From Steve Lenham Sent Sat, Nov 3rd 2018, 17:30
On 02/11/2018 16:09, Travis Thatcher wrote: > I've been having some trouble with a Farfisa Syntorchestra and i suspect > its the AY-1-0212 top octave gen. but I want to be sure before I go > about trying to find a replacement. Currently once the synth warms up > the E, F# and sometimes G# notes crackle or do not play at all when in > Trombone setting on the poly synth side. I scoped the pins of the > AY-1-0212 and I can see the fluctuation on the square waves of the > offending voices. even with the dividers removed. The 0212 does get warm > and I'm not sure if this is normal - not to hot to touch but warm. I've > gone over the solder connections and actually replaced with wires for > the pin of the 0212 responsible for F# to the divider chip. The weird > thing is that with the other poly switches everything works just fine - > the divider seems to bump up an octave for those others, judging by the > waveforms. After a strenuous morning of DIY, I thought I would relax with some vintage schematics and somebody else's problem :-) Those schematics show that the Syntorchestra has two master oscillators, one used for the Trombone voice and one for all the other poly voices. I presume this is because more than one voice can be selected simultaneously, in which case the separate clocks prevent them all being phase-locked together. ICb2 is a dual one-shot oscillator IC, here used to generate the two master clock signals. It's a device I had never heard of (SN29730) and The Internet has barely heard of it either, but it isn't doing anything particularly clever - just generating two high frequency clock signals. The master clock for the Trombone comes from pin 8 and has its frequency divided by two by ICb3 (7473 J-K flipflop). The other master clock comes from pin 5. This explains the difference in note frequencies you observed, which is normal. ICb1 (7400 quad NAND gate) simply selects which of the two master clocks is passed to the TOG IC.It is controlled by the signal marked "A", which comes from the Trombone selector switch. Now, while I can't absolutely rule out the AY-1-0212 TOG, it seems far too much of a coincidence that the Trombone has its own master clock and only the Trombone voice is misbehaving. I can't think of a fault that would cause the TOG (which is only a complex frequency divider) to work perfectly at one frequency but not at half that frequency. Before you do anything else, check that there is a stable voltage level on ICb1 pin2. This is the control signal from the Trombone selector switch and should go to -5V and stay there when Trombone is selected. It could be as simple as tarnished switch contacts causing dropout on this signal! After that, for me the #1 suspect is ICb1 (the master clock generator) and its associated solder joints and connections to the other ICs. I'd suggest two tests that don't require changing rare ICs: 1. Temporarily disconnect ICb3 pin 1 from ICb2 pin 8 and link it to ICb2 pin 5 instead. This runs the Trombone voice from the same master clock as all the other voices, though still at half the frequency. If the Trombone now behaves, the problem is with the Trombone half of ICb2. 2. If there is still a problem, reverse the previous temporary modification. Then, temporarily disconnect ICb1 pin 12 from ICb3 pin 13 and link it to ICb1 pin 4 instead. This runs the trombone voice from the same master clock as all the other voices but this time without the divide-by-two. If the Trombone now behaves, the problem is with divider ICb3. If there is STILL a problem then it has to be either the selector ICb1 (which you have already changed), a bad solder joint somewhere or - least likely IMHO - the TOG! Either way, reverse the previous temporary mod before proceeding. Please report back with any new information! BTW, be aware that all the circuitry we are discussing runs on 0V and -5V rather than the more usual +5V and 0V. Cheers, Steve L. Benden Sound Technology