[Test-Equipment] Trying to use a function generator as a tunable filter
Jeff Furman
jfurman at ocs.net
Sun Dec 13 19:10:54 EST 2009
I didn't see a direct answer to this question, perhaps I missed it in the
responses--
The usual function generator (identified by its capability of sine,
square, and, importantly, triangle waveforms) has a dedicated design to
fundamentally generate triangle waveforms, with a square wave generated in
the feedback. A sine wave is formed from the triangle waveform by a
nonlinear circuit that is independent from the triangle wave oscillator.
Some examples are: B&K 3010, HP3310A, HP3300A, as examples.
the B&K 3010 and the HP3300 manuals including their theory of operation
and schematics are accessible at the BAMA reflector:
http://bama.edebris.com/manuals/
You will see these are dedicated oscillator circuits, rather than filters
with feedback arrangements to allow oscillation.
In contrast to the function generators, oscillators such as wien bridge
(HP200CD, etc) and bridged-tee and twin-tee (can't remember which ones
right now) circuits may be heroically rewired to make use of their filter
networks as tunable selective bandpass or band reject filters.
I saw a paper once in the Journal of the Acoustic Society of America that
showed a modification of the HP200AB to be a bandpass filter.
Note that the HP200AB is a straightforward single ended circuit, in
contrast to the HP200CD's fancier balanced circuit.
If you examine the circuit for the HP330B and HP334 harmonic distortion
analyzers at the above web site, it's easy to see the wien bridge network
in the HP330B (the HP334 uses a wien bridge as well, but includes
automatic tuning circuits.) That's the worked example for the band reject.
If you can find the Heathkit harmonic distortion analyzer, it uses a
similar circuit.
A bandpass implementation can be found in GR1232A (again on the BAMA
mirror site) which uses an interesting Hall network for tuning with a
single pot instead of ganged capacitors. It also uses twin-tee fixed
frequency filters. I don't know any oscillators using the Hall network,
though.
Jeff Furman AD6MX
More information about the Test-Equipment
mailing list