AVR-Chat

For my scope testpoints, in my data paths, I always leave
6pin headers for a simple D-A, based on only 6 resistors:
1R,2R,4R,8R,16R,32R.  R's are 1% values, multiples  usually
of 2.048K or another standard value.  Standard 1% values are
not always spot-on the power-of-2 value, however, but I select them
easily with a DVM.  I tie the common sides of the resistors all together
and to the scope probe or audio amp.  It look pretty good, and sounds
reasonable, when the digital outputs are modern CMOS parts.
Sometimes the 6 resistors are permantly soldered onto the board,
standing up(to save space), for a permanent analog testpoint.
Ken



> At 09:08 AM 10/01/2006, you wrote:
>>you can never know
>>which way the value will be off, or by what specific quantity.
>>
>>   So, with 16 devices each specified to be within + or - 1%, the overall
>>tolerance of the network is plus or minus 16%.
>
> Which could be 0%...in theory of course :-). I have used a Burns R2R
> resnet
> for the past 12 years or so to generate sounds for pedestrian crossing
> signals, it is not a HIFI system but never noticed any difference in sound
> in the some 20,000 units made (shipped to a few countries too).
>
> By the way the cost is nowhere near the >$2 mentioned  before, it takes
> very
> little room, don't need a buffer as I use the audio amp for buffering.
>
> Regards
>
> John Samperi
>
> ********************************************************
> Ampertronics Pty. Ltd.
> 11 Brokenwood Place Baulkham Hills, NSW 2153 AUSTRALIA
> Tel. (02) 9674-6495       Fax (02) 9674-8745
> Email: john@ampertronics.com.au
> Website  http://www.ampertronics.com.au
> *Electronic Design * Custom Products * Contract Assembly
> ********************************************************
>
>
>
>
> Yahoo! Groups Links
>
>
>
>
>
>
>