> > Does anybody the difference in circuit design that would cause

this

> > effect Can I easily modify my newer A-155 to behave the same way
>
> Newer 155's have a small mod on them, which I believe is to stop it
> stepping more than one step when manually clocking it (and it's

also

> the thing that stops you clocking it very fast!). Underneath 'Board

4

> (Controller Board)' (one of the small ones), on your newer 155 you
> should see a capacitor and resistor soldered together and bridging

a

> couple of pins on an IC. I suspect this mod may be what stops newer
> ones behaving this way. I know Bakis (for one) has removed this mod
> on his 155's, so perhaps if he is listening he can quickly try
> clocking it with a sawtooth to see if it does clock 'randomly'
> (It'll save me getting my soldering iron out, but I will carry on
> looking at it to see if I can figure the exact mechanism out...)

I can indeed confirm that removing this mod causes the behaviour you
have observed with the sawtooth clock input, particularly noticeable
with a falling sawtooth. I simply de-soldered the junction between
the extra resistor and capacitor. The (comparatively) slow fall of
the sawtooth causes havoc with the logic - it cannot decide if it
should be 'high' or 'low', quickly flipping between both states until
the ramp has come down enough for it to be sure. This 'noise' on the
logic clocks the sequencer very fast, thus giving the appearance of
randomly jumping backward and forward through all the steps. The mod
prevents this from happening. (Without it, as previously mentioned,
it is now possible to clock the sequencer at frequencies well into
the audio range - nearly 6kHz from my A-145 at full bore!)

Tim