[sdiy] Digital control

[mark verbos]

I didn't see the Mbase01 scematic, I just looked at the PCB and saw 
those chips.  Obviously, using a DtoA converter that is shared for 
everythign is the cheapest way to go.

Mark



Johannes Öberg wrote:

>Thanks for your help guys!
>
>Are the Mbase01 schematics available somewhere, or does anybody know
>how they implemented the DCR's? How did the Polysix or other early
>patchmemory synths do it?
>
>The digital resistor chips are to expensive for me unfortunately. If I
>would try to build a digitally controlled analog synth (of course,
>this is more like dreaming than considering) I would need alot of
>digitially controlled "pots". What's the _cheapest_ way of doing this
>with an acceptable reproducability and resolution (say at least 6
>bits) ? IIRC the old polysynths used to use a single D/A going into
>these crazy mega-sample'n'hold chips, going into VCR's. Are these
>SnH's still an alternative? Can't it be done in a simpler way, if one
>accepts poorer stability or something?
>
>The other alternative seems to be multiplying DAC's, but this can't be
>a cost-effective way of doing it, can it? Or did I misunderstand? I'm
>believing you mean that I should use one DAC per pot. It sure seems
>compellingly simple enough though.
>
>Shouldn't it be possible to use a totally el-cheapo method, something
>like a single transistor as the VCR and either a giant sample'n'hold
>thing or perhaps even the old one-bit-D/A from the microcontroller? Or
>am I totally naive? Any linearity problems could be compensated for in
>software, methinks. As most of the 'digitally controlled resistors'
>would replace pots for non-timecritical stuff (like Attack rates) slow
>settling time (or what it's called) due to slow samplerate for the D/A
>wouldn't be a problem.
>
>/Johannes
>
>  
>