[AMRadio] Speech processing technologies and methods?

Donald Chester k4kyv at charter.net
Thu Dec 19 13:00:25 EST 2013

The UNAMPLIFIED D-104 is an excellent choice when using vintage amateur
radio transmitters. But one note of caution: the D-104 or any other crystal
mic should work into a minimum load resistance/impedance of around 5
megohms. This is according to the Astatic company in documentation they
published for decades. Working the mic into less  than optimum load
resistance kills the lower frequency range of D-104, which is actually
pretty good, and makes it sound more like a tin-can telephone.

A crystal microphone is approximately equivalent to an ideal a.c. generator
with a 500 pf capacitor in series. Naturally, with 500 pf of series
capacitance, the lower frequencies will appear only  when the load
resistance is extremely high. Unfortunately, many of the 1950s and 60s
vintage amateur  transmitters used a much lower grid resistance in their
first audio stage. It is not uncommon for the stock grid resistor to be only
one or two  megohms or lower. I recall a popular transmitter but can't
remember which one, that uses a grid resistor on the order of 100K!

I would recommend changing the input resistor at  least to 4.7 megs, and
preferably up to 10 megohms, although higher values of grid resistance may
result in unstable grid bias and plate current, and some individual tubes
are more sensitive to this than others. The RCA receiving tube manual
recommends a grid resistor of no more than 0.5 megohms for most tubes, but
up to 4.7 megs seems to work OK with the majority of good tubes. Check the
voltage across the cathode resistor if one is used, and see if this  changes
with the higher value grid resistor. Another good test, and one that can be
used with circuits that don't use a cathode resistor, is to measure the
voltage right at the plate pin, using a good quality DVM or VTVM that does
not load down the plate resistor.

My own D-104 works into 20 megohms of resistance. This is accomplished by
wiring the mic for balanced output and feeding it into a push-pull pre-amp
using two tubes carefully selected for balance and absence of hum, with a 10
meg grid resistor for each tube. Instead of the two sections of a 12AX7, I
use a pair of 6F5 octal-base tubes, which are electrically identical to one
section of an 'AX7, but separate tubes allows for  selection for a matched
pair. I used to  run a dynamic microphone mixed in with the D-104, but this
system has worked so well that I haven't used the dual microphones for over
a couple of years.

Don k4kyv

This email is free from viruses and malware because avast! Antivirus protection is active.

More information about the AMRadio mailing list

This page last updated 19 Nov 2017.