Hi all,

I was experimenting over the weekend to design a 1:1 Balun and a 1:9
Unun. Unfortunately I have been not very happy with my results.

Finally I realized that my Vectronics SWR showed a frequency depending
SWR of up to 1:2 at higher frequncies, if I used simple 1,5 inch wire
clips with a 50 Ohm resistor connected to the PL plug.

My setup was: the Vectronics, the Balun or Unun and a termination
resistor, either 50 Ohms or 450 Ohms, very tiny carbon resistors. The
wires of the Balun and Unun have been just an inch or so on each side.

By connecting the resistors as short as possible to the PL plug of the
Vectronics, I did see a 1:1, so it's the setup that makes a mess.

How can I measure a Balun or Unun, what setup do I need?

Thanks for your replies in advance,
73s de Wolfgang
OE1MWW

You didn't mention the specific frequency, and that can make a big
difference. I would expect wire to give you roughly four nanohenries
per centimeter, and if you add 12cm*4nH/cm=48nH in series with 50
ohms, you get to a 2:1 SWR at about 120MHz. That doesn't account for
the shunt capacitance the wires also introduce, which would reduce the
SWR somewhat, or increase the frequency at which you see 2:1. In
general, you should connect the load either immediately at the balun
or unun output, or through transmission line with impedance equal to
the load (or at least the same as the impedance of the transmission
line you plan to connect).

I'm assuming your meter doesn't respond to external fields, and only
to forward and reflected on the line, but if it does respond to
external fields, that's also an issue.

Cheers,
Tom


Wolfgang K. Meister wrote in message . ..
Hi all,

I was experimenting over the weekend to design a 1:1 Balun and a 1:9
Unun. Unfortunately I have been not very happy with my results.

Finally I realized that my Vectronics SWR showed a frequency depending
SWR of up to 1:2 at higher frequncies, if I used simple 1,5 inch wire
clips with a 50 Ohm resistor connected to the PL plug.

My setup was: the Vectronics, the Balun or Unun and a termination
resistor, either 50 Ohms or 450 Ohms, very tiny carbon resistors. The
wires of the Balun and Unun have been just an inch or so on each side.

By connecting the resistors as short as possible to the PL plug of the
Vectronics, I did see a 1:1, so it's the setup that makes a mess.

How can I measure a Balun or Unun, what setup do I need?

Thanks for your replies in advance,
73s de Wolfgang
OE1MWW

On 2 Feb 2004 10:07:11 -0800, (Tom Bruhns) wrote:

You didn't mention the specific frequency, ...

Tom - I was testing between 3 (7) and 30 Mhz

73s de Wolfgang

Wolfgang,

Then it's a bit of a puzzle. With two 4 cm pieces of wire connecting
a 50 ohm resistor to your meter, I wouldn't expect 2:1 SWR at 30MHz.
Maybe I can find time to try it on a good network analyzer today and
see what it looks like. As I understand your first posting, you see
the error with no balun involved, just the 50 ohms connected to the
meter through short ~4cm wires, and the error goes away when you
shorten the leads to practically nothing. Could your meter be in
error? You can check for a 2:1 SWR at 30MHz if you put 150pF in
series with 50 ohms, with very short leads right at the meter (or
through a short piece of 50 ohm line). It's always good to be able to
check your meter with some known loads--they don't have to be
precision loads for usual amateur work. For example, 50 ohms + 150pF
= 2.00:1 at 30MHz, and if you have a 5% error in the resistor and 5%
in the capacitor, you could be as low as 1.91:1 or as high as 2.11:1,
which is too small an error to worry about for any ham application I
can think of.

Cheers,
Tom


Wolfgang K. Meister wrote in message . ..
On 2 Feb 2004 10:07:11 -0800, (Tom Bruhns) wrote:

You didn't mention the specific frequency, ...

Tom - I was testing between 3 (7) and 30 Mhz

73s de Wolfgang

(Tom Bruhns) wrote in message om...

Then it's a bit of a puzzle. With two 4 cm pieces of wire connecting
a 50 ohm resistor to your meter, I wouldn't expect 2:1 SWR at 30MHz.
Maybe I can find time to try it on a good network analyzer today and
see what it looks like.

Did that. With two 100 ohm 1% 0805 SMT resistors in parallel right at
an SMA jack, using a freshly calibrated network analyzer at 30MHz, I
got about 60dB return loss, or about 1.002:1 SWR. Putting the same
two resistors on a pair of 4cm long 28AWG wires formed into a circular
loop, I got 17.3dB return loss at 81 degrees, which is about 1.32:1
SWR and matches up really well with about 73nH of series inductance
plus a tiny bit of excess series resistance. That's just about the
inductance my calculator tells me it ought to be.

So...I'd look for some other reason for the 2:1 SWR reading.

Cheers,
Tom

"Wolfgang K. Meister" wrote in message
...

....

How can I measure a Balun or Unun, what setup do I need?

You need a termination resistor... you can make one
by soldering a resistor inside a PL-259, or you can
buy one (they are available for 50ohms as BNC format).
__
Steve
KI5YG/EA
..

On Wed, 04 Feb 2004 01:35:59 GMT, " Stephen Cowell"
wrote:

You need a termination resistor... you can make one
by soldering a resistor inside a PL-259, or you can
buy one (they are available for 50ohms as BNC format).

Steve - my setup was:

- Vectronics SWR Bridge
- the Balun (Unun) connected to the PL connector with very short wires
- other side of Balun (Unun) terminated with a 50 ohm or 450 ohm
resistor.

The wires on the Balun, on each side, are around 1,5 inch long.

Is this a solution: (?)

a.) connecting a 50 ohm resistor with wires, approx. length of the
total wire length of the Balun, to the Vectronics
total wire length: same as wires on each side of the balun ;-)
b.) note the SWR over the range
c.) connect the Balun with the termination resistor
d.) calculate the difference ?

73s
Wolfgang

"Wolfgang K. Meister" wrote in message
...
On Wed, 04 Feb 2004 01:35:59 GMT, " Stephen Cowell"
wrote:

You need a termination resistor... you can make one
by soldering a resistor inside a PL-259, or you can
buy one (they are available for 50ohms as BNC format).

Steve - my setup was:

- Vectronics SWR Bridge
- the Balun (Unun) connected to the PL connector with very short wires
- other side of Balun (Unun) terminated with a 50 ohm or 450 ohm
resistor.

The wires on the Balun, on each side, are around 1,5 inch long.

Is this a solution: (?)

a.) connecting a 50 ohm resistor with wires, approx. length of the
total wire length of the Balun, to the Vectronics
total wire length: same as wires on each side of the balun ;-)
b.) note the SWR over the range
c.) connect the Balun with the termination resistor
d.) calculate the difference ?

I don't think you'll get far with that... it could be
+ or - reactance and still have the same SWR.
What construction are you using for the transformers?
Are they toroidal? This is best, since they are
self-shielding... you have to watch what's in the
near-field of unshielded stuff.
__
Steve
KI5YG/EA
..

Wolfgang K. Meister wrote in message . ..
....
Is this a solution: (?)

a.) connecting a 50 ohm resistor with wires, approx. length of the
total wire length of the Balun, to the Vectronics
total wire length: same as wires on each side of the balun ;-)
b.) note the SWR over the range
c.) connect the Balun with the termination resistor
d.) calculate the difference ?

No, certainly not! It's a complex quantity, R+jX, that determines the
scalar quantity, SWR. If you add two complex quantities, each of
which has a 2:1 SWR, you could come out with a 4:1 SWR, or you could
come out with a 1:1 SWR, or anything in between. So the difference
between two SWR readings will only serve to confuse you.

Also note that it is not just "wires" you are connecting between the
meter and the load, but a transmission line. The spacing of the wires
can make a large difference in the result, especially as they become a
significant fraction of a wavelength long.

If you've tested the SWR meter with a few known loads and it's giving
you the right readings, then I'd say it will probably about as
accurately tell you the SWR of the impedance presented to it by the
balun and whatever load you put on the other side of the balun. If
the SWR meter does not give you the right results with known loads,
then I'd say you are firing shots in the dark.

Cheers,
Tom

(For the purists and nit-pickers, as used above, SWR for any load
impedance Zload is defined as (1+|r|)/|(1-|r|)|, where
r=(Zload-Zref)/(Zload+Zref), and Zref is simply the reference
impedance to which the meter is (assumed to be) calibrated.)