[sdiy] BBD findings 1

[JH.]

Made a few experiments with my BBD clock.
( http://jhaible.heim.at/matrix_fx/jh_matrix_fx_bbd_clock_sch.pdf )

Built 2 BBD circuits, one with TDA1022 for short delay, and one with
M3005 for long delay.

I just want to share some observations, even though I'm sure it's not
all new. No listening tsts yet - just sine generator.

TDA 1022 takes 900 kHz clock (that's the highest my circuit goes) with
no problem. Remarkably clean output.

MN3005 _survives_ 900kHz clock, but gets a little warm. But don't ask
about sound quality. Well, yes, do ask. The higher the clock rate, and the
higher the input signal frequency (!), the smaller the clipping level.
This means, you can run 1Vp @1kHz thru the chip at "normal"  clock
range, but the sine will become a saw wave (or worse) at higher clock
rates. However, with 200mVp you can get a lot higher. Guess why
some flangers keep the signal level a lot smaller than data sheet values ...

Also, clock rate changes the optimal DC bias setting at the input.
Well, unfortunately it doesn't change the DC bias (a trimmer), which
means that depending on clock frequency range, your clipping changes
from symmetrical to unsymmetrical.

Conclusion: BBDs suck (;->).
No, I was just joking.
Conclusion: BBDs are such complex sound shaping devices, that it would
be increadibly hard to emulate their behaviour by other means. No wonder
that (AFAIK) no one has nailed it yet.

JH.