[sdiy] on vactrols

[JH.]

My 2 cents:

the use of 2 Vactrols (in potentiometer configuration) is mandatory, because
of
temperature drift, if you want the presets to be halfway predictable.
If you don't (allowing some little randomness), one Vactrol might be fine.

Vactrol have a huge benefit over LM13700 and the likes: They are free of
offset
voltage.

Unless you have to reproduce the behaviour of one specific type of vactrol,
and if
you have time and want to save money, you might simply build your own
vactrols.
Glue a LED and a LDR together, and seal it against environmental light with
black
paint. There's other options than using a LED even, although a LED might be
your
first choice in most applications. Four LDRs placed around a small
incandescent
lamp are the heart of the legendary Schulte Compact A Phasing. (Two groups
of 4 LDRs around a bulb, actually.) A LDR in combination with a small
electroluminiscent panel is the core of the famous LA2 studio compressor.
(You run the audio signal of the sidechain right thru the EL element without
rectifying it - just boost amplify it to some 50 volts with a little tube
amplifier.)
There have also been vactros with neon lamps, but I forgot what they have
been for.
To trigger thyristors, maybe.

Don't underestimate the slow response of Vactrols in "static" applications.
When I dial up the ring modulation depth on my PolyKorg clone, it's hard
to get precise in-between values because I have used rather slow Vactrols.
The fact that the response time depends on the momentary resistance of the
LDR,
and whether you're going to increase or decrease it, complicates things
further.
Because your potentiometer, built from two Vactrols, will never have a
constant
resistance between its two end positions, as an ordinary potentiometer
would.

Examples of vactrols used as potentiometers in a preset recall system can be
found
on my PolyKorg pages, like this one showing the before mentioned ring
modulator
"potentiometer"
http://www.oldcrows.net/~jhaible/polykorg/jh_3200_signal_3of4.gif
and the according LED control circuit
http://www.oldcrows.net/~jhaible/polykorg/jh_3200_signal_2of4.gif .

JH.



> In a message dated 8/20/05 11:47:36 AM, philmacnutt at mac.com writes:
>
> <<  Thanks for all the info from everyone.
>
> Quick question, does the performance (speed, etc) of the vactrol matter
much
> if i am just using them across potentiometers with a preset unit
(programmer)?
>  that is, the values are only changed once when setting up the
programmers,
> then they stay put.  I'm a little out of my knowledge-zone on this, so
forgive
> me if i am not making sense... >>
>
> It sounds like that wouldn't really be a very critical application, so the
> vactrol response speed would not be so important.  Just know that when you
> change to a new preset, there will be a brief settling time for the new
parameter
> values.  It may or may not even be noticeable, unless you're playing a
note on
> the exact same beat as your patch change.
>
> Keep in mind that there may be certain parameters that would be much
better
> served by going through a OTA/VCA, such as an LM13700.  For instance, any
sort
> of signal level attenuation that you would normally need to use a
3-terminal
> attenuating potentiometer to control (as opposed to a pot in a 2-terminal
> "rheostat" or "variable resistor" configuration), such as LFO modulation
amount
> controls, or audio signal level controls.  You would need two vactrols to
> simulate an attenuating pot (or one of those special dual-element
vactrols), and that
> would get expensive very fast -- not to mention, the response to CV would
not
> be nearly as predictable and linear as with an OTA, and the response of
two
> vactrols would not likely match very well.  LM13700's cost about $1.50
these
> days (sometimes even less), and each one contains two separate complete
VCA's.
>
> MB in Las Vegas (where Austin, TX seems like a cool, tropical paradise)
>