Ferrite transformer losses, 43-foot vertical and the K3

[Phil Salas]

Because of some earlier discussions here, I wanted to actually measure
losses in a 4:1 ferrite transformer.  I wanted this info as I have a
home-brew 43-foot vertical and these transformers are what seem to be
recommended for "matching" to this antenna.

I used a FT240-61 ferrite toroid which has a permeability of 125.  I chose
16-gauge speaker wire to experiment with.  This is because I want to
eventually use high voltage wire, and 16-gauge is the largest gauge
2-conductor high-voltage wire I could find (McMaster-Carr 9634T701 @
$2.65/foot).  I built a 4:1 unun, as I am feeding an unbalanced vertical
antenna.  And I decided to go with a voltage balun as this is a simpler
structure than a current balun or unun.

With a little experimentation, I was able to build a very good 1.8-30 MHz
4:1 unun.  This consists of 12-turns of the 16-gauge speaker wire on the
FT-240-61.  As the voltage balun is a little inductive causing degradation
at the higher frequencies, I tuned this out with a 33pf capacitor across the
50 ohm input.  This gave me a transformer with 1.2:1 SWR at 1.8 MHz, but
less than 1.1:1 from 3.5-30 MHz.  In order to measure loss, I built a second
identical transformer and connected these back-to-back.  I measured
insertion loss with both an Array Solutions PowerMaster, and a Tektronix
TDS-2200 digital oscilloscope.  I made all measurements with 20 watts of RF
power on my workbench.  Bottom line:  Loss through both transformers was
less than ½-watt (20 watts forward power) from 1.8-30 MHz.  This is just a
little over 1% of loss in each transformer.  Even if my measurements are off
by a factor of two, this is still pretty much insignificant loss.

Next I installed one of these transformers at the base of my 43-foot
vertical.  My radial system isn't the best in the world because of the space
I'm restricted to.  I have about a dozen random-length radials with lengths
up to about 50-feet.  My transmission line is 60-feet of Andrew ½-inch
Heliax that transitions to a 3-foot section of LMR-400 inside my house going
to the K3.  My Array Solutions PowerMaster is located immediately at the
output of the K3.  The SWR measured with the PowerMaster was as follows:

160:  4.9:1
80:  6.3:1
60:  3.3:1
40:  3.2:1
30:  3.2:1
20:  3:1
17:  2.1:1
15:  1.9:1
12:  1.4:1
10:  2.2:1

Obviously, these mismatches are easily handled by the internal K3
auto-tuner.  And line loss is minimal because the mismatch isn't very high,
and the transmission line is very low loss.

The 16-gauge speaker wire on the FT240-61 core seems to be working fine even
with 600 watts out of my ALS-600 amplifier.  However, I do have some of that
expensive McMaster-Carr high-voltage wire on order.

Anyway, I just thought I'd share these measurements with the group.

Phil - AD5X

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[N4LQ-2]

Phil. Questions:
1. Why would one use a balun when both the antenna and coax are unbalanced?
Wouldn't a unun be appropriate?
2. What are the swrs at the balun? The swr at the K3 doesn't tell us much
since the length of the coax affects it greatly.
Steve Ellington N4LQ
[hidden email]
----- Original Message -----
From: "Phil & Debbie Salas" <[hidden email]>
To: <[hidden email]>
Sent: Friday, September 19, 2008 4:21 PM
Subject: [Elecraft] Ferrite transformer losses, 43-foot vertical and the K3

> Because of some earlier discussions here, I wanted to actually measure
> losses in a 4:1 ferrite transformer.  I wanted this info as I have a
> home-brew 43-foot vertical and these transformers are what seem to be
> recommended for "matching" to this antenna.
>
> I used a FT240-61 ferrite toroid which has a permeability of 125.  I chose
> 16-gauge speaker wire to experiment with.  This is because I want to
> eventually use high voltage wire, and 16-gauge is the largest gauge
> 2-conductor high-voltage wire I could find (McMaster-Carr 9634T701 @
> $2.65/foot).  I built a 4:1 unun, as I am feeding an unbalanced vertical
> antenna.  And I decided to go with a voltage balun as this is a simpler
> structure than a current balun or unun.
>
> With a little experimentation, I was able to build a very good 1.8-30 MHz
> 4:1 unun.  This consists of 12-turns of the 16-gauge speaker wire on the
> FT-240-61.  As the voltage balun is a little inductive causing degradation
> at the higher frequencies, I tuned this out with a 33pf capacitor across
> the
> 50 ohm input.  This gave me a transformer with 1.2:1 SWR at 1.8 MHz, but
> less than 1.1:1 from 3.5-30 MHz.  In order to measure loss, I built a
> second
> identical transformer and connected these back-to-back.  I measured
> insertion loss with both an Array Solutions PowerMaster, and a Tektronix
> TDS-2200 digital oscilloscope.  I made all measurements with 20 watts of
> RF
> power on my workbench.  Bottom line:  Loss through both transformers was
> less than ½-watt (20 watts forward power) from 1.8-30 MHz.  This is just a
> little over 1% of loss in each transformer.  Even if my measurements are
> off
> by a factor of two, this is still pretty much insignificant loss.
>
> Next I installed one of these transformers at the base of my 43-foot
> vertical.  My radial system isn't the best in the world because of the
> space
> I'm restricted to.  I have about a dozen random-length radials with
> lengths
> up to about 50-feet.  My transmission line is 60-feet of Andrew ½-inch
> Heliax that transitions to a 3-foot section of LMR-400 inside my house
> going
> to the K3.  My Array Solutions PowerMaster is located immediately at the
> output of the K3.  The SWR measured with the PowerMaster was as follows:
>
> 160:  4.9:1
> 80:  6.3:1
> 60:  3.3:1
> 40:  3.2:1
> 30:  3.2:1
> 20:  3:1
> 17:  2.1:1
> 15:  1.9:1
> 12:  1.4:1
> 10:  2.2:1
>
> Obviously, these mismatches are easily handled by the internal K3
> auto-tuner.  And line loss is minimal because the mismatch isn't very
> high,
> and the transmission line is very low loss.
>
> The 16-gauge speaker wire on the FT240-61 core seems to be working fine
> even
> with 600 watts out of my ALS-600 amplifier.  However, I do have some of
> that
> expensive McMaster-Carr high-voltage wire on order.
>
> Anyway, I just thought I'd share these measurements with the group.
>
> Phil - AD5X
>
> _______________________________________________
> Elecraft mailing list
> Post to: [hidden email]
> You must be a subscriber to post to the list.
> Subscriber Info (Addr. Change, sub, unsub etc.):
> http://mailman.qth.net/mailman/listinfo/elecraft
>
> Help: http://mailman.qth.net/subscribers.htm
> Elecraft web page: http://www.elecraft.com

--------------------------------------------------------------------------------



No virus found in this incoming message.
Checked by AVG - http://www.avg.com
Version: 8.0.169 / Virus Database: 270.7.0/1679 - Release Date: 9/18/2008
5:03 PM

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[w5tvw]

When a high VSWR exists with a toroid ferrite balun due to a high inductive
or capacitive reactance and the toroid begins to heat, the losses will rise
to quite unacceptable losses and can possibly destroy the balun itself, even
tough the tuner used appears to have reduced the VSWR on the input side of
the circuit to a very low value.

Generally a 4:1 or 1:1 transformer type balun should be preferably used for
just a resistance transformation, not where there is a highly  reactive
component on the output side of the balun.  Lots of people "get away" with
this situation, but it isn't a very good idea.  I commonly did this for
years with very large ferrite cores and a vanilla high pass "T" network
tuner (like the many MFJ and other "T" network tuners)

For the last 15 years I have used nothing but the common Parallel type
balanced line link coupled tuner configuration and had extremely good
results with the old fashioned and sometimes very cranky to get setup right
circuit.

"choke" type baluns (the ones that traditionally use a large number of
ferrite beads on a length of coaxial cable) are much less troublesome than
the transformer type.  If your "balun" setup runs cool, then you probably
have hit upon a length of feeder that is "just right" and you are "OK".  If
it is running warm then you are treading on dangerous ground and things may
be getting ready to surprise you one day with a catastrophic failure,
especially when you run the legal limit!

This no matter what the VSWR meter says between the tuner and the ferrite
balun in question.

73,

Sandy W5TVW
----- Original Message -----
From: "n4lq" <[hidden email]>
To: "Phil & Debbie Salas" <[hidden email]>; <[hidden email]>
Sent: Friday, September 19, 2008 8:54 PM
Subject: Re: [Elecraft] Ferrite transformer losses, 43-foot vertical and the
K3

> Phil. Questions:
> 1. Why would one use a balun when both the antenna and coax are
> unbalanced?
> Wouldn't a unun be appropriate?
> 2. What are the swrs at the balun? The swr at the K3 doesn't tell us much
> since the length of the coax affects it greatly.
> Steve Ellington N4LQ
> [hidden email]
> ----- Original Message -----
> From: "Phil & Debbie Salas" <[hidden email]>
> To: <[hidden email]>
> Sent: Friday, September 19, 2008 4:21 PM
> Subject: [Elecraft] Ferrite transformer losses, 43-foot vertical and the
> K3
>
>
>> Because of some earlier discussions here, I wanted to actually measure
>> losses in a 4:1 ferrite transformer.  I wanted this info as I have a
>> home-brew 43-foot vertical and these transformers are what seem to be
>> recommended for "matching" to this antenna.
>>
>> I used a FT240-61 ferrite toroid which has a permeability of 125.  I
>> chose
>> 16-gauge speaker wire to experiment with.  This is because I want to
>> eventually use high voltage wire, and 16-gauge is the largest gauge
>> 2-conductor high-voltage wire I could find (McMaster-Carr 9634T701 @
>> $2.65/foot).  I built a 4:1 unun, as I am feeding an unbalanced vertical
>> antenna.  And I decided to go with a voltage balun as this is a simpler
>> structure than a current balun or unun.
>>
>> With a little experimentation, I was able to build a very good 1.8-30 MHz
>> 4:1 unun.  This consists of 12-turns of the 16-gauge speaker wire on the
>> FT-240-61.  As the voltage balun is a little inductive causing
>> degradation
>> at the higher frequencies, I tuned this out with a 33pf capacitor across
>> the
>> 50 ohm input.  This gave me a transformer with 1.2:1 SWR at 1.8 MHz, but
>> less than 1.1:1 from 3.5-30 MHz.  In order to measure loss, I built a
>> second
>> identical transformer and connected these back-to-back.  I measured
>> insertion loss with both an Array Solutions PowerMaster, and a Tektronix
>> TDS-2200 digital oscilloscope.  I made all measurements with 20 watts of
>> RF
>> power on my workbench.  Bottom line:  Loss through both transformers was
>> less than ½-watt (20 watts forward power) from 1.8-30 MHz.  This is just
>> a
>> little over 1% of loss in each transformer.  Even if my measurements are
>> off
>> by a factor of two, this is still pretty much insignificant loss.
>>
>> Next I installed one of these transformers at the base of my 43-foot
>> vertical.  My radial system isn't the best in the world because of the
>> space
>> I'm restricted to.  I have about a dozen random-length radials with
>> lengths
>> up to about 50-feet.  My transmission line is 60-feet of Andrew ½-inch
>> Heliax that transitions to a 3-foot section of LMR-400 inside my house
>> going
>> to the K3.  My Array Solutions PowerMaster is located immediately at the
>> output of the K3.  The SWR measured with the PowerMaster was as follows:
>>
>> 160:  4.9:1
>> 80:  6.3:1
>> 60:  3.3:1
>> 40:  3.2:1
>> 30:  3.2:1
>> 20:  3:1
>> 17:  2.1:1
>> 15:  1.9:1
>> 12:  1.4:1
>> 10:  2.2:1
>>
>> Obviously, these mismatches are easily handled by the internal K3
>> auto-tuner.  And line loss is minimal because the mismatch isn't very
>> high,
>> and the transmission line is very low loss.
>>
>> The 16-gauge speaker wire on the FT240-61 core seems to be working fine
>> even
>> with 600 watts out of my ALS-600 amplifier.  However, I do have some of
>> that
>> expensive McMaster-Carr high-voltage wire on order.
>>
>> Anyway, I just thought I'd share these measurements with the group.
>>
>> Phil - AD5X
>>
>> _______________________________________________
>> Elecraft mailing list
>> Post to: [hidden email]
>> You must be a subscriber to post to the list.
>> Subscriber Info (Addr. Change, sub, unsub etc.):
>> http://mailman.qth.net/mailman/listinfo/elecraft
>>
>> Help: http://mailman.qth.net/subscribers.htm
>> Elecraft web page: http://www.elecraft.com
>
>
> --------------------------------------------------------------------------------
>
>
>
> No virus found in this incoming message.
> Checked by AVG - http://www.avg.com
> Version: 8.0.169 / Virus Database: 270.7.0/1679 - Release Date: 9/18/2008
> 5:03 PM
>
> _______________________________________________
> Elecraft mailing list
> Post to: [hidden email]
> You must be a subscriber to post to the list.
> Subscriber Info (Addr. Change, sub, unsub etc.):
> http://mailman.qth.net/mailman/listinfo/elecraft
>
> Help: http://mailman.qth.net/subscribers.htm
> Elecraft web page: http://www.elecraft.com

--------------------------------------------------------------------------------



No virus found in this incoming message.
Checked by AVG - http://www.avg.com
Version: 8.0.169 / Virus Database: 270.7.0/1682 - Release Date: 9/20/2008
10:24 AM

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Don Wilhelm-4]

Sandy,

You are correct that using a balun (either 4:1 or 1:1) is not
necessarily the most efficient solution.  The 'old fashioned' link
coupled tuner will most always be more efficient.

When the feedline input impedance (and that has nothing to do with the
characteristic impedance of the feedline) is close to the output
impedance of the balun, the balun will be just almost as efficient as
the link coupled balanced tuner, but that rarely happens in practice.

Yes, using a balun following an unbalanced transmatch is a compromise.  
It lends itself to easy bandswitching and its associated convenience.  
If one is searching for the most efficient antenna tuning mechanism,
then either dedicated resonant antennas are required, or one must accept
the inconveniences of changing coils in a simple balanced link coupled
tuner or accept the compromises of an easy bandswitching system.  The
old Johnson Matchbox was an effort to provide bandswitching convenience
with a link coupled balanced tuner, but even it has limited matching
range compared to the simple single-band tuner designs.

Bottom line, one must either accept the compromises dictated by the
conveniences of bandswitching or accept the inconveniences of using the
most efficient tuners that can be constructed.  There is no 'best of all
worlds'.

73,
Don W3FPR

Sandy wrote:

> When a high VSWR exists with a toroid ferrite balun due to a high
> inductive or capacitive reactance and the toroid begins to heat, the
> losses will rise to quite unacceptable losses and can possibly destroy
> the balun itself, even tough the tuner used appears to have reduced
> the VSWR on the input side of the circuit to a very low value.
>
> Generally a 4:1 or 1:1 transformer type balun should be preferably
> used for just a resistance transformation, not where there is a
> highly  reactive component on the output side of the balun.  Lots of
> people "get away" with this situation, but it isn't a very good idea.  
> I commonly did this for years with very large ferrite cores and a
> vanilla high pass "T" network tuner (like the many MFJ and other "T"
> network tuners)
>
> For the last 15 years I have used nothing but the common Parallel type
> balanced line link coupled tuner configuration and had extremely good
> results with the old fashioned and sometimes very cranky to get setup
> right circuit.
>
> "choke" type baluns (the ones that traditionally use a large number of
> ferrite beads on a length of coaxial cable) are much less troublesome
> than the transformer type.  If your "balun" setup runs cool, then you
> probably have hit upon a length of feeder that is "just right" and you
> are "OK".  If it is running warm then you are treading on dangerous
> ground and things may be getting ready to surprise you one day with a
> catastrophic failure, especially when you run the legal limit!
>
> This no matter what the VSWR meter says between the tuner and the
> ferrite balun in question.
>
> 73,
>

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[w5tvw]

Absolutely, Don!

My main message was to watch the baluns for heating.  During a contest or a
long winded transmission, you can really screw up a nice ferrite balun
QUICKER than one thinks.

73,

Sandy W5TVW

----- Original Message -----
From: "Don Wilhelm" <[hidden email]>
To: <[hidden email]>
Sent: Saturday, September 20, 2008 8:59 PM
Subject: Re: [Elecraft] Ferrite transformer losses, 43-foot vertical and the
K3

> Sandy,
>
> You are correct that using a balun (either 4:1 or 1:1) is not
> necessarily the most efficient solution.  The 'old fashioned' link
> coupled tuner will most always be more efficient.
>
> When the feedline input impedance (and that has nothing to do with the
> characteristic impedance of the feedline) is close to the output
> impedance of the balun, the balun will be just almost as efficient as
> the link coupled balanced tuner, but that rarely happens in practice.
>
> Yes, using a balun following an unbalanced transmatch is a compromise.
> It lends itself to easy bandswitching and its associated convenience.
> If one is searching for the most efficient antenna tuning mechanism,
> then either dedicated resonant antennas are required, or one must accept
> the inconveniences of changing coils in a simple balanced link coupled
> tuner or accept the compromises of an easy bandswitching system.  The
> old Johnson Matchbox was an effort to provide bandswitching convenience
> with a link coupled balanced tuner, but even it has limited matching
> range compared to the simple single-band tuner designs.
>
> Bottom line, one must either accept the compromises dictated by the
> conveniences of bandswitching or accept the inconveniences of using the
> most efficient tuners that can be constructed.  There is no 'best of all
> worlds'.
>
> 73,
> Don W3FPR
>
> Sandy wrote:
>> When a high VSWR exists with a toroid ferrite balun due to a high
>> inductive or capacitive reactance and the toroid begins to heat, the
>> losses will rise to quite unacceptable losses and can possibly destroy
>> the balun itself, even tough the tuner used appears to have reduced
>> the VSWR on the input side of the circuit to a very low value.
>>
>> Generally a 4:1 or 1:1 transformer type balun should be preferably
>> used for just a resistance transformation, not where there is a
>> highly  reactive component on the output side of the balun.  Lots of
>> people "get away" with this situation, but it isn't a very good idea.
>> I commonly did this for years with very large ferrite cores and a
>> vanilla high pass "T" network tuner (like the many MFJ and other "T"
>> network tuners)
>>
>> For the last 15 years I have used nothing but the common Parallel type
>> balanced line link coupled tuner configuration and had extremely good
>> results with the old fashioned and sometimes very cranky to get setup
>> right circuit.
>>
>> "choke" type baluns (the ones that traditionally use a large number of
>> ferrite beads on a length of coaxial cable) are much less troublesome
>> than the transformer type.  If your "balun" setup runs cool, then you
>> probably have hit upon a length of feeder that is "just right" and you
>> are "OK".  If it is running warm then you are treading on dangerous
>> ground and things may be getting ready to surprise you one day with a
>> catastrophic failure, especially when you run the legal limit!
>>
>> This no matter what the VSWR meter says between the tuner and the
>> ferrite balun in question.
>>
>> 73,
>>
> _______________________________________________
> Elecraft mailing list
> Post to: [hidden email]
> You must be a subscriber to post to the list.
> Subscriber Info (Addr. Change, sub, unsub etc.):
> http://mailman.qth.net/mailman/listinfo/elecraft
>
> Help: http://mailman.qth.net/subscribers.htm
> Elecraft web page: http://www.elecraft.com

--------------------------------------------------------------------------------



No virus found in this incoming message.
Checked by AVG - http://www.avg.com
Version: 8.0.169 / Virus Database: 270.7.0/1682 - Release Date: 9/20/2008
10:24 AM

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Jim Brown-10]

On Sat, 20 Sep 2008 20:25:51 -0500, Sandy wrote:

>When a high VSWR exists with a toroid ferrite balun due to a high inductive
>or capacitive reactance and the toroid begins to heat, the losses will rise
>to quite unacceptable losses and can possibly destroy the balun itself, even
>tough the tuner used appears to have reduced the VSWR on the input side of
>the circuit to a very low value.

The word "balun" is far too broad and confusing in the context of this
statement. Dissipation in a common mode choke wound with coax, whether on a
toroid or only a string of beads (both are so-called "current baluns") is
UNRELATED TO SWR. It is DIRECTLY related to IMBALANCE in the antenna. Further,
the higher the choking impedance, the lower the dissipation.

There appear to be significant gaps in your understanding of baluns and common
mode chokes. My tutorial includes an extensive discussion of common mode chokes
wound with coax. http://audiosystemsgroup.com/RFI-Ham.pdf 

73,

Jim K9YC




_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com