sync24 specifications

[Fraser, Colin J]

> -----Original Message-----
> From: Byron G. Jacquot [mailto:thescum at surfree.com]
> Sent: 21 June 2000 00:19
> To: synth-diy at node12b53.a2000.nl
> Subject: RE: sync24 specifications
> 
> (thinking from a programmatical standpoint, the DIN clock 
> line probably
> interrupts the processor, telling it that it's time for the 
> next step.

If you look at the 303 / 606 schematics, the interrupt comes from an
oscillator that fires every 1.8 ms (IIRC).
The din clock line has to be polled by the CPU.

The processor is running essentially two different tasks - playback of a
sequence, and scanning of the keys & leds.
My guess is one task is handled by the regular interrupt routine, the other
by the normal program loop.
Either the clock line is polled during the interrupt routine, or it is
continuously polled during the normal program loop.

If it only polls the line while it is executing the interrupt routine, then
a pulse of less than 1.8 ms would be too short to be picked up reliably - it
might fall between two interrupts.
The interrupt oscillator period is adjustable, so on a badly adjusted
303/606, the clock pulse may need to be longer.
If the clock pulse is polled during the normal program loop, and the
interrupt routine takes say 50% of it's 1.8ms cycle to complete, then the
clock pulse would need to be at least 0.9ms long to ensure that it is not
missed during the interrupt service.
I've not looked at the 808 schematics for a while, but I think it works it
in a similar way.

Someone with too much spare time might like to investigate.
I find 2ms sync pulses have worked with every device I've tried to sync
(808, 909, 303, 606, 202 ,707, 727, MPC percussion computer)

My 12c508 based midi-sync chip is at
http://www.octavo.demon.co.uk/colinf/m2d/m2d.htm
This only needs 2 8 pin ICs, 6 resistors, a cap, and two sockets. No PSU
needed.


Colin f