[sdiy] Choosing a DAC

[Stewart Pye]

There is still the issue of accuracy though.
For a 12 bit DAC with an INL of  +/- 1LSB with a 5V span,  steps could
be off by as much as 5V / 4095 * 2 =  2.4mV.

The best version of the AD558 had 1/4 LSB INL. Over a 5V span the error
for that could be 5V / 256 * .25 * 2 = 9.7mV.
Someone please correct me if I stuffed up my calculations there.

I might just get a 12 bit DAC with +/- 1LSB INL and see how it goes....

Regards,
Stewart.

Tom Wiltshire wrote:
> I was just about to suggest exactly the same thing. Everyone is 
> banging on about needing a 12-bit DAC as a minimum. This isn't true if 
> you're doing the glide function using analogue afterwards, and the 
> Pro-One is the living proof!
>
> T.
>
>
> On 7 Jul 2009, at 05:39, Olav Martin Kvern wrote:
>
>> Hi Stewart,
>>
>> Do look at the way that Sequential Circuits set up the DAC in the 
>> Pro-One. It's an Analog Devices AD558, an 8-bit DAC with an 
>> adjustable range. They use it as a 6-bit DAC, and feed it note values 
>> from 0 to 63 for just over a 5-octave range at 1 volt per octave. 
>> Glide is handled by a separate circuit. It's really pretty brilliant 
>> (in my opinion).
>>
>> With all of the incredible devices that are available today, it's 
>> easy to get hung up on DAC accuracy and resolution, but we can still 
>> learn a few tricks from circuitry from 1980.
>>
>> Thanks,
>>
>> Ole
>> _______________________________________________
>> Synth-diy mailing list
>> Synth-diy at dropmix.xs4all.nl
>> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>>
>
>
>