IVP <joecolquitt <at> clear.net.nz> writes:

>=20
> Hi James,
>=20
>=20
> I agree that interrupts and look-up tables could be used to simplify
> the coding. There is some groundwork needed to generate the data
> but after that it's fairly straightforward. Even without NCO (such as
> the F84) the pins can be alternated with BSF/BCF at the end of
> each interrupt half-cycle
>=20
> So, you'd have an interrupt for when to reload the timer, one table
> for the reloads, one for the sweep period, (the number of interrupts),
> possibly one for the initial load but that could be done via a goto
> and the DIP switch
>=20
> If you choose a short interrupt period (eg 100us) you'll have finer
> control at the Hz level but it does mean a lot of changes (big tables)
>=20
This approach works fine, main limitation is resolution (Hz) that can be
achieved.  I implemented square wave generator in PIC, although in C.  For
the audio frequencies that might not be an issue, depending on you
requirements. =20

Another approach which will likely provide a better resolution, is to use
DDS (direct digital synthesis), but remove the whole "generate clean
sine-wave" part.  A good example is http://www.g4jnt.com/PIC_DDS.pdf, there
was also discussion about DDS on PIClist.  The project linked above is in
assembler. =20

Sergey




--=20
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.