Dear LF group,
 
I have a question to an audio reference signal generator and, as always, you 
will have some good suggestions...
 
Since today i have also installed Spectrum Lab by DL4YHF. There, i found the 
possibility to make a correction of the transceiver frequency drift by applying 
a audio reference signal. This is very useful for me, since my K2 has a 
temperature frequency drift of arround 20Hz (e.g. when opening the window, snow 
outside).
 
If i understood it correctly, i have to apply this signal to e.g. the left 
audio channel (microphone input at the PC), if my RX signal is at the right 
channel(?). 
 
One way to generate that reference signal could be to use my DDS-VFO at e.g. 
f=600,0Hz. But than i cannot use it any more to generate a test signal in the 
137kHz range.
 
So i thought about building a reference signal generator, using DCF77 at 
77,50000000000000000000000000000000000000000000000000000000000000000 kHz ;-) 
There are several ways to realise that and also many ready to use circuits. But 
i want to build my own(!). I have an idea about the circuit but perhaps it is a 
good idea to ask YOU if this will give a stable reference signal:
 
Input stage should be a standard ferrite rod antenna, brought to resonance by a 
C, of course. Than a simple amplifier using a BF245C or so. Then, further 
amplification until the voltage waveform gets (almost) rectangular. Perhaps a 
schmitt trigger stage, such as CMOS4093 to get a proper rectangular signal. 
Then, 4 D-Flipflops CMOS4013 to divide the signal by 128. The resulting 
frequency should be 605,4687500000000000000000000000000000000 Hz. If this will 
be a solution, it would cost not even 10EUR...
 
My question, or worries, are about the following: What about the noise , that 
is added to the voltage of DCF77? Will the frequency vary a few pHz ( ;-) ) due 
to that noise? And what about the trigger level of the schmitt trigger: This 
cheep circuit may not have a constant trigger level and thus the periode will 
vary from pulse to pulse? What about QSB of the signal of DCF77? This will vary 
the voltage (between constant time steps, when du(DCF)/dt is changing within 
two steps) at the input of the schmitt trigger and thus the output frequency 
will be f+ x*du/dt? In other words: It is possible to use the frequency 
77,5000... kHz itself, or do i have to read out the message schema, that is 
provided, such as the cheep digital clocks do it?
 
There are several modules available, that provide a 1Hz output signal or so. 
But what i want is a frequency in the NF range, that can be applied to the 
soundcard or a direct conversion receiver. Since my interest is normally based 
on LF operation, not to normal frequeny generators, this should not become a to 
big project, so easy but homemade circuits will be choosen :-)
 
What do you think?
 
Stefan/DK7FC
 

<<winmail.dat>>