[sdiy] BBDs

[Seb Francis]

anthony wrote:

>> Is it really true you can run the MN3207 or MN3007 up to 1.5Mhz?  The 
>> datasheets rate them only up to 200 or 100 KHz.
>> Do you think this would this overheat the chips, or degrade the sound 
>> quality?
>
>
> Yeah it's easy and it works. I've done it without a buffer with a 
> MN3207. Something happens as it gets out of its range. I think the 
> datasheet mentions that the clock biasing becomes nonlinear after 
> 200kHz or something like that. There's also ever the hint of the 
> effect 'fading' as the frequency gets past the rated range. The 
> Reticon ones seem to fall flat right above their rated range. You can 
> feel it 'drop out' or something. I was using a crummy clock though. 
> The datasheet refers to this as 'dispersion' I think. No big deal with 
> those though since their range is about 1.5 MHz anyway.
>
Thanks for your reply Anthony, I'll give it a try!  I guess the buffer 
necessity is more to do with the clock driver chip used than the BBD itself.

>> I guess the driver chip (e.g. MN3102) would cope fine as long as it 
>> was buffered as you describe.
>
>
> I wouldn't even bother with the MN3102. 4046's and 4047's are a lot 
> cheaper. (Their numbers are close but they're different...) The MN3102 
> makes the Vgg supply a tiny bit easier but that's about it.
>
I'm going to use a MN3102 since I already bought a few .. they were 
pretty cheap actually and I figured the quality of the Vgg supply would 
be better than just running a resistor off the power rail.  Not sure how 
much difference this would make, but for a few extra pence I'm happier 
to take the easy option :)

>
>>> A lot of the earlier flangers and choruses used SAD1024's and 
>>> SAD512's and R5106's which have a higher rated max clock speed. 
>>> These seem to be the more prized of the vintage flangers.
>>>
>>
>> Or perhaps an MN3009 or MN3209 would be good? (still only rated up to 
>> 100KHz, but only 256-satge)
>
>
> The less stages the easier it is to overclock it. Less capacitaive 
> reactance.
>
Again I guess this comes down to the clock driver output impedance and 
power dissipation.  I was more thinking these shorter BBDs would be 
better because they wouldn't need to be clocked so high to get really 
short delay times.

Seb