Analog RoF gain ripple that tracks the MZM bias controller, not the fiber

Started by LunaWei · 2026-11-16 01:48 UTC · 24 views · 2 replies
L Admin
[Technical article / FAQ, rfforum Editorial Team]

An analog RFoF link whose RF gain or SFDR ripples on a seconds-to-minutes scale is often the Mach-Zehnder bias loop hunting, not fiber dispersion or connector dirt. If the ripple correlates with bias voltage or second-harmonic lock error, stop the fiber scrub first.

1. What the bias loop owns
- Quadrature lock via second-harmonic null (or dither) keeps SFDR; a loose loop walks the operating point and paints RF gain and even-order products as "fiber fade".
- Optical power and photocurrent change with the same walk. A gain ripple that tracks bias DAC / error signal is electronics until proven otherwise.
- Cold start vs soaked lock points differ. A link that "settles after lunch" is often bias thermal, not the spool in the rack.

2. Freeze list before you re-polish connectors
- Bias mode (open-loop set-and-forget vs closed second-harmonic / dither), error signal, and whether ripple frequency matches the loop
- Optical power / photocurrent logged on the same time axis as RF gain
- Freeze bias, open the loop briefly, or hold DAC: if ripple dies, the fiber was never the patient
- OMI / drive level: compression can amplify bias sensitivity

3. Habit
Never close an "RFoF intermittent fade" ticket without a bias-error vs RF-gain overlay. Archive lock mode, soak time, and photocurrent next to every SFDR claim. If holding bias kills the ripple, schedule fiber work later.

4. Practical gate
- Scope or log bias error vs RF detector for one ripple period
- Hold / freeze bias for two minutes; re-check gain
- Only then chase connectors, EDFA, or dispersion

Discussion: on your last RoF gain hunt, did the fix land in bias lock, drive/OMI, or a real optical path — and do you require a bias-hold A/B before opening a fiber ticket?

— rfforum 编辑部 / Editorial Team
L Admin
Addendum — bias-hold A/B before a fiber ticket: overlay bias-error or second-harmonic lock versus RF gain for one ripple period, then freeze or hold the DAC for two minutes. If the ripple dies, the spool and the connectors can wait.

Traveler on the same sheet as any SFDR claim: lock mode (open-loop set-and-forget vs closed second-harmonic / dither), photocurrent, OMI / drive, and soak time (cold start vs after lunch). Compression at the drive can amplify bias sensitivity — write the drive level too.

Open: on your last RoF gain hunt, did the fix land in bias lock, drive/OMI, or a real optical path — and do you require a bias-hold A/B before opening a fiber ticket?

— rfforum 编辑部 / Editorial Team
N Member
Good stuff, bookmarked. Will try this approach in our next build.
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.