Temp. compensated VCO

[uzs159 at ibm.rhrz.uni-bonn.de]

Hi Henri, and list,

Henri Kovalainen wrote:
> 
>  The board pic is now available at
> 
>  http://lyseolukio2.kajaani.fi/~dh/compvco.ps

Great work! Cool! Looks very good. 

>  I found out that (at least in my case, it was around 0.5v) the saw
>  amplitude varied quite much. So you might have to twiddle the feedback
>  resistor (470k) on the buffer amp (which also sums the offset trimmer).

In my original schematics this was a 47k resistor, with this you should 
get about 5Vpp saw amplitude.

>  The squarewave is generated with an op-amp comparator, so purists may
>  find it to be too slacky, but it was good enough for me. I haven't had
>  any latching problems with the comparator, but (supposingly) the
>  compensation amp config has latched a couple of times when I have whacked
>  it with a hard negative pulse from a lfo etc. 

This can indeed happen, if you overdrive the tempco-amp, since there is 
the possibility that some of the opamps can go beyond the common mode 
range. But in "normal" operation (with CVs in the range of 0-15V)
it should not happen.

>Also a small negative spike
>  might be present at the bottom of the saw wave, which didn't bother me,
>  and if I ever design a waveshaper board for this one I'll include a
>  clipping amp that will shave these off. The resistors values it the cv
>  summing amp depend on which way you hook the pitch pots. I didn't connect
>  the arrays leftover tranny to the negative rail (this was suggested in
>  Rene's schematics) which hasn't caused any problems that I'm aware of.
I did it this way, and had no problems.

>  Hop, skip and jump with a wire if you want this to be done. I have used
>  IRF530s instead of BUZ20s too, and they all seem to work properly.

I did use BUZ20, BUZ71 and BS108 and all worked fine. If one doesn't
like MOSFETs as switches one can use BJTs as well.


>  Also, I would like to point out that only proper feedback is appreciated,
>  any criticism or flaming is ignored and therefore futile.
>  If you don't like the design, skip it.
Hehe, yeah ;->

Well I should perhaps point out that there is a minor bug in my 
schematic: (I did recently realize this when reading the archives of 95)
The pins 6 and 2 of the 555 should go to the other end of 
the Hi-freq trimmer TP 2 (otherwise this trimmer is useless).
Fortunately one can change it easy on the board. The trace that 
goes to the upmost terminal of T1 should go to the lowest of TP2.

Two other remarks: I do not see why you changed R25 to 47K. The 
555s reset points are 1/3 and 2/3 of supply voltage this gives about 
5Vpp. It worked for me.
And one should include a small resistor into the V+ supply of the 555
to improve the effect of the 680n buffer caps.
This can be done by putting in a 100 Ohms resistor instead of the wire 
bridge that goes away from C7, and replacing the wire bridge from IC5
to another point that has +15V.

Bye
 -Rene