was telling me. 


In trying to guess what you are trying to match with a 50 to 12.5 ohm balun the
only good guess is some sort of shortened dipole.  This will not respond well to
a broadband approach because it will have a low impedence at resonance and high
impedance everywhere else.  The approaches that I can see working for this type
of antenna are LC networks rather than broad band transmission line
transformers.  You can probably make it work with a non-inductive resistance by
experimenting with different mixes of cores until you find the range of
interest.  If you are trying to match a shortened whip you will need a unun
which is similar, but wired differently than a balun.
 Willis 'Cookie' Cooke
K5EWJ




________________________________
From: Al Lorona <[hidden email]>
To: Elecraft_List <[hidden email]>
Sent: Sun, August 8, 2010 1:02:22 AM
Subject: Re: [Elecraft] Backward balun

Thanks for all of the excellent advice.

To sum up, today I measured a 4:1 current balun backwards and Tom was pretty
right on... the high frequency cutoff is so severely lowered that it is unusable

for my application. I did not test a 4:1 voltage balun.

I think one approach to take is to use a unun to transform down to 12 ohms and
then a conventional 1:1 balun following that to transform to a balanced feed.
I'll try this and report back if it doesn't work.

I know that there were several who suggested Sevick's book-- and I hope I'm not
commiting sacrilege by saying that I've gotten very little understanding of
balun theory from that book. It's a great cookbook for constructing baluns (and
ununs), but not so great to learn why one is doing what he's doing. But that's
just me.

It is also very surprising that I found almost no reference to a 50-to-12.5 ohm
balun on the web. You can certainly find one for purchase, but you can't find
the instructions on how to wind one. You have to buy the book, I guess!

This is surprising to me because I can think of many applications where one
needs to transform 50 ohms to a lower (balanced) impedance, including the one I
am currently battling.

Thanks again to everybody.

Good weekend,

Al  W6LX
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html



     
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html