(more)lfo with 90deg lag?

[Haible Juergen]

	>just thought of a way, consider a saw LFO.
	>Now consider this as input to a diode breakpoint 
	>generator, set to make a triangle over the same period.
	>
	>And, let us have a second diode breakpoint generator 
	>to make the 3 segment waveform corresponding to one
	>period's worth of 90deg out of phase triangle.
	>
	>paul perry melbourne australia (going to bed)

Build an ordinary triangle / square generator.
(Integrator + schmitt trigger)
Then take the sign of the triangle (comparator ...)
and feed it into a second integrator
In theory, that would be enough. In practice,
you need a second feedback loop that will be inactive
as long as everything is like in theory, but which will
decently correct the switching points in case the second 
integrator starts to drift out of range.

(Not my own idea; got it from Tietze/Schenk, Halbleiterschaltungstechnik)

JH.