Measuring Losses in an ATU

[Edward A. Dauer]

I loved Ron’s (AC7AC’s) alternative method for measuring power losses in an ATU (reprinted below.)  But if I remember my high school physics correctly, wouldn’t one have to know the specific heat, and the mass, of every bit of matter in the enclosure that participates in the thermal equilibrium in order to calculate the energy loss from the temperature rise?  That made me wonder if Ron offered it tongue-in-cheek.

That suspicion in turn called to mind the classic story of the bored physics class, in which the students were given a barometer and a battery-powered clock and asked to describe how they would determine the height of a nearby building.  One answer was to drop the barometer from the roof, use the clock to measure how long it took to hit the ground, and then working backwards from the known rate of gravitational acceleration calculate the distance it traveled.  Another was to lower the barometer on a string from the roof just to the point where it missed grazing the ground, start it swinging and then measure the period of the pendulum which is, friction ignored, a constant function of its length.  Third (for the pilots in the crowd), measure the barometric pressure at the roof and at the ground and, using the standard barometric pressure reduction rate and assuming a standard temperature lapse rate, calculate the difference between the two altitudes.  Finally, one fellow who was not destined to become a physicist but rather something else entirely, scoffed at having to climb to the roof.  He said, we’ll just go one flight down to the basement – where we give the superintendent the barometer and the clock as a bribe for him to tell us how tall the building is.

I know it’s OT.  Sorry.  But I haven’t had an excuse to retell that one in years.

Ted, KN1CBR




On 11/3/16, 3:16 PM, "Elecraft on behalf of [hidden email]" <[hidden email] on behalf of [hidden email]> wrote:

   
    A more direct approach would be to isolate the ATU in a very well insulated
    enclosure, transmit, and measure the temperature rise until it stabilizes.
    You can then calculate how much energy is being dissipated by the ATU and
    becoming heat. That requires the ATU be carefully isolated from the rest of
    the KX3 and be in a very well insulated enclosure to avoid loss of heat to
    the environment, something not practical to do with a KX3. It is, however, a
    valid way to determine losses in external devices like an RF transformer
    (e.g. "balun").
   
   

______________________________________________________________
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]

[K9MA]

I haven't tried it, but one way to measure the loss of the ATU would be
to connect a second, identical ATU "backwards" to a 50 Ohm load. This
second ATU would be tuned to present the highest or lowest resistive
load possible.  One way this could be done would be by connecting a
resistor across the transceiver output, matching it, and reading out the
ATU component values.  The second ATU would be set to the same values,
which should result in a match from the first ATU without retuning.  The
measured loss would be twice that of a single ATU.

That said, it's very likely the maximum ATU loss would be at the highest
frequency of interest and highest or lowest impedance the ATU can
match.  Highly reactive loads might make it worse.  I suppose an
automated test could explore all the possible combinations.

This could all be simulated, of course, if good models of the components
are available.  I suspect most of the loss is in the inductors, but one
needs to know more than their DC resistance.

73,

Scott  K9MA

--
Scott  K9MA

[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]

[K9MA]

On 11/3/2016 22:24, Ron D'Eau Claire wrote:
> Since, at HF, there is very little dielectric losses most of the loss is
> resistive. That means the greatest loss is at whatever condition of
> frequency and impedance transformation produces the greatest circulating
> currents. You are quite right; RF resistance is different from dc
> resistance. It's generally much higher due to the "skin effect".

The core losses could also be significant.

> One of the widest range ATU circuits is the T-match, but it is capable of
> producing a low SWR to the transmitter at settings the result I huge
> circulating currents in the coil. More than a few Hams have discovered the
> coil in their MFJ T-match tuner collapsed in a heap as the poly insulating
> bars on the coil melted at a fraction of the rated power. It's easy to have
> one consume more than 50% of the RF power applied.

The L network is the lowest Q matching network (without a transformer),
so it has the lowest circulating currents.  T and Pi networks have
higher Q, but don't require a continuously variable inductor.

At extreme impedances, T and Pi network tuners can indeed be so lossy
that most of the power is dissipated in the tuner itself.  One of mine
once achieved a match to a shorted feedline, with predictable results.

73,

Scott  K9MA

> 73, Ron AC7AC
>
> -----Original Message-----
> From: Elecraft [mailto:[hidden email]] On Behalf Of K9MA
> Sent: Thursday, November 3, 2016 4:28 PM
> To: [hidden email]
> Subject: Re: [Elecraft] Measuring Losses in an ATU
>
> I haven't tried it, but one way to measure the loss of the ATU would be to
> connect a second, identical ATU "backwards" to a 50 Ohm load. This second
> ATU would be tuned to present the highest or lowest resistive load possible.
> One way this could be done would be by connecting a resistor across the
> transceiver output, matching it, and reading out the ATU component values.
> The second ATU would be set to the same values, which should result in a
> match from the first ATU without retuning.  The measured loss would be twice
> that of a single ATU.
>
> That said, it's very likely the maximum ATU loss would be at the highest
> frequency of interest and highest or lowest impedance the ATU can match.
> Highly reactive loads might make it worse.  I suppose an automated test
> could explore all the possible combinations.
>
> This could all be simulated, of course, if good models of the components are
> available.  I suspect most of the loss is in the inductors, but one needs to
> know more than their DC resistance.
>
> 73,
>
> Scott  K9MA
>
> --
> Scott  K9MA
>
> [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]
>

--
Scott  K9MA

[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]

[k6dgw]

I once [1959] measured the efficiency of a 10 KW FM broadcast
transmitter using a variant of Ron's suggestion.  Came within 0.9% of
the value eventually measured on a water-cooled dummy load -- which was
about the accuracy of my K&E Log-Log Duplex Vector rule.  Never
underestimate the ingenuity of us old guys. :-)

73,

Fred K6DGW
Sparks NV USA
Washoe County DM09dn


On 11/3/2016 4:12 PM, Dauer, Edward wrote:
> I loved Ron’s (AC7AC’s) alternative method for measuring power losses
> in an ATU (reprinted below.)  But if I remember my high school
> physics correctly, wouldn’t one have to know the specific heat, and
> the mass, of every bit of matter in the enclosure that participates
> in the thermal equilibrium in order to calculate the energy loss from
> the temperature rise?  That made me wonder if Ron offered it
> tongue-in-cheek.
______________________________________________________________
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]