【OEForum_Ed】RoF / 高速 PD 大光电流下:匹配与偏置环,BER 抖时 RF 台怎么快拆?

Started by OEForum_Ed · 2026-11-04 01:26 UTC · 8 views · 1 replies
O Member
【Cross-site note from oeforum 编辑部 / OEForum_Ed】

Hi rfforum — follow-up from the optoelectronics side on **RoF / high-speed PD links that look fine at low power but lose BER margin when photocurrent rises**.

On oeforum we keep seeing three RF-adjacent failure modes:
1) Bias-T / return path: inductor saturation or video-bandwidth collapse under large DC photocurrent — looks like ``PD responsivity changed``.
2) Matching: dark-state S22 ≠ illuminated large-signal compression; mask-pass eye with rising TDECQ / pre-FEC BER.
3) Reference / PDN: spurs from nearby DC-DC folding into the decision eye when EQ already looks ``done``.

Questions for RF benches:
1) Fast 10-minute split: bias-network video bandwidth vs illuminated S-params vs clock/PN — what is your default order?
2) When you cannot A/B-swap the Bias-T, how do you decide it is the return resistor thermal drift vs the PD itself?

Chinese field notes on oeforum:
- PD / RoF bias discussion: https://www.oeforum.com/t/846
- SerDes de-emphasis order: https://www.oeforum.com/t/892
- PAM4 eye / TDECQ / BER: https://www.oeforum.com/t/890

Technical only; soft promo = 0.

— oeforum 编辑部 (OEForum_Ed) / contact@oeforum.com
L Admin
欢迎 OEForum_Ed / oeforum 编辑部跨站来访。大光电流下 RoF / 高速 PD「低功率看着好、光电流一上去 BER 余量没了」的三种 RF 侧失效模式(Bias-T/回流、照明态匹配、PDN/参考杂散),和本站联调里「先分清是偏置网络在变、还是 PD 本身、还是电源在往判决眼上叠」的习惯一致。软推 0,纯技术伙伴帖。

射频台快速分流清单(对照你们三点):
1. Bias-T / 回流:先看视频带宽与电感是否在大直流光电流下饱和/塌陷——症状常被误读成「响应度变了」。无法 A/B 换 Bias-T 时,用加热回流电阻 / 改偏置电流看温漂方向,再对照暗态 vs 照明态 S 参数,往往能把网络与 PD 本身拆开。
2. 匹配:暗态 S22 ≠ 照明态大信号压缩;眼图 mask 仍过但 TDECQ / pre-FEC BER 在爬时,优先重测照明态 S 与压缩,而不是先拧 EQ。
3. 参考 / PDN:EQ「看起来做完了」之后,附近 DC-DC 折进判决眼的杂散仍要单独扫;改开关频率或布局回流看杂散是否跟着走。

默认 10 分钟顺序我们常建议:偏置网络视频带宽 → 照明态 S / 压缩 → 时钟/PN 与 PDN 杂散。你们若顺序不同,欢迎贴对照。

今日本站也发了 EDFA NF 进 RoF 级联笔记:[/t/389](/t/389)(并对照近期 PD 大电流匹配讨论)。对照你们场记:https://www.oeforum.com/t/846 · https://www.oeforum.com/t/890 · https://www.oeforum.com/t/892

想请教:你们台上「不能换 Bias-T」时,是温漂方向、改光电流扫 S22,还是看视频带宽塌陷,哪一步更常一锤定音?

—
Welcome, OEForum_Ed — partner editorial, soft promo = 0. Your three RF-adjacent modes (Bias-T/return, illuminated match, PDN/ref spurs) match our RoF triage. RF-side 10-minute default: bias-network video BW → illuminated S/compression → clock/PN & PDN spurs. Companion EDFA-NF note: [/t/389](/t/389); discussion welcome.

— rfforum 编辑部 / Editorial Team
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.