On Sat, 10 Jul 2004 01:04:22 +1200, you wrote:

>> Would a PIC12F675 using internal osc with code to control the PWM and
>AD's, with a linear or log pot as input, or with the PIC using it's A/D =
pin
>as input.  The PIC code would be minimal, and certainly offset if the =
volume
>produced is high.
>Would that work?
>>
>
>No hardware PWM I think. And cost is rather high compared with =
alternatives.
>Using Digikey as a guide (there MAY be cheaper but DK is easy to get
>comparison prices from). 12F675 is about $US1.35/1000s
>
>The tiny11 (also no hardware PWM) is $0.25 or less in that quantity! The
>question as to suitability of low frequency PWM is the only issue.
>
>_____________________________
>
>> ATTiny13 would seem a good fit - internal RC osc, on-chip ADC,  =
internal
>RC at up to 9.6MHZ
>you could get about 0.1uS PWM resolution in software, so about 0.2%
>resolution at 20KHz
>>
>
>A nice processor, but again, price is too high. About $US0.75/1000s
>And still no hardware PWM AFAIK.
>My PWM spec of at least 6 bits of resolution at 20 Khz means a PWM bit
>changes at 1/(20,000 x 2^16) =3D 0.78 uS
>At 20 Mhz you probably get about 15 MIPS (even though Atmel say 20 MIPS =
at
>20 Mhz. If this sat in a tight loop doing nothing but PWm you would =
probably
>make it. 15MIPs * .78uS =3D ~ 12 instructions per PWM bit! Not really =
enough
>to run interupts. Reading an A2D in there would get tough. I've little =
doubt
>that Scott D could coax out the magic needed, but it still costs more =
than
>the analogue solution and does the PWM and nothing else

You can get one-cycle software PWM resolution on a PIC or AVR, using =
either a short jump table to
trim the PWM period, or timer interrupts.=20
e.g.=20

addwf pc
nop
nop
nop

will give you a 2 to 5 cycle trim for W values of 3 to 0

Timer int would normally be the best way as you'd have plenty of =
foreground time to play with. Any
trickiness dealing with the whole 0..100% range can be overcome, if =
necessary with more than one
piece of code.=20

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsubscribe-request@mitvma.mit.edu