I am developing a circuit which uses the cd74hc4046a chip at low frequency (~ 300 Hz). With R1 ~ 900k and C1 ~ 22 nF the chip seems to yield an f0 in the correct range. I am using phase comparator #1. The feedback loop is active but is a follower (unity gain). R ~ 20k and C~ 100 nF.
If I simply power the chip (no input to sig_in), I can see a stable square wave output from the VCO on a scope when the coupling is set to 50 Ohm & DC. If I then introduce a signal at sig_in from an external function generator, I have to switch the coupling to 1M Ohm and AC or the chip will not lock, presumably because the VCO is (then) loaded too much (?). After changing the coupling on the scope one does observe that the chip is able to lock robustly and has a wide lock range.
Now the problem: The circuit I am developing essentially uses the VCO output to drive a piezo element, and after a couple steps that signal is routed back to sig_in. Each section of the circuit has been isolated and tested, and on their own they work fine. When I wire up the entire sequence, which is:
VCO --> buffer amp --> piezo --> charge amp --> custom schmitt trigger --> sig_in
I find that I cannot observe any output from the VCO whatsoever. I.e. it's not that the circuit won't lock, it's that the VCO acts like it won't even turn on. But I am using phase comparator #1, so it should oscillate at f0.
To test this I tried to route the VCO output directly to the sig_in pin, both with a straight wire and capacitively coupled. The lock indicator (pin 1) pulls high, however, again I cannot see any output from the VCO on the scope.
My questions are:
1) Are there load requirements for the VCO output? Nothing germane seems to be written in any of the manuals I have for the chip.
2) What does one have to do to make the VCO cough up an output for the sequence present in the application of interest (listed above)?
2a) I tried adding a 50-Ohm load to the VCO output, in parallel with the input to the buffer amp...this did not work. One might imagine that a simple filter to introduce a phase shift might help, however I have not tried this yet.
Please advise.
