Mythbusters: V/I ratio is forced to Z0
 

I always see a few that delight in attacking MFJ. Find me another company
that produces products for a reasonable price for the Ham.

Do they get any props at all, hell no. Some idiot ****es and moans that the
MFJ $19.95 what ever is less accurate than their favorite $350.00 supplier.

Why would you not "trust" an MFJ to measure anything?

If you can't give a specific reason to avoid an MFJ product, then what is
gained by your supercilious remark.


"Dave" wrote in message
...

"Cecil Moore" wrote in message
om...
Owen Duffy wrote:
This has nothing to do with the stated myth: Measurements with a Bird
43 of the conditions on the Thruline section are invalid unless it has
some minimum length of 50 ohm line on both sides of itself.

Would you be willing to make the same statement about an MFJ wattmeter?

now your are just trying to muddy the waters... i wouldn't trust an mfj to
measure anything!

What is a 50-ohm environment. ???
 

Dave wrote:
... or you get one 1/e reduction for each
distance equal to the spacing which in rg-58 or rg-8x that i used is
something like 1mm or less.. so how far down do you want to be?

Is that 1/e reduction for power or voltage?

in 4.5mm
you are down to 1% which is well under the accuracy of these cheap meters,
and much less than the length of even the connectors on the meters.

Kevin Rhodes said, "... the lowest order undesired mode should reduce
intensity by a factor of 1/e in about 2.03"..." for Rg-213. 2.03" is
a lot greater than the spacing in RG-213.
--
73, Cecil http://www.qsl.net/w5dxp

Mythbusters: V/I ratio is forced to Z0
 

Owen Duffy wrote:
Is it not time you put in some time on the experimental side to
support / validate your conceptual contribution?

Maybe you can help. I have an MFJ-259B antenna analyzer and a
dual-trace 100 MHz o'scope. Let's say I have an unmarked 2 inch
piece of coax and I don't know the characteristic impedance.

Using the MFJ-259B and o'scope, how could I ascertain the Z0 of
that 2 inch piece of coax? This is similar to the problem of
determining the physical Z0 of a poorly designed SWR meter.
--
73, Cecil http://www.qsl.net/w5dxp

V/I ratio is forced to Z0:was Mythbusters
 

On Sat, 15 Oct 2005 20:37:11 GMT, Cecil Moore wrote:
http://www.mfjenterprises.com/man/pdf/MFJ-816.pdf
What do you think is the purpose of the 10pf variable cap
if not to vary the voltage in the voltage divider?

You don't know, do you? :-)
It adjusts the frequency correction at the high end of the meter's
frequency range. Too bad this version of the manual doesn't lay it
out for Xeroxing - many of their other manuals do. This is a
commonplace of the Bruene design.

The question is: Between the "transmitter" terminal and
the "antenna" terminal, what determines the physical
characteristic impedance of the sampling circuit?

You don't know that one either, do you?

The turns ratio and the 82 Ohm load resistor IS the Z of sampling
circuit; the capacitor provides the e phase reference and tracking
across frequency. It is very lightly loading as a series load by
design and as evidenced by Dave's measurements.

Odd that you couldn't have said this yourself - rather ordinary stuff.

What is a 50-ohm environment. ???
 

On Sat, 15 Oct 2005 20:49:42 GMT, Cecil Moore wrote:
in 4.5mm
you are down to 1% which is well under the accuracy of these cheap meters,
and much less than the length of even the connectors on the meters.
Kevin Rhodes said, "... the lowest order undesired mode should reduce
intensity by a factor of 1/e in about 2.03"..." for Rg-213. 2.03" is
a lot greater than the spacing in RG-213.
so much for relying on an obscure poster quoted indirectly by
paraphrase to a new context. Such is third hand information.

What is a 50-ohm environment. ???
 

On Sat, 15 Oct 2005 19:02:58 -0000, "Dave" wrote:


"Cecil Moore" wrote in message
om...
Dave wrote:
don't bother, its a figment of cecil's imagination and creative
vocabulary.

Actually, the question logically follows from the reflection model.
Under what boundary conditions does a piece of transmission line
force Vfor/Ifor = Vref/Iref = Z0 to a specified accuracy?

then look up article 3.19 in 'fields and waves in communications
electronics' which is where they derive the fringe effects for a step change
in spacing between two plates. this is where they refer you for calculating
the effects of evanescent modes from a more complex problem stating that the
results are identical with the static case in 3.19... in the derivation of
the complex case of a step change in a cable they show that a single lumped
capacitance added at the step is an adequate representation of the
discontinuity when calculating the evanescent modes below cutoff. in 3.19
if you disregard the fields along the length of the step it ends up in an
equation: Z=h/pi(exp(pi*W/V0)-1-pi*W/V0_j*pi) the important part is
obvioulsy the decay factor in the exponential which goes as exp(-x/h) where
h is the separation in the planes... or you get one 1/e reduction for each
distance equal to the spacing which in rg-58 or rg-8x that i used is
something like 1mm or less.. so how far down do you want to be? in 4.5mm
you are down to 1% which is well under the accuracy of these cheap meters,
and much less than the length of even the connectors on the meters.

Dave, I note the work above estimating the rate of decay of the effect
of the discontinuity, and we have seen another analysis of the decay
rate (mis)quoted.

Doesn't the impact of the discontinuity at a distance x depend on the
magnitude of the products of the discontinuity times the attenuation
at distance X.

These discussions do not seem to have attempted to estimate the
magnitude of the products of the discontinuity.

Think about all the places where we do not preserve nominal Zo, and
whether significant power is diverted to other than the dominant mode
and subsequently lost. Look inside a HF transceiver, it is full of
discontinuities between the PA collector and the coax socket. Design
practice for HF transceivers does not consider potential losses from
evanescent modes.

If there were significant dissapative losses from such
discontinuities, don't all the texts on stub tuner design need to be
rewritten.

Where is the experimental evidence that significant power is diverted
in practical circuits and transmission line discontinuities?

My experiment reported at the beginning of the thread could be varied
to include another Bird at the dummy load, and to compare the power
loss between Birds with the expected cable loss (in the dominant mode)
to indicate whether there was another dissapative loss mechanism at
work.

I won't waste the time on the experiment because I anticipate that
instrument errors and cable specification errors would swamp any
likely effect of the discontinuities. IMHO, a well designed experiment
would require more accurate measurement of the dominant mode loss of
the 75 ohm cable, and more precise and accurate power measurement
equipment.

To demonstrate that point, I have just performed a test at 10.1MHz
with the Bird 43 and the 5.27m of 9275 on the load side with o/c load.

The measured return loss was -0.36dB. (Always seems un-natural
specifying a loss as -dB). I estimate the return loss should have been
-0.25dB (using the calculator at http://www.vk1od.net/tl/tllce.php),
leaving an unaccounted for -0.11dB. The differences are smaller than
the tolerances / accuracy of the equipment.

Perhaps it does demonstrate that the loss caused by the discontinuity
is not significant in terms of the specified accuracy of the Bird 43.

Owen
--

What is a 50-ohm environment. ???
 

"Cecil Moore" wrote in message
m...
Dave wrote:
... or you get one 1/e reduction for each distance equal to the spacing
which in rg-58 or rg-8x that i used is something like 1mm or less.. so
how far down do you want to be?

Is that 1/e reduction for power or voltage?

the equation gives the potential and flux functions, so its the
voltage/current essentially. no body calculates power functions unless they
are doing something strange.


in 4.5mm you are down to 1% which is well under the accuracy of these
cheap meters, and much less than the length of even the connectors on the
meters.

Kevin Rhodes said, "... the lowest order undesired mode should reduce
intensity by a factor of 1/e in about 2.03"..." for Rg-213. 2.03" is
a lot greater than the spacing in RG-213.
--
who is kevin rhodes? what reference books is he cited in as calculating
that and using what method? and that seems like an awfully long distance
for a fringe effect from a change in diameter of the conductors... but that
is just a gut feeling based on my understanding of the field equations.

V/I ratio is forced to Z0:was Mythbusters
 

"Cecil Moore" wrote in message
m...
Dave wrote:
ok, just for you i swapped all the cables around. and you are wrong, it
is not indistinguishable from 75 ohms, it is easily picked out as a 50
ohm section.

What causes that?
--

obviously between the tx and ant terminals it looks like a 50 ohm
transmission line.

ok, you guys have fun now, the rain is over and i probably won't be around
to argue with cecil as much the next few days. i trust someone will keep an
eye on him and make sure he doesn't twist the problem around so it can only
be solved with s parameter analysis referencing that optics web site
formulas he likes so much.

What is a 50-ohm environment. ???
 

"Owen Duffy" wrote in message
...
On Sat, 15 Oct 2005 19:02:58 -0000, "Dave" wrote:


"Cecil Moore" wrote in message
. com...
Dave wrote:
don't bother, its a figment of cecil's imagination and creative
vocabulary.

Actually, the question logically follows from the reflection model.
Under what boundary conditions does a piece of transmission line
force Vfor/Ifor = Vref/Iref = Z0 to a specified accuracy?

then look up article 3.19 in 'fields and waves in communications
electronics' which is where they derive the fringe effects for a step
change
in spacing between two plates. this is where they refer you for
calculating
the effects of evanescent modes from a more complex problem stating that
the
results are identical with the static case in 3.19... in the derivation of
the complex case of a step change in a cable they show that a single
lumped
capacitance added at the step is an adequate representation of the
discontinuity when calculating the evanescent modes below cutoff. in 3.19
if you disregard the fields along the length of the step it ends up in an
equation: Z=h/pi(exp(pi*W/V0)-1-pi*W/V0_j*pi) the important part is
obvioulsy the decay factor in the exponential which goes as exp(-x/h)
where
h is the separation in the planes... or you get one 1/e reduction for each
distance equal to the spacing which in rg-58 or rg-8x that i used is
something like 1mm or less.. so how far down do you want to be? in 4.5mm
you are down to 1% which is well under the accuracy of these cheap meters,
and much less than the length of even the connectors on the meters.

Dave, I note the work above estimating the rate of decay of the effect
of the discontinuity, and we have seen another analysis of the decay
rate (mis)quoted.

Doesn't the impact of the discontinuity at a distance x depend on the
magnitude of the products of the discontinuity times the attenuation
at distance X.

These discussions do not seem to have attempted to estimate the
magnitude of the products of the discontinuity.

Think about all the places where we do not preserve nominal Zo, and
whether significant power is diverted to other than the dominant mode
and subsequently lost. Look inside a HF transceiver, it is full of
discontinuities between the PA collector and the coax socket. Design
practice for HF transceivers does not consider potential losses from
evanescent modes.

If there were significant dissapative losses from such
discontinuities, don't all the texts on stub tuner design need to be
rewritten.

Where is the experimental evidence that significant power is diverted
in practical circuits and transmission line discontinuities?

My experiment reported at the beginning of the thread could be varied
to include another Bird at the dummy load, and to compare the power
loss between Birds with the expected cable loss (in the dominant mode)
to indicate whether there was another dissapative loss mechanism at
work.

I won't waste the time on the experiment because I anticipate that
instrument errors and cable specification errors would swamp any
likely effect of the discontinuities. IMHO, a well designed experiment
would require more accurate measurement of the dominant mode loss of
the 75 ohm cable, and more precise and accurate power measurement
equipment.

To demonstrate that point, I have just performed a test at 10.1MHz
with the Bird 43 and the 5.27m of 9275 on the load side with o/c load.

The measured return loss was -0.36dB. (Always seems un-natural
specifying a loss as -dB). I estimate the return loss should have been
-0.25dB (using the calculator at http://www.vk1od.net/tl/tllce.php),
leaving an unaccounted for -0.11dB. The differences are smaller than
the tolerances / accuracy of the equipment.

Perhaps it does demonstrate that the loss caused by the discontinuity
is not significant in terms of the specified accuracy of the Bird 43.

Owen
--
yes, this is true also. the initial amplitude of the evanescent modes is
proportional to the size of the discontinuity. when going from 50 to 75
ohms that is a relatively small discontinuity to start with and the modes
decay exponentially from that starting point. there are probably bigger
losses due to interconnections between components and the connectors on the
case than are caused by the actual change in cable impedance.

What is a 50-ohm environment. ???
 

On Sat, 15 Oct 2005 20:49:42 GMT, Cecil Moore wrote:


Kevin Rhodes said, "... the lowest order undesired mode should reduce
intensity by a factor of 1/e in about 2.03"..." for Rg-213. 2.03" is
a lot greater than the spacing in RG-213.

Cecil,

Go back and read your own post:


You quoted Kevin Rhodes:

(1/45m)**2 = (1/.203")**2 + kz**2
Clearly, kz must be imaginary to make this work. thus an
evanescant, non-propagating wave:
kz**2 = (1/45m)**2 - (1/.203")**2
To the accuracy used to date, the first term on the right
is negligible, so the decay rate, alpha, can be estimated:
alpha**2 = - (kz)**2 = (1/2.03")**2

How does rearranging the terms here increase 0.203" to 2.03"?

In any event 0.203" is an overestimate of the spacing, 0.203" is
actually half the overall diameter of RG213.

I don't really understand this mathematics, do you understand the
thing you are citing?

Owen
--