Gene Fuller wrote:
What has become quite clear from this lengthy thread and the experiments
reported is that what you seek is impossible.

Not true. The question has been answered and it surprised me. Only
a half inch or so of ordinary 50 ohm coax is required to ensure the
SWR on that half inch of coax is the same as the SWR reported by
the SWR meter.

I didn't ask the original question in a very understandable
manner because I don't know of anyone who understood my original
question.

And who would really care to know such a thing?

History:
Reg and I were arguing about a year ago whether two feet of RG-400
was long enough to ensure that the SWR reported by the SWR meter
was the same as the SWR on the RG-400. I said it was but had no
references that covered the subject so had no proof. There was
no answer to my question forthcoming from this newsgroup at the
time, so I asked the question over on sci.physics.electromag. I
received an answer on that newsgroup but, unfortunately, didn't
question the math. The answer was one foot of RG-213 but the answer
was off by a magnitude due to an accidental shift in a decimal point.
The correct answer was about an inch to reduce the undesired mode
by a factor of 1/e^5, i.e. 1/0.2=5 where 0.2 is the conductor
spacing, e.g. estimated for RG-213.

The error was caused by using 2" instead of 0.2" as the
conductor spacing.

I mistakenly ran with the one foot answer but, wrong as it was,
proved that my two feet of RG-400 was long enough. Now Dave has
set the record straight and caused the decimal point error to
be exposed by Owen and me at about the same time.

The correct answer is surprising to me. 1/2 inch of ordinary
50 ohm coax is enough to ensure a 50 ohm environment in that
1/2 inch piece of coax, i.e. to ensure Vfor/Ifor=Vref/Iref=50.

Nobody else may care, but I learned a lot and thank everyone
who contributed.
--
73, Cecil http://www.qsl.net/w5dxp