[sdiy] Spice model for a Vactrol ???

[jhaible at debitel.net]

> Anyone have a good idea for making a spice model
> of a Vactrol ???


Sure. I'm working on this.


> I'm thinking of calling the LED a current controlled source (of voltage
> or
> current) but I can't figure out how to make the output look like a
> variable
> resistance.


Here's how I do it in PSpice (other Spice Simulators will probably
have similar building blocks, but not the same ones):

ABM3/I Model (that's an Analogue Behavioural Model with 3 Voltage
inputs and one floating current source output.
Just feed back the two current source nodes to two of the voltage
inputs. The 3rd input is used to control the resistance.
In order to avoid division, I make the 3rd input a _conductance_
input rather than a resistance input.
You can just write the equation into the model:
I = ( V1 - V2 ) * V3 , where V3 is the G value.
That's all you need for a static linear model.
Now run the G values thru a lookup table (also in PSpice), and
you have the nonlinearity. Easy.

What I have _not_ done yet, is the dynamic behaviour. Not
too difficult,either: Model a VC Lag with different time constants
for rising and falling values, and control them in real time with the
momentary G value.

I don't know if I will do this dynamic stuff, though. The static
nonlinear model is quite useful as it is.


JH.




> 
> Right now I don't care about approximating the actual resistance curve
> vs
> input current... just want to make it work :^)
> 
> H^) harry
> 
> 
> 




-------------------------------------------------
debitel.net Webmail