[sdiy] BBD clock bleed

Tom Wiltshire tom at electricdruid.net
Tue Oct 23 19:10:47 CEST 2018


So, to summarise:

1) Clock balance trim. 10K to replace R38/R41.This can help a bit, but you won’t see a huge effect.
2) Proper pre-BBD and post-BBD filtering. Even if the clock noise is ultrasonic and can’t be heard, it can cause problems further down your audio path. This is the most important. Bonus points for adding tracking filters to avoid cutting the bandwidth when not required. 
3) Keep CP1 and CP2 PCB traces short and neat and away from other sensitive lines. I put them on one side of the board and the OUT1 and OUT2 traces on the other side, and that seemed to work pretty well.

BBDs are noisy in general, and quite a few original 1980’s circuits wrap a compander (NE570/SA571 or similar) around them to try and keep it under control.

==================
       Electric Druid
Synth & Stompbox DIY
==================

> On 23 Oct 2018, at 15:36, Quincas Moreira <quincas at gmail.com> wrote:
> 
> Great info folks, thanks very much! I think for this video I will heed Erica's request and keep it as a flanger or slapback. It sounds pretty awesome doing that !  I already have two Pico DSP and one Erica digital delay for the longer repetitions...
> 
> Cheers and a great day to all!
> 
> Q
> 
> 
> 
> On Tue, Oct 23, 2018 at 9:26 AM <rburnett at richieburnett.co.uk> wrote:
> I doubt that pursuing any sort of "out-phasing" technique like that to 
> reduce the clock feed-through would ultimately be very fruitful.
> 
> The amplitude of the clock feed-through likely depends on the 
> instantaneous clock frequency.  The phase shift of the feed-through may 
> also vary with frequency, and your integrator to convert the clock from 
> square to triangle would add 90 degrees worth of phase shift to start 
> with anyway...  It would therefore be hard to achieve any significant 
> degree of cancellation over a range of clock frequencies, in my opinion.
> 
> I'd be looking at a static LPF or some sort of tracking filter to remove 
> the clock feed-through, once I was happy that the original design was 
> working properly, or at least as good as it can do.
> 
> -Richie,
> 
> 
> 
> On 2018-10-23 14:36, Quincas Moreira wrote:
> > I’ve thought of using na integrator to derive a triangle from the
> > clock pulse, then an OTA to distort it into a sine, then invert it and
> > sum it to the BBD output thus cancelling most of the bleed but now
> > we’re talking a full on daughterboard, lol
> > 
> > Sent from my iPad
> > 
> >> On Oct 23, 2018, at 8:12 AM, Tom Bugs <admin at bugbrand.co.uk> wrote:
> >> 
> >> There appears to be minimal input and output filtering also.
> >> 
> >> 
> >>> On 23/10/2018 14:07, Vladimir Pantelic wrote:
> >>> no output balance trimmer between out1 and out2 on this one
> >>> 
> >>>> On 23.10.2018 14:46, Quincas Moreira wrote:
> >>>> Sorry Folks, please follow this link
> >>>> https://drive.google.com/open?id=1O5kKkaJZqom61W8WQKQCfm3gLF83goz0
> >>>> 
> >>>> Sent from my iPhone
> >>>> 
> >>>> On Oct 23, 2018, at 06:13, Tom Wiltshire <tom at electricdruid.net
> >>>> <mailto:tom at electricdruid.net>> wrote:
> >>>> 
> >>>>> The list strips attachments. Can you post a link?
> >>>>> 
> >>>>> 
> >>>>>> On 23 Oct 2018, at 08:30, Quincas Moreira <quincas at gmail.com
> >>>>>> <mailto:quincas at gmail.com>> wrote:
> >>>>>> 
> >>>>>> thanks folks. It’s a new
> >>>>>> module, not yet released V3205 based. Strange that you don’t see 
> >>>>>> the
> >>>>>> pdf attachment, I’ll try again in the morning
> >>>>>> 
> >>>>>> Sent from my iPhone
> >>>>>> 
> >>>>>> On Oct 22, 2018, at 23:59, Pete Hartman <pete.hartman at gmail.com
> >>>>>> <mailto:pete.hartman at gmail.com>> wrote:
> >>>>>> 
> >>>>>>> As others have said a lot depends on which BBD we're talking 
> >>>>>>> about.
> >>>>>>> 
> >>>>>>> Most of my BBD experience is with Blacet's Time Machine.  It uses
> >>>>>>> the MN3005 which has balanced outputs, which can be trimmed to
> >>>>>>> minimize the clock bleed, and it also uses clock dependent 
> >>>>>>> filtering
> >>>>>>> and compansion to help reduce the noise further.  I have also 
> >>>>>>> played
> >>>>>>> some with the Doepfer A-188-2 which uses a tapped BBD, and that 
> >>>>>>> does
> >>>>>>> *not* have balanced outputs, so the bleed is harder to tame.
> >>>>>>> 
> >>>>>>> I looked at the Erica site as well, and honestly, I can't tell 
> >>>>>>> which
> >>>>>>> one you're asking about either.  Let us know/attach the schem, 
> >>>>>>> and
> >>>>>>> I'm sure someone will have suggestions :)
> >>>>>>> 
> >>>>>>> Pete
> >>>>>>> 
> >>>>>>> 
> >>>>>>> 
> >>>>>>> On Mon, Oct 22, 2018 at 8:55 PM Quincas Moreira 
> >>>>>>> <quincas at gmail.com
> >>>>>>> <mailto:quincas at gmail.com>> wrote:
> >>>>>>> Hey friends! I'm working in a review of the new Erica Synths BBD
> >>>>>>> module, and I'm getting massive bleed of the clock frequency 
> >>>>>>> (though
> >>>>>>> it looks sinusoidal rather than square)
> >>>>>>> Any tips on minimizing this bleed? Anytime the clock steps into
> >>>>>>> audible range it's super loud
> >>>>>>> 
> >>>>>>> Schematics included...
> >>>>>>> 
> >>>>>>> Thanks!!
> >>>>>>> 
> >>>>>>> 
> >>>>>>> 
> >>>>>>> --
> >>>>>>> Quincas Moreira
> >>>>>>> Synth Diy Guy
> >>>>>>> _______________________________________________
> >>>>>>> Synth-diy mailing list
> >>>>>>> Synth-diy at synth-diy.org <mailto:Synth-diy at synth-diy.org>
> >>>>>>> http://synth-diy.org/mailman/listinfo/synth-diy
> >>>>>>> _______________________________________________
> >>>>>>> Synth-diy mailing list
> >>>>>>> Synth-diy at synth-diy.org <mailto:Synth-diy at synth-diy.org>
> >>>>>>> http://synth-diy.org/mailman/listinfo/synth-diy
> >>>>>> _______________________________________________
> >>>>>> Synth-diy mailing list
> >>>>>> Synth-diy at synth-diy.org <mailto:Synth-diy at synth-diy.org>
> >>>>>> http://synth-diy.org/mailman/listinfo/synth-diy
> >>>> 
> >>>> _______________________________________________
> >>>> Synth-diy mailing list
> >>>> Synth-diy at synth-diy.org
> >>>> http://synth-diy.org/mailman/listinfo/synth-diy
> >>>> 
> >>> _______________________________________________
> >>> Synth-diy mailing list
> >>> Synth-diy at synth-diy.org
> >>> http://synth-diy.org/mailman/listinfo/synth-diy
> >> 
> >> --
> >> BugBrand LTD
> >> UK company No. 07199808
> >> VAT No. GB 988 2629 57
> >> 1 Ninetree Hill
> >> BRISTOL BS1 3SB
> >> United Kingdom
> >> www.bugbrand.co.uk
> >> 
> >> 
> >> ---
> >> This email has been checked for viruses by AVG.
> >> https://www.avg.com
> >> 
> >> _______________________________________________
> >> Synth-diy mailing list
> >> Synth-diy at synth-diy.org
> >> http://synth-diy.org/mailman/listinfo/synth-diy
> > 
> > _______________________________________________
> > Synth-diy mailing list
> > Synth-diy at synth-diy.org
> > http://synth-diy.org/mailman/listinfo/synth-diy
> 
> 
> -- 
> Quincas Moreira
> Synth Diy Guy
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at synth-diy.org
> http://synth-diy.org/mailman/listinfo/synth-diy




More information about the Synth-diy mailing list