Tempco's

[Joe Farler]

owner-synth-diy at horus.sara.nl:
> 
> Hi
>      If what has been said on the list i scorrect re: the tempco in Gene's 
> VCO circuit needing to go *UP* in resistance with increasing temperature 
> then those of us in UK definitely *CAN'T* use Farnell PN 143-584.
>      I'm investigating the possibilities either of getting a correct 
> positive tempco for the 1K part or else replacing the 56K with a negative 
> tempco. If we use the 10K equivalent to 143-584, we'd need a normal 180R 
> resistor where the 1K was. Will a load of 10180R be too low for the circuit 
> that was designed with a load of 57000R in mind?
> Answers please..... I've got one of the British tempco catalogues, and I'm 
> awaiting another.
> We live in hope.....
> chrisc
> 

All I did was put the negative Ik tempco in the upper leg of the divider with a 100 R
in the lower and change the feedback resistor to 18 K  ( from 100 K ). There is still a 10 K
trimmer, of course, to set 1V/oct.  All you need to do here is get 18 mV out for every 1 volt in and this works fine.  The osc has been stable for a week now with only minute fine tune adjustments called for.  It sounds great as well :)

As regards the circuit being designed with a load of 57,000 R in mind, thats not
really the case.  The original Electronotes design has a 2 K tempco  feedback
resistor and a 100 R to 390 R divider.  The tempco was moved to the divider to avoid
problems with inductance in the feedback loop of the summer. ( with some brands
of tempco )

There is an excellent explanation in the archive, posted by Gene, and sourced from
an Electronotes article on Tempco theory.

Regards

Joe