[sdiy] Trying to establish confidence in my test equipment

[Oren Leavitt]

A very easy way to tune the V/Oct tracking on resonant filters (SV, 
bandpass, lowpass, etc) using the scope X/Y inputs

Connect the sawtooth output of a calibrated VCO to the X axis of the 
scope and to the input of the filter to be adjusted. About 100Hz is good 
start frequency.
Connect a variable control voltage to the control voltage inputs of both 
the VCO and the filter.
Connect output of the filter to the Y axis of the scope.
Turn up the resonance and set the filter's initial freq so that you see 
a few cyles of a diminishing sine wave (but not self oscillation).

Turn the VCO/VCF control voltage freq up.
If the sine wave pattern compresses, the filter is going sharp.
If it expands, it is going flat.
The filter tracks the VCO perfectly when there is no change at all in 
the pattern on the scope as the control voltage is swept up and down. No 
compressing/expanding.

- Oren



harrybissell wrote:
> "Rykhaard D.A.M.I.A.N." wrote:
> 
> 
>>Hey Harry!  Lissou...... leesewjus ....... them funky patterns!
>>HARRY???  How do I do that on my scope?  :O  I've got 2 single trace
>>scopes.  Is it possible with one of them?  (A B&K and a Leader).  I
>>have 2 probes now, if that'd help. ?? :)
>>I'd like to learn that tuning trick! :O :) :)
> 
> 
> Lissajous patterns.
> 
> You need to be able to drive the scope X axis directly (external
> input, replacing the sweep)
> 
> The X axis will be the reference frequency. It must be stable, and its
> much nicer if it is a sinewave (easier to interpret) but a triangle
> is OK.  A square is really useless...
> 
> The Y axis is the normal input. Again a sine wave is the easiest, as
> above.
> 
> If the X and Y frequencies are perfectly in tune, the display will be a
> circle,
> oval, or straight line.  It will not move.
> 
> If they are close. the pattern will 'tumble' around between these
> shapes...
> at the beat frequency.  It is easy to pick out even a sub Hz movement
> this way.
> 
> If you double the input (Y) frequency... there will be two peaks in the
> X axis
> and the wave will look like a figure eight lying on its side... or a
> straight
> line. It will not move.  If the frequency is not perfect, the wave will
> tumble at
> the difference frequency, as before.
> 
> If you half the frequency, the figure eight will be vertical (not
> horizontal)
> and will not move, unless there is a difference then it will tumble (as
> above)
> 
> You can easily see low integer ratios, like 1:1 2:1 3:1 4:1... or 1:2
> 1:3 1:4...
> 
> or even fractional rations (a fifth will be 3:2).
> 
> If they are precisely in tune, the display will not tumble.
> 
> Now... it can be HARD to tell for some ratios (or odd waveforms) what
> the ratio
> is.  I hook up the signals to a small amp as well, so I can hear the
> relationships.  When you hear it getting near an octave, THEN look at
> the scope
> to fine tune it.
> 
> For setting up a VCO, set the reference oscillator to the midrange of
> the VCO
> calibeation setting, and adjust to see like two octves up, and two
> octaves down.
> 
> As long as the reference oscillator will hold the frequency, and you
> don't touch it... the VCO can be calibrated 'ratiometrically' and the
> absolute frequency
> is not important. It will be 1/4, or 1/2, or 1/1, or 2, or 4 times the
> reference
> frequency.
> 
> I usually use a good digital tuner these days... AND the scope to prove
> that
> the octave relationships are correct. I can't trust the tuner
> completely. :^P
> 
> H^) harry
>