[sdiy] Single vs Dual Op Amp Supplies

[Matthew Smith]

Thanks for your replies, David.  Quoting from both in no particular order:

> Don't use a uC!  That's a complete copout!  Do it with analog, like a real
> man!  ;->
...>
 > And I'm actually dead serious about you resisting the temptation to
 > rely on digital techniques for this.  You'll never learn analog design
 > that way.

I believe that you speak wisely there.  With a) >99% of my experience 
being in the digital field and b) uCs selling for as much as or less 
than a glass of beer, my inclination is to seek a digital fix for just 
about every problem. Whilst this may help polish my digital skills, it 
does NOT help the analogue cause which is what half this exercise is about.

 > Yes, you can get A triangle from A square.  However, you need to
 > convert a square of any frequency at, say, 0 to 5 V, to a triangle at
 > that same frequency at 0 to 5 V.  The problem is, the current you send
 > to the integrator will be constant if you simply drop a constant
 > voltage across a resistor, say.  This means that the triangle will
 > have the same slope regardless.  Hence, only one certain frequency
 > will have the same amplitude as the square.  Anything slower will
 > increase in amplitude until the integrator opamp hits the rail.
 > Anything faster will decrease in amplitude until it becomes inaudible.

Understood - this is thus the crux of the matter.

 > Clear as mud?  I may develop a circuit to do what I've just described.
 > If I do, I'll send you the schematic, or if you've got access to
 > MultiSim, I'll send you the simulation file.

Don't have MultiSim nor have I ever looked at circuit simulation 
software before.  I tend to do it the 'old way' - build it, then stick a 
mutimeter/'scope on it.  Be interested to see the schematic - shouldn't 
prejudice my attempt at a solution too much!

Now I think I should set up a project directory and copy and paste all 
the relevant bits into a file in there.  What I thought was a little 
incidental bit looks to be a full-blown area for investigation.  Order 
of work probably needs to be:

1) Get the MIDI interface/assigner working, even if I don't include the 
arpeggiator code yet.
2) Build basic oscillator and have it receiving note data from (1).
3) Output from (2) to be offset and into split supply analogue section.
4) Make sure that output from (3) still looks like output from (2).
5) Tackle square to triangle problem.

That should take me a goodly while as I've got a ton of other projects 
both related and unrelated to attend to so won't even think what point 
(6) will be yet!

Cheers

M

-- 
Matthew Smith
Smiffytech - Technology Consulting & Web Application Development
Business:      http://www.smiffytech.com/
Blog/personal: http://www.smiffysplace.com/
LinkedIn:      http://www.linkedin.com/in/smiffy
Skype:         msmiffy
Twitter:       @smiffytech