[sdiy] MFOS Soundlab Oscillators not oscillating

[James R. Coplin]

Thanks to everyone for all the help on this.  It appears that *both* MF104
transistors were bad in oscillator 1&2. Replacement parts arrived, made the
replacement and now everything works.  Thanks again!

James R. Coplin

-----Original Message-----
From: David G. Dixon [mailto:dixon at interchange.ubc.ca] 
Sent: Monday, June 01, 2009 4:46 PM
To: 'James R. Coplin'; 'synthdiy DIY'
Subject: RE: [sdiy] MFOS Soundlab Oscillators not oscillating

> I'm having serious problems with a Soundlab here I'm assembling.
> Everything
> works perfectly except the oscillators.  Both oscillators are not
> oscillating.  I've been pulling my hair out for about a month on this
> before
> posting to try to minimize the chances that it's me doing something dumb.
> I've checked the components dozens of times (really) and did a continuity
> check between all the points in the circuit and all the points check and
> supplies and grounds are all connected.  I can't find any bridges
> anywhere.

Well, the SoundLab doesn't do much if the oscillators don't oscillate!  ;->
But seriously, I've built one of these, and the oscillators worked fine, so
take heart.  You will prevail!

> I had originally built it with TL072 and TL074 ICs but replaced those with
> the LF444 chips as specified in the schematic as I heard some folks had
> trouble with the TLxxx series.  I had hoped this would solve the problem
> but
> still no luck.

The venerable TL074 works just fine here.  The LF444 has better input specs,
but you really would not be able to tell the difference.

I hope you used IC sockets!!!

> Here are the voltages for each of the pins for IC5 on the
> schematic, a LF444.  Does anyone have any thoughts on what might be
> causing
> this fail?
> 
> Pin / Voltage:
> 1 / -0.66v
> 2 / 0v
> 3 / 0v
> 4 / 9.25v
> 5 / 0v
> 6 / 0v
> 7 / .02v
> 8 / -7.62v
> 9 / 2.9v
> 10 / -.63v
> 11 / -9.13v
> 12 / 0
> 13 / 0
> 14  / .05v
> 

Let's have a look.  Pins 4 and 11 are the rails.  They look about right (you
must be using a wall wart rather than batteries).

Pins 5-6-7 are the CV summer and 1/50 gain block, so 1V in should give 0.02V
at pin 7.  This is trimmed to 0.018V by the 100-ohm trimmer for 1V/oct
tracking.  Pin 5 is grounded, and pin 6 will be at virtual ground, so 0V is
expected at both of these.

Pins 1-2-3 are providing a constant current source (9V dropped across 1M, or
9uA) for the expo-converting transistor Q1.  The emitter current of this
trannie is one diode drop, or about 560mV for a 2N3904 (I believe, if it's
the same as 2N3906 in the other direction) at base ground potential.  Hence,
at 18uA (9uA from both Q1 and Q3), one would predict a voltage drop across
the 10k resistor R64 of about 180mV which, on top of the diode drop, would
give a voltage at pin 1 of -740mV.  You are reading -660mV, which seems
about 80mV too low.  This would lead me to believe that perhaps one of your
transistors (probably Q3) is not sourcing any current...?  Again, pin 3 is
grounded and pin 2 is at virtual ground, so 0V are expected here as well.

Pins 12-13-14 are the ramp integrator.  Again, pin 12 is grounded and pin 13
is at virtual ground, so 0V are expected.  The voltage of pin 14 should be
increasing linearly and then suddenly falling.  This is your ramp wave.  The
fact that this is not happening means that this integrator is not receiving
any current, which again points to Q3.  My guess is that Q3 has been
installed backwards.

Pins 8-9-10 are the ramp reset comparator.  Pin 9 is receiving a reference
voltage of 9V*47k/147k = 2.9V, so this is correct.  The input at pin 10 is
-0.63V, which is below the reference voltage, so pin 8 should be putting out
a voltage which is two diode drops above the negative rail, or about -7.6V,
which it is.  Hence, there is nothing wrong with the opamp at all.

Check the JFET to make sure it is installed in the correct orientation, but
it won't switch until the comparator flips positive.  If the integrator were
receiving current from Q3, but the JFET was not doing its job, then the
voltage at pin 14 would increase until it was two diode drops below the
positive rail (7.6V) and sit there.  The fact that it is sitting at -0.63V
means that it is not receiving any current, and there is probably nothing
wrong with the JFET.

Hence, all fingers point to transistor Q3, which is probably in backwards
(most likely), or was fried by applying the soldering iron too long during
installation (a bit less likely, as my experience is that these little
bastards are actually pretty robust, unless you're a total klutz with the
soldering iron, which you probably are not).

> Any help would be greatly appreciated!  Thanks!

Well, I hope my little analysis helps you solve the problem.

> James R. Coplin
> 
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy