[sdiy] Help, I'm Desperate! (Charge Injection with DG408)

[David G Dixon]

Thank you everyone for the excellent suggestions!

I also decided lastnight that it couldn't be charge injection.  The glitches
are just too big and not really filterable.

I concur that it is probably an issue with the logic source.  Each
more-significant-bit comparator is feeding the threshold to the next
comparator, so there is probably a natural cascading effect in these bits.
I've also not used any positive feedback, and maybe this will speed up the
comparators a bit.  I also thought briefly at one point about using Gray
code, but didn't act on it.  It might not require a huge change to the
circuit... just a different approach to controlling the comparators.  Lots
to think about.  I am also veering towards the idea of using a DG407, as I
think having only one chip will help, as it did with the DG409.

I have a bit more experimenting to do with the circuit, but can't get to it
until later today -- doctor's appointment and department Christmas party
await.
 

> -----Original Message-----
> From: Synth-diy [mailto:synth-diy-bounces at synth-diy.org] On 
> Behalf Of mskala at ansuz.sooke.bc.ca
> Sent: Tuesday, December 04, 2018 6:47 AM
> To: Steve Lenham
> Cc: synth-diy at synth-diy.org
> Subject: Re: [sdiy] Help, I'm Desperate! (Charge Injection with DG408)
> 
> On Tue, 4 Dec 2018, Steve Lenham wrote:
> > The ideal - though not always possible - way to eliminate these 
> > intermediate states is to make the system synchronous. You feed the 
> > asynchronous address lines to the inputs of a D-type latch, 
> then clock 
> > the latch at a point where
> 
> Another thing often done to reduce glitches in this sort of 
> situation is to use a Gray code.  Instead of using the 
> digital words in the sequence 000, 001, 010, 011, 100, 101, 
> 110, 111, where all of them can change between successive 
> values, one might use them in the sequence 000, 001, 011, 
> 010, 110, 111, 101, 100.  That way, only one bit changes 
> between successive words.  This is done for things like 
> absolute position encoders where you have a slow mechanical 
> system driving digital logic and you can't depend on both 
> bits changing at the same time when two change.
> 
> I don't really understand how the digital logic is being used 
> in David's design and I fear that trying to retrofit Gray 
> coding onto what he's currently got might require so much 
> change to the way the comparators are being used, as to be 
> impractical.  But it *is* a useful way to reduce glitches 
> when it's applicable.
> 
> --
> Matthew Skala
> mskala at ansuz.sooke.bc.ca                 People before tribes.
> http://ansuz.sooke.bc.ca/
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at synth-diy.org
> http://synth-diy.org/mailman/listinfo/synth-diy