Elecraft › [K2]

Can I measure antenna impedance with K2?

Classic

List

Threaded

26 messages Options

LA7NO

Re: Can I measure antenna impedance with K2?

Hi Don,

Thanks for your explanation.
Yes, I can follow you in this reasoning, and totally agree.
I was clearly not reading carefully enough earlier posts, and was jumping
to conclusions.

Per-Tore / LA7NO

On 1 August 2014 03:32, Don Wilhelm <[hidden email]> wrote:

> P-T,
>
> No, a conjugate match will assure 100% power transfer.
>
> Look at it this way - what the conjugate match says is that if you cut a
> transmission line at any point, looking one way at that cut point, you will
> have some impedance - example is 70 + j20. Now look the other way and the
> impedance will be the conjugate match - 70 - j20. That is the condition
> that exists.
> It also happens to be the condition for maximum power transfer.
>
> Since that cut can be made at any and all points along the transmission
> line - think what would happen if the efficiency at each of those points
> (when connected together) were 50% - nothing would get to the antenna. So
> we know that 50% is *not* the efficiency of any and all junctions of any
> conjugate match.
>
> Mixing the conjugate match concept with the maximum power transfer theorem
> is getting us into confusion - there are 'holes' in that combination. Yes,
> they work together, but not seamlessly.
> If a generator has an internal impedance of 50 ohms, the maximum power
> transfer will be only if the load to that generator is also 50 ohms. That
> says nothing about a conjugate match.
>
> Now to further complicate things, the internal impedance of a generator
> has nothing to do with the efficiency of that generator - especially if we
> are discussing a PA output stage. Bringing the conjugate match concept
> into the internal design of a PA stage is in error - it just does not work
> that way. The conjugate match only applies to the output of that amplifier
> stage.
>
> In other words, if we send 10 watts into a feedline (or ATU), all 10 watts
> will go to the load (antenna) except for losses in the feedline.
>
> I am not about to embark on the design and efficiency of a PA stage on
> this reflector, so take the conjugate match and maximum power transfer
> theorem only to the terminals at the PA stage and all will make sense.
> They do not apply to the internals of that stage.
>
> 73,
> Don W3FPR
>
> On 7/31/2014 7:32 PM, Per-Tore Aasestrand wrote:
>
>> Hi Don,
>>
>> On 1 August 2014 01:16, Don Wilhelm <[hidden email]> wrote:
>>
>> If one uses the voltage divider example, yes the maximum efficiency is
>> 50%,
>>
>>> But the output of a PA stage is not a resistor, and the collector load
>>> "resistance" is set by the designer to produce the output power desired.
>>>
>>> I fully agree.
>>
>> But will not a conjugate match also imply a max efficiency of 50%
>>
>> P-T
>> ______________________________________________________________
>> Elecraft mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/elecraft
>> Help: http://mailman.qth.net/mmfaq.htm
>> Post: mailto:[hidden email]
>>
>> This list hosted by: http://www.qsl.net
>> Please help support this email list: http://www.qsl.net/donate.html
>> Message delivered to [hidden email]
>>
>>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]
>

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

73,

Per-Tore / LA7NO

Jim Brown-10

Re: Can I measure antenna impedance with K2?

In reply to this post by Don Wilhelm-4

Much confusion and misunderstanding about impedance matching at the
output of an RF power amplifier. The output impedance (source impedance)
is NOT necessarily the same as the rated Load impedance. Rather, the
rated Load Z is the Z that the output stage is happy driving based on
its dynamic characteristics (what we old farts, raised on hollow state
devices called the "Load Line"), and will often be LESS THAN the rated
Load Z. Yes, the matching network should be tuning out the reactance,
and it should be providing the resistive load Z that the rig wants to
see, but this Z will rarely be the source Z of the output stage, and
will often be much lower.

But the real question here is, why expect a transceiver, designed nearly
20 years ago, to be suitable for measuring the impedance at the
transmitter end of a piece of coax connected to an antenna when so many
EXCELLENT devices capable of that measurement are available at
remarkably low cost, and with great power and flexibility?

There are the AIM products, OK but expensive for what you get, and my
favorite, the German designed, UK built, VNWA, a 1.5 GHz Vector Network
Analyzer that cost me $750 delivered to my home in W6 three years ago
with calibration loads.

I export data from this unit in Touchstone format (a plain text format
for data exchange) to SimSmith (freeeware, excellent) and let it compute
the complex Z at the antenna end of the coax, having measured the length
of the coax using the TDR capability of the VNWA. I can also expert
data from the VNWA on this sort of measurement to AC6LA's excellent
Excel spreadsheets, and also the data on a sample length of any piece of
transmission line (coax or other) to compute fundamental properties of
the transmission line. I can also use SimSmith to design matching
networks using stubs and discrete components.

http://k9yc.com/PacificonSmithChart.pdf

So with all of this analytic capability with very good accuracy
available at very low cost, why would you want to use a K2 and far less
elegant methods to do much less?

73, Jim K9YC

On 7/31/2014 6:32 PM, Don Wilhelm wrote:
> what the conjugate match says is that if you cut a transmission line
> at any point, looking one way at that cut point, you will have some
> impedance - example is 70 + j20. Now look the other way and the
> impedance will be the conjugate match - 70 - j20. That is the
> condition that exists.
> It also happens to be the condition for maximum power transfer.

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

David Woolley (E.L)

Re: Can I measure antenna impedance with K2?

In reply to this post by LA7NO

The efficiency would be even less, in practice, if designed that way.

The power transfer theorem isn't useful in most real world cases. In
practice, the optimum load for a PA will be very different from its AC
resistance (the reactive part should be corrected for by the fixed part
of the output matching circuitry). For an AF PA the AC resistance will
generally be much lower than the optimum load line resistance, because
of the heavy feedback used. For RF PA's there will be lots of parasitic
impedances and relatively little, if any, feedback will be applied, so
the situation is more complex. Without feedback, I have a feeling the
resistive component will be high, as it would be at AF, without
feedback, but I've not looked into the details. (The K2 low power PA
does have some negative feedback, so it is possible that it acts as a
voltage source, i.e. has an AC resistance much lower than optimum load.)

Generally, however, the optimum load is determined much more by the
target output power and the available power supply voltage. To a first
approximation, for ideal class B push pull, you want Vcc**2/RL to be
twice(?) the PEP. You then transform 50 ohms to that optimum value,
which is probably reactive, with the fixed part of the matching network.

It is the 50 ohm design value for the load that the KATn matches to the
antenna, not the actual source impedance of the transmitter. If you
look towards the transmitter, you will not see the complex conjugate of
the feedline input impedance, however, if you assume 50 ohms and the ATU
L and C values, you will get the complex conjugate of the feedline input
impedance.

--
David Woolley
Owner K2 06123

On 01/08/14 00:32, Per-Tore Aasestrand wrote:

>
> On 1 August 2014 01:16, Don Wilhelm <[hidden email]> wrote:
>
> If one uses the voltage divider example, yes the maximum efficiency is 50%,
>> But the output of a PA stage is not a resistor, and the collector load
>> "resistance" is set by the designer to produce the output power desired.
>>
>
> I fully agree.
>
> But will not a conjugate match also imply a max efficiency of 50%

Wes (N7WS)

Re: Can I measure antenna impedance with K2?

In reply to this post by k6dgw

Actually, the situation *never* occurs. Any practical transmission line
impedance is complex (reactive), hence a purely resistive load is not a complex
conjugate.

BTW, Zo and Vp vary with frequency too, making the situation even more
"complex", pardon the pun.

Wes N7WS

On 7/31/2014 4:20 PM, Fred Jensen wrote:
> [snip]
>
> If the feedpoint impedance of your antenna is totally resistive, *and* is
> exactly equal to the characteristic impedance of your feedline, then length
> only matters when calculating loss. This situation hardly ever occurs for any
> real hams however. :-)

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Wes (N7WS)

Re: Can I measure antenna impedance with K2?

In reply to this post by Jim Brown-10

Excellent!

On 8/1/2014 12:56 AM, Jim Brown wrote:

> Much confusion and misunderstanding about impedance matching at the output of
> an RF power amplifier. The output impedance (source impedance) is NOT
> necessarily the same as the rated Load impedance. Rather, the rated Load Z is
> the Z that the output stage is happy driving based on its dynamic
> characteristics (what we old farts, raised on hollow state devices called the
> "Load Line"), and will often be LESS THAN the rated Load Z. Yes, the matching
> network should be tuning out the reactance, and it should be providing the
> resistive load Z that the rig wants to see, but this Z will rarely be the
> source Z of the output stage, and will often be much lower.
>
> But the real question here is, why expect a transceiver, designed nearly 20
> years ago, to be suitable for measuring the impedance at the transmitter end
> of a piece of coax connected to an antenna when so many EXCELLENT devices
> capable of that measurement are available at remarkably low cost, and with
> great power and flexibility?
>
> There are the AIM products, OK but expensive for what you get, and my
> favorite, the German designed, UK built, VNWA, a 1.5 GHz Vector Network
> Analyzer that cost me $750 delivered to my home in W6 three years ago with
> calibration loads.
>
> I export data from this unit in Touchstone format (a plain text format for
> data exchange) to SimSmith (freeeware, excellent) and let it compute the
> complex Z at the antenna end of the coax, having measured the length of the
> coax using the TDR capability of the VNWA. I can also expert data from the
> VNWA on this sort of measurement to AC6LA's excellent Excel spreadsheets, and
> also the data on a sample length of any piece of transmission line (coax or
> other) to compute fundamental properties of the transmission line. I can also
> use SimSmith to design matching networks using stubs and discrete components.
>
> http://k9yc.com/PacificonSmithChart.pdf
>
> So with all of this analytic capability with very good accuracy available at
> very low cost, why would you want to use a K2 and far less elegant methods to
> do much less?
>
> 73, Jim K9YC
>

dl2ki

Re: Can I measure antenna impedance with K2?

In reply to this post by Don Wilhelm-4

Hi Don,

is it correct that the Lxx.x value in the tuner menu is determined via the relay positions of the individual partial inductances? Or is the value measured?

If the value is determined via the relay positions, the accuracy probably depends on whether the coil data match those from the circuit diagram as closely as possible. In the KAT2 manual there is a note that the individual inductances do not necessarily have to correspond exactly to the values in the circuit diagram.

This could explain some discrepancies in the measurement results I performed.

73, Wolfgang
DL2KI