problems on integer N PLL harmonic locking
It's a Integer PLL with N=8,
During PLL test, we find the 8th-order harmonic of the reference signal can make the oscillator itself locking at 8*fref , and the pll loop seems to have no help on the tuning of the PLL.
It's called something like "Injection and Locking or Pulling" effect, and does anyone have similar experience or How can I reduce the coupling and make the PLL work normal ?
Look forwards,
8 times Fref or 8 times Fvco ? They are totally different.
If PLL locks 8th harmonic of the VCO, there is either a programmation error or 8th harmonic has higher RF level (uncommon?, 8th harmonic should normally be much lower).Reduce this level by aid of Low Pass Filter.
If PLL takes 8th harmonic of Fref as a comparison frequency,the same thing may happen.Fref signal is too bad and therefore 8th harmonic is higher than fundamental one.Use a clearer reference signal.
Hello BigBoss,
Thanks for your feedback. I think you are wrongly understood what is my meaning.
It is a measurement results, and the circuit works well in simulation.
It means due to the 8 times harmonic of fref couples to the circuitry near the ring oscillator, and also due to the divider ratio is 8 , and the oscillator should also works around 8*fref after the coarse(PVT) calibration, so then the oscillator itself locking to 8times of fref (something like injection and locking) without the help of the PLL loop. It also will cause the fine tuning of the oscillator out of work.
Finally, the spectrum of output of the oscillator is still like a perfect locking ,but it is not the phase locking ,just frequency locking (like a frequency mutiplexer)
is this a trick question? Insert a lowpass filter at the oscillator output.
Ps, this PLL is a digital PLL, and so the higher order harmonic of the reference should be around the chip which is easily to make DCO failure for the work