[Elecraft] balanced tuner

[Tom W8JI]

>I heard some one say that a balanced tuner is very 
>expensive to build. I used some quarter inch copper tubing 
>from the hardware store to build a four inch coil. This is 
>the tubing for a ice maker. I then built a four inch coil 
>and tapped it to a multi position switch from radio shack. 
>I had a  used large old condenser and hooked the whole 
>thing up in a L network design. I left the ground 
>connection floating on both input and output and connected 
>it directly to ladder line. A current balun was made out of 
>25 ft of coax and put at the input of the tuner. Works 
>great on my 330 ft loop. No second coil is necessary. 
>Everything is balanced and nothing heats up. Kind of ugly 
>looking, but the electrons don't seem to mind.>>

Bruce,

I looked at this issue extensively when the whole myth about 
moving baluns came out.

Moving the balun to the input of an unbalanced network does 
NOT make life on the balun or the system easier for the 
truly difficult problem, common mode isolation. It does not 
change a thing to the better for common mode currents or 
isolation, and it actually makes the system worse on higher 
bands where network physical size and unwanted stray 
capacitance affects balance.

It takes exactly the same common mode impedance and common 
mode current and voltage capacity in the balun if it is 
located at the tuner output or at the tuner input when the 
network is a floating unbalanced  network. The core (if 
used) will get just as hot, and current unbalance (except 
for stray capacitance or network transmission line effects) 
will be exactly the same.

If you draw it on paper and trace the path from one lead of 
the transmission line you will see exactly what  mean. There 
is a direct connection from one side of the balanced antenna 
terminals to the balun, and this means the balun  has 
EXACTLY the same common mode problems. The only thing you 
modify is the differential impedance, and it is extremely 
easy to solve that issue with any current balun so you 
really just fix something that is largely a non-issue to 
start with.

Now if you used a real balanced network with series 
impedances in each leg and in particular some perfectly 
ground reference point for the shunt elements, you would 
make common mode life easier for the balun BUT the drawback 
is you now have a balanced voltage source which may or may 
not supply balanced currents. The symmetry of the network is 
also critical. You have, in essence, exactly the same 
expense and difficulty as simply building a balanced tuner 
of any standard configuration.

A manufacturer would be misleading customers if it claimed 
they had a balanced tuner when using an non-symmetrical 
floating network with a balun on the input. It would be no 
better than the same balun on the output, and likely much 
worse on upper bands.

I'm afraid there is no free lunch. It has to be a balanced 
network which means at least double the cost of the 
expensive components, or you can simply build a good balun 
and use an unbalanced network on the balun input. Building a 
very good balun for the output is less than half the overall 
cost of using an expensive true-balanced network, I know 
because I priced this stuff out dozens of times.

There might be a marketing or sales advantage to customers 
who feel good about a balun on the input, but that would be 
dishonest or incompetent engineering by the manufacturer to 
claim it did anything for system performance.

73 Tom