Network tuner losses can be roughly divided into two groups: 1. basic
network losses, and 2. inadequate component losses.
1.Every network will show a loss factor (called "delta" by Terman) that is
a function of the load to be matched vs. the input impedance. Not all
loads--relative to the input impedance--permit low values of delta for
any given network configuration. For the standard CLC Tee network,
optimal matching conditions tend to coincide with maximum C values on the
output side, even though there may be other settings that will effect the
match. Even so, the delta may not be low in some cases, and the network
itself will show losses even with the very best component quality.
Some basic equations for estimating delta for various networks have been
developed by ZL1LE. His equations and some basic notes on network losses
are contained in a note at my site, and the equations are part of his
network ATU program which is a part of the HAMCALC collection of
utilities. These sources will help you estimate losses that are due
solely to the conditions to be matched and the choice of networks to make
the match.
2. Component losses represent inadequate values and designs of components
that result in additional losses. Of course, if a given network design of
a CLC tuner does not have adequate output capacitance range, an optimal
setting for lowest delta cannot be found. However, even with sufficient
range, components can be inadequately structured so that they dissipate
additional power--or so that they fail to lose the heat generated and
therefore begin to create destructive effects on the component or on other
components in the immediate area. Basic coil and capacitor design play a
major role in this set of losses, not to mention component sizing for the
anticipated power level and range of deltas. However, one should not
forget component cleanliness as a contributing factor to this class of
losses. Switch construction and connecting wire sizes also make a
difference. Tom Rauch's recent design that improves the structure of the
coil for matching network use is an example of design upgrades that can
make a difference in this department.
Once one has designed all components of a network tuner to be fully
adequate to the task at hand, there are still basic delta losses. For
some matches, delta may be fractional and losses with high Q components
can be less than 1%. (This is rare, and losses in the 2% to 5% range tend
to be a practical minimum.) However, some loads may force very high values
of delta, and losses will be much greater, even with the best components.
The best solution is then to do one of two things: a. change the load
to be matched so that a low value of delta is obtained, or b. change the
network configuration to one that yields a lower value of delta for the
matching situation. In many amateur installations, each option may be
equally difficult to implement, so that living with delta losses may be
necessary in the short run. In such cases, careful attention to component
quality is a must. However, option b. becomes a live option if
we decide that our present network tuner may have component shortcomings
that we wish to correct with a new unit of commercial or home
construction.
I hope this helps a little in sorting the types of losses that occur with
network tuners.
-73-
LB, W4RNL
L. B. Cebik, W4RNL /\ /\ * / / / Tel: (423) 938-6335
1434 High Mesa Drive / \/ \/\ ----/\---
Knoxville, Tennessee /\ \ \ \ / / || / http://www.cebik.com
37938-4443 USA / \ \ \ \ || e-mail: cebik@utk.edu
--
FAQ on WWW: http://www.contesting.com/towertalkfaq.html
Submissions: towertalk@contesting.com
Administrative requests: towertalk-REQUEST@contesting.com
Problems: owner-towertalk@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|