【Editorial】PLL lock-detect is high, but the carrier is not on frequency: false lock, a sticky pin, or the reference?

Started by OfficialEditorRF · 2026-11-14 05:18 UTC · 14 views · 1 replies
O Member
【Editorial Discussion】

A lock-detect pin that is already high is not the same thing as a carrier sitting on the intended frequency. The pin is a convenience for firmware and bring-up; it is not a frequency measurement. When the indicator and an independent frequency check disagree, it is worth writing down which of the usual suspects you have actually cleared before you retune the loop or swap the part.

1) Confirm the carrier with something that does not share the lock-detect path. A counter, a spectrum analyser, or a second receiver referenced to a known standard will tell you whether the output is on frequency, on a harmonic or subharmonic, or merely nearby and noisy. If the pin is high while the carrier is clearly elsewhere, treat the indicator as a separate question from the loop itself.

2) False lock is a real loop behaviour, not only a pin fault. A phase-frequency detector can settle into a stable but incorrect state, especially around a harmonic, a strong spur, or a reference path that is itself dirty. Changing the reference level, briefly opening the loop, or forcing a controlled relock and then repeating the independent frequency check often separates a false lock from a pin that is simply stuck high.

3) The lock-detect circuit has its own thresholds, filters, and pull-ups. A sticky or poorly filtered pin can stay high while the loop is still cycling, or go high on a periodic beat that is not a true lock. If you only look at the pin, you will not see this. Scope the pin and the output together, and note whether “locked” survives a small change in temperature, supply, or reference level.

4) Do not skip the reference. A wrong crystal, a mis-programmed divider, or a reference that has moved will put the carrier in the wrong place even when the loop is honestly locked to what it is being given. In that case lock-detect can be telling the truth about the loop and still leave you off the intended channel. Check the reference frequency and the programmed ratios before you blame the VCO or the pin.

Open question: the last time lock-detect was high and the carrier was not on the intended frequency, what actually closed the gap — a false lock in the loop, a misleading pin, or the reference and divider programming? Please share the kind of synthesiser and how you proved it; no customer names or unverified numbers needed.
A Member
Thanks for sharing, very thorough. We learned this lesson the hard way too.
广告 AdRenesas 瑞萨微控制器芯片产品照片Renesas Electronics微控制器、模拟与射频/时钟方案 · MCUs, analog, RF and timing访问官网 →免费展示 · 资料来自公开信息 · 如需撤下请联系 ads@rfforum.com
广告 Ad广告位招商 / Advertise hereads@rfforum.com · 刊例 Rate cardT2 · 728×90
Log in or register to reply.