[sdiy] Digi-Mod schematic

[Seb Francis]

Neil Johnson wrote:
> Hi Seb,
>
>   
>> Still interested to know why you think the BAT54S would have been a bad 
>> idea.  Is it only the extra capacitance or are there other parasitic 
>> effects that I've overlooked?
>>     
>
> 1 - You're adding 20pF in parallel to 220pF, so what effect does that 10% difference mean?  What is the temperature coefficient of that additional 20pF?
>
> 2 - Diodes are not the ideal devices we might wish them to be.  They conduct as soon as a potential difference is supplied, what we might refer to as the leakage current in the 'off' state.  What effect will that have on your circuit?  By moving the diodes (or removing them completely) you can eliminate this source of error/confusion/headaches.
>
>   

Yes, makes sense.  So does seem like a good idea to find a way to get 
rid of them.


> Thought: try not to use leading or trailing underscores to disambiguate net names - its very difficult to see them unless you zoom in on the diagram, hence the confusion earlier.
>   

Was trying to make the labels fit into the tiny sub-circuit box on page 
1.  But you're right, it's not clear and I should just make the box bigger!


> On the issue of power supplies, you seem to assume that the +5V coming in to your board, and providing the analogue supply, is going to be clean enough for your application.  Is this a safe assumption?  Given that you're putting so much effort into your analogue signal path, I think a weak spot is your power supply scheme.
>   

No, I'm specifically not using the +5V power input for analog power.  
The power coming into the board (labelled "+5V_", soon to be labelled 
"EXT+5V" :) is not used for any of the analog audio path.  The +5V for 
the CODEC VA comes from the +15V supply, via U5 (78L05).

I am indirectly using EXT+5V for the CV buffers and CV ADC, but via a 
regulator U4 and LC network FB3/C13.

Thanks again for your input.
Seb