Re: Software PLL - the math that i did looks too simple to be true - please confrim

On Tuesday, February 8, 2011 11:39:55 AM UTC-5, Tim Wescott wrote:

  ...

> Robert, there's a heck of a lot more different types of phase detectors=
=20
> than just multiplying the two waveforms together.  Each one has its own=
=20
> impact on how the loop behaves, some making it easier to design the=20
> loop, some harder.  So far you seem stuck on the "phase detector=20
> multiplies signals" concept, and can't seem to move beyond it.

I will amplify that. Again, the CD4046 provides elementary examples. The ch=
ip has two phase detectors and pins to connect either of them. One is an XO=
R with, IIRC, Schmidt-trigger inputs, and the other is a state machine that=
 compares the timing of rising edges. The XOR provides a quadrature lock an=
d the characteristics you know. The edge detector provides an in-phase lock=
 and guaranteed eventual lock-up provided the oscillator's range makes that=
 possible. You notes your needs and you makes your choice. I think you'll f=
ind the data sheet worth looking at.

Jerry