Capacitor reset with FET/analogue switch

[Rick Jansen]

In message <9603161539.AA66917 at horus.sara.nl> you write:
> 
> In sawtooth VCO designs usually a capacitor is charged
> with a current, and reset with an analogue switch of 
> some kind (an old discussion ;-).
> 
> I still haven't figured out all the ins and outs of the
> reset process, particularly wether a FET or MOS-switch
> in the form of a HC4066 ic is suitable or not. Theoretically
> it is not.

I got the answer (I hope) from Mr. Hill, one of the authors
of The Art Of Electronics:

|>But, how do these devices cope with the
|>theoretical max current? 10 ohms and 5 volts
|>results in 0.5 A initially... 

|>> No problem at all.  The FET just looks like a resistor, and 
|>> the high current flows and makes the larger voltage drop and 
|>> the internal heating rate is also high.  But it lasts such a 
|>> short time that the thermal mass keeps the junction 
|>> temperature to very safe levels.  

|>> The FET's negative tempco property (at high currents) means 
|>> there isn't any current hogging or localised heating.  

|>> Also remember E = 0.5 C V^2 

|>> This is is a pretty small amount of energy dissipated in the 
|>> FET; Unless the capacitor is 40,000 uF and you get a half 
|>> watt-second and significant junction temp rise...

So I guess this is the answer to why FET's and HC4066-like
switches do not break down when using them for capacitor reset.

Hill also suggests 2n7000 or vn01 or vn2222 parts as discrete
components for this purpose. They are smaller than a HC4066 ic,
but of course not as easy to get hold of.

Rick Jansen
__
rick at sara.nl   http://www.sara.nl/Rick.Jansen
_____________________________________________
S&H's a module and s&h's looking good