Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

[Moved]: Locking range of very low bandwidth PLL

Status
Not open for further replies.

whedaboard

Newbie
Joined
May 3, 2013
Messages
2
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,323
Hi there.

I have a confusion with the locking ability of very low bandwidth PLL/FreqSynthesizer:

Talking about a Type-2 PLL, consisting of PFD, ChargePump, 2nd-order LPF ((R+C1)//C2), VCO and N-divider (let's assume N=1 for simplicity).

If we design the PLL bandwidth very very low, to tolerate the high input jitter, or in the case of a Clock-Recovery PLL which has extremely low loop bandwidth to tolerate the occasional "11" or "00", then we have to have very very low loop gain to make the loop stable (enough phase margin in open loop transfer function) due to the open loop transfer function.

However, if we want to have a very low loop gain, given that Kvco and PFD gain are probably inflexible, the only freedom in design is the charge pump current Icp. If we have a very small Icp, then locking range of the PLL will be very small, because the the loop is so weak that it doesn't have enough gain to pull the VCO to the target locking frequency.

Then I'm wondering, how can these kinds of very low bandwidth PLL even work/lock? I know there're techniques like AFC, which sweeps the Vctrl of VCO and limits the Vctrl within a small voltage range to assist lock. But does it mean if I don't use assisting technique, the loop by itself just simply won't work?
 

You can use a dual loop PLL.
You can look at dual loop PLL as having one very low bandwidth path (but high gain) that keeps changing the 'center frequency' of the VCO. During the initial lock time, you can increase the bandwidth of the path to get fast lock, and then when you are close to lock, you kill the bandwidth to ensure stabiliy.
Look up Eric Naviasky's paper on the same.
 

I have seen a loop filter where a simple silicone diode was placed together with R,C components. For a large signal from analog phase detector the loop gain was different than for a low signals (when oscillators were close to each other). It was AFAIR in 'Phase Lock Loops and Frequency Synthesis' by VF Kroupa.
 

I have seen a loop filter where a simple silicone diode was placed together with R,C components. For a large signal from analog phase detector the loop gain was different than for a low signals (when oscillators were close to each other). It was AFAIR in 'Phase Lock Loops and Frequency Synthesis' by VF Kroupa.
Shorting resistors while loop seems like a nice idea.
But I wonder what guarantees that if the loop is unstable to begin with, it will lock?
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top