[sdiy] ASM-1 VCO questions--please help

[Magnus Danielson]

From: <synth at charlielamm.com>
Subject: [sdiy] ASM-1 VCO questions--please help
Date: Sat, 15 Mar 2003 12:54:54 -0800 (PST)

> I am slowly getting this to work but still have a few issues.
> 
> 1) Pitch of VCO when summing opamp is set to 0V rel. to ground doesn't
> seem to be working correctly. I have read in the archives that the VCO
> should produce a frequency of around 900hz when there is no modulation
> (which I read to mean: Pin 1 of cntl op amp summer, or the "op amp" side
> of the 56K resistor R13, reads 0V on fluke)
> 
> My VCO produces more like 6K.

What is your R15?

If it is 150k you should expect about 9k, where as if you have 1.5M you should
expect about 900 Hz.

Since the ASM-1 has had both variants there you have it. The current
schematics online shows a 150k resistor and thus 9k is more appropriate.

        3            3            1         1
f  = -------- = ----------- = ---------- = ---- = 9090,909090... Hz
 0   R   * C    150k * 2,2n   50k * 2,2n   110u
      15    2

If you run with different values, just toss them in there instead.

The whole point about setting the op-amp side of R13 to 0V is that you remove
the effect of scale errors etc. and have a know setup. If you want you can
insert a trimmer in series with R15 to trim the oscillator down to C#9. This
should greatly simplified further trimming.

> 2) The thing won't scale.  It goes up in pitch when I play my kbd, but I
> can't get it to track evenly beyond 1 octave.  Maybe related to issue 1?

Whichever frequency you have with pin 1 at 0V, you need to trimm the scale to
match 1V/oct. This is what P2 is for. If you only use P1 and P2 then you have
to walk up and down the scale and trim offset and scale alternatively.
If you add a trimmer to R15 you trim the component offset separately, then
trim scale (P2) and then trim CV scale offset (P1) and then you should be all
done.

> 3) I have tried adding a 10K trim in series with the 150K resistor to give
> a high-note tweak.  No help.

You should hit C#9 (8869.844 Hz).

If you really have about 6k then you should look at correct values at R15,
C2, R19 and R20.

If you've got R19 and R20 wrong you end up with having both the wrong
frequency, wrong amplitude and with a DC offset (on Sawtooth output).
Both the DC offset and amplitude is EASY to check, but they should be checked
on the U3 output (5Vpp, 2.5V DC).

If you really have 6 kHz with R15 at 150k and C2 at 2,2 nF, then I would not
be supprised to find R19 = R20 and the Vref thus being 7.5 V.

Naturally, you could have a gazillion of other possible flaws...

> 4) This is my own board layout.  I am willing to entertain that I "didn't
> lay it out correctly" although the basic functionality (ramp + pulse out,
> PWM, PW, general behaviour of the VCO) seems to be OK.
> 
> The thing just won't scale.

Hmm...

> Any suggestions?  I am using 1% for all resistors in the exp converter and
> 5V ref for the comparator.  Everything else is 5%.  I am using a 22pf
> ceramic cap for C3 and a 2n2 Mylar C2.

Seems fair.

> I am using a 1K 1% metal film instead of the tempco.

Worked for me (temperature sensitivity expected flaw naturally).

Cheers,
Magnus