[HBR] Dissipation factor of materials

[Walt Hutchens]

Dan K9WEK said:

> My guess is that the small dissipation  factor of PVC will not affect the Q
> enough to measure.

Yep. The amount of PVC (or polystyrene) in the RF field of the coil is
too small to matter. There's resistance in the wire, energy is
radiated, the metalwork near the coil isn't perfectly conducting, the
tuning capacitor and other capacitors have losses: These dominate in
normal situations.

The antenna coil's effective Q will be controlled by the radiation
resistance of the antenna; this is doubtless one reason that receiver
performance in extreme strong signal conditions can be improved by
using a short (NON-resonant) antenna.

As a general design principle the less coupling you can use to the
antenna, the better. Make that HBR work properly with the
W6TC-specified antenna coil rather than coupling more tightly to
overcome poor sensitivity!

Offhand I'd guess that with a well-made HBR and its coils the dominant
factor for the MIXER coil Q is the resistance of the coil winding. Got
any solid silver enamel wire in #20 size?

However this coil probably could have too HIGH a Q: At a Q of 200 on
80M you have a half-power bandwidth of 20kcs or less which will
greatly weaken most of the 500kcs of unwanted signals and the higher Q
you have the more trouble there will be with both tracking and
instability in the RF and/or mixer stage. It's not uncommon to find
this coil loaded with a resistor in commercial designs.

The oscillator coil Q is important for stability. While osc. stability
against actual tuning changes -- the tube elements move, the chassis
expands, wiring moves, vibrates, etc. -- can't be improved by high Q,
transit time and resistance effects in the tube structure ALSO change
the frequency of oscillation and high Q will minimize this problem.

Transit time and resistance effects (the latter might be leakage
between the heater and cathode) change the frequency by changing the
phase shift within the tube. Such phase shift changes force the
oscillator to change to the frequency at which the tuned circuit
provides the (new) necessary amount of phase shift to meet the
requirement of EXACTLY 360 degree loop phase shift. The higher the Q,
the less frequency change is needed to cause a given amount of phase
shift change.

Transit time changes with applied voltage; you can regulate the HV but
the filament temperature also (effectively) changes it, and that's
harder to regulate.

Although the whole superhet idea is to move the main part of the
selectivity to a fixed frequency at which selectivity is easy, a
selective front end is ALSO important in a high performance superhet.
Why? Because the usual triode and pentode mixer circuits are
non-linear on the signal path -- that's HOW they do the mixing between
the desired signal and the local oscillator. However they will also
mix undesired SIGNALS with each other and this mixing can create
products at the desired signal frequency.

Thought experiment: By 100% perfect international agreement a given 20
kcs of the 80M band is totally vacant one evening -- all the usual
signals there have moved to other parts of the band. Is the center of
that 20 kcs dead silence? NO -- it will have a high noise level, due
to signals at other frequencies mixing in your front end so they
produce new sigs in the quiet zone.

In fact, thunderstorms -- an average of 3600 worldwide at any given
time -- create extremely powerful broadband noise which not only
includes the frequency of a desired signal but mixes to cause even
more garbage on that frequency. It's not just a crowded band that
makes a good front end helpful.

Since the garbage products appear at the desired signal frequency, it
is impossible to get rid of them with a selective IF. The only help is
to minimize them by weakening the undesired signals BEFORE they reach
the mixer. That's what high-Q front end coils -- especially the mixer
coil -- do for you.

Multiple RF stages help by giving you another high Q coil to cut down
off channel signals. However there's substantial complication of the
design and in HBR-type receivers I don't think that would be
justified.

Note also that the more the gain in the RF stage the more mixing of
unwanted strong signals will occur in the mixer and subsequent stages.
RF stage tubes should be chosen for strong signal handling ability and
NOT run at higher-than-necessary gain. Either go with W6TC's
recommendation or use a 6EH7 and keep the RF gain control no higher
than necessary.

Walt
KJ4KV