Shop OBEX P1 Docs P2 Docs Learn Events
clock out phase delay — Parallax Forums

clock out phase delay

steprogsteprog Posts: 227
edited 2009-12-26 06:18 in Propeller 1
Hi Guys,
I was just wondering.· Im rigging a prop to be a LO frequency generator.· I need to "delay" or rather adjust the phase of these clocks to optimize the demodulation Im trying to do.· Im outputing frequencies ranging from 50 to 70MHz. I would like to adjust the delay of these signals.· Is this doable?
Thanks,
Greg

Comments

  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2009-12-24 19:26
    Using two counters, you can send out square waves that are in quadrature with each other (i.e. I and Q). This is done by presetting the phase registers. You may, however, find that the output is unsuitable for RF work, due to the phase jitter.

    -Phil
  • steprogsteprog Posts: 227
    edited 2009-12-24 19:54
    Hi Phil,
    Not doing quadular on this just adjusting the delay to pic everything up on one leg. Like a lock-in amplifier. I don't know much about phase jitter, doesn't that depend on the clock used?
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2009-12-24 21:20
    steprog said...
    I don't know much about phase jitter, doesn't that depend on the clock used?
    If the clock originates from one of the Prop's counters (e.g. NCO driving a PLL), the jitter will be a function of FRQx: the more "one" bits there are, the more jitter there will be.

    -Phil
  • steprogsteprog Posts: 227
    edited 2009-12-24 21:47
    Im wondering if that will be negated by the simple fact that the input frequency and the·LO frequency will be generated from the same source. In other words whatever jitter is created should be the same for the input of the mixer and the LO of the mixer.· I maybe okay.· However, my concern is phase shift.· I need to be able to shift the clock to optimize the phase.
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2009-12-24 22:45
    I'm really not sure from your description what it is you're trying to do. But if you have two counters running at the same frequency, but with different values in PHSx (i.e. phase shifted), you will get relative jitter between them. Again, the amount of jitter will be determined by the number of "one" bits in FRQA and FRQB.

    Also, when you change the phase shift on the fly, the correct way to do it is to increment FRQx, then decrement it by same amount. Incrementing or decrementing PHSx directly will only change its shadow register.

    -Phil
  • steprogsteprog Posts: 227
    edited 2009-12-24 23:02
    Thanks Phil,
    It looks like I need to do my homework on the counters. Just for your information Im trying to modulate a laser beam running through a fiber optic Im doing this with several beams and combining these beams together. Using a detector I can disect the beam by demodulating with the given frequency. Modulated to baseband I get the amplitude of the waveform that correlates to the phase of the laser. By doing this I can use a control loop to sync up the phase of the individual beams. Thereby having coherent light from multiple sources. I only know half of what Im doing but its fun.
    Thanks for the help,
    Greg
  • steprogsteprog Posts: 227
    edited 2009-12-26 06:18
    Phil Pilgrim (PhiPi) said...
    Using two counters, you can send out square waves that are in quadrature with each other (i.e. I and Q). This is done by presetting the phase registers. You may, however, find that the output is unsuitable for RF work, due to the phase jitter.

    -Phil

    Phil,
    I've gone through the counter description but haven't found a way to do the quadrature output you describe. Can you point me to more details on this?
    Thanks,
    Greg
Sign In or Register to comment.