From mj20279@janus.swipnet.se Sat Sep 28 12:14:35 1996 28 Sep 96 15:14:16 +0500 28 Sep 96 15:13:41 +0500 Comments: Authenticated sender is 28 Sep 96 03:58:11 +0500 by gaia.swip.net (8.7.6swip/8.7.3) with SMTP id JAA02179 for ; Sat, 28 Sep 1996 09:50:18 +0200 (MET DST) Date: Sat, 28 Sep 1996 09:50:30 +0200 (MET DST) From: Ricard Wolf Cc: Analogue Heaven Subject: MG-1 trigger speed enhancement mod Hi! Don't know whether I've emailed/posted this in various ways before, but here's a more structured description of how to speed up your MG-1's trigger rate: Moog/Realistic MG-1 fast gate response -------------------------------------- The MG-1 is unusual in that it has a pseudo-polyphonic voice which although a bit of a gimmick, can be useful as a third oscillator if nothing else. However, the addition of a polyphonic voice has crated a slight problem in the gate generation circuit. Normally on synths of this type, each key has two make switches, one which connects to a resistive divider for CV generation, and the other which is connected in parallel with the other key contacts, used to generate the gate. On the MG-1 however, the Poly voice uses one set of key contacts to simply switch the signals on off from the pitch divider chain for the voice, leaving one other make function for the CV. So how is the gate generated? It turns out that the Poly signal is taken out and smoothed with an RC circuit. When the Poly signal is present, i.e. a key pressed, the gate signal is activated, whereas when all keys are released, no Poly signal is available and the gate goes to the inactive state. The RC circuit smoothes the Poly signal so that the gate doesn't repeat at the audio frequency. Problem is, that the RC circuit smoothing function is a bit too sluggish as set from the factory, meaning that if a key is pressed in rapid succession, the gate signal doesn't have time to return to the inactive state, and the second trigger is lost. The effect is subtle, and can at first be mistaken for sloppy playing, bad return springs in the keys, or badly mounted keyboard switches. The keyboard simply feels sluggish in a subtle way. The remedy is to change resistor R47 on the bottom PCB (the one under the keys). The original value is 100k, and I've reduced this to 30k in my machine, but the optimum value may very well vary a bit between units, depending on the tolerances of other components. The basic idea is, go as low as you can without the gate starting to repeat at the frequency of the Poly voice note played. This is more likely to happen with low notes, or groups of low notes (clusters). Best way is to set up an AD patch (sustain switch out) with fast attack and decay, and play a few notes and clusters at the bottom end of the keyboard. This way, you'll notice if the gate starts repeating. Happy triggering! /Ricard -- Ricard Wolf (ricard.wolf@swipnet.se) Analog synth, railroad and Citroen GS fanatic -- "Ich bin der Musikant mit Taschenrechner in der Hand" --