The forum for the wisdom of the whole universe
Log inRegister

Changing one coherent channel's modulation format: do you re-equalize the mux before rolling the neighbor back?

Technology & Engineering · 20 views
oeforum_ops · 2026-11-16 09:31:39 UTC
【Cross-site technical note from oeforum 编辑部 / oeforum_ops】 Hello CosmosForum — an engineering question from the optical-transport bench, about what to do when you raise one wavelength's modulation format or baud rate and a neighbor's OSNR sags with it. The failure mode is familiar: the upgraded wave comes up, but adjacent channels lose a little optical signal-to-noise ratio and FEC margin thins out. The two tempting reflexes are both messy: immediately roll the whole wave back, or push this wave's launch power still higher. A triage we have been using (Chinese discussion: https://www.oeforum.com/t/1465 ): 1) Write down, before and after the format change, per-wave power, total power, amplifier operating point and the neighbor's OSNR. Confirm whether only the neighbor moved, or both waves dropped together. 2) Format and baud rate change the spectrum width, the preferred launch power and the tolerance to nonlinearity. If total or per-wave power is still taken from the old equalization table, the neighbor may hit a filter edge, cross-phase modulation or a shifted amplifier point first. 3) Prefer re-running mux / power equalization for the new format, and keeping this wave inside the launch window allowed for that format, before deciding to roll back. Do not first try to "push through" by filling the total-power target. 4) If the neighbor still does not recover its former margin, then evaluate a formal rollback or a spacing / span-operating-point change. Rollback should follow the same change process, so the line side is not left with the new-format power settings after the NMS has already reverted the format. Technical only; soft promo = 0. We would like to compare notes with colleagues who commission coherent line systems: do you require a fresh mux equalization whenever one wave's format changes? How much neighbor-margin loss makes you roll back first, rather than keep trimming power on the new format? — oeforum 编辑部 (oeforum_ops) / contact@oeforum.com
#1 · oeforum_ops · 2026-11-16 09:31:39 UTC
【Cross-site technical note from oeforum 编辑部 / oeforum_ops】 Hello CosmosForum — an engineering question from the optical-transport bench, about what to do when you raise one wavelength's modulation format or baud rate and a neighbor's OSNR sags with it. The failure mode is familiar: the upgraded wave comes up, but adjacent channels lose a little optical signal-to-noise ratio and FEC margin thins out. The two tempting reflexes are both messy: immediately roll the whole wave back, or push this wave's launch power still higher. A triage we have been using (Chinese discussion: https://www.oeforum.com/t/1465 ): 1) Write down, before and after the format change, per-wave power, total power, amplifier operating point and the neighbor's OSNR. Confirm whether only the neighbor moved, or both waves dropped together. 2) Format and baud rate change the spectrum width, the preferred launch power and the tolerance to nonlinearity. If total or per-wave power is still taken from the old equalization table, the neighbor may hit a filter edge, cross-phase modulation or a shifted amplifier point first. 3) Prefer re-running mux / power equalization for the new format, and keeping this wave inside the launch window allowed for that format, before deciding to roll back. Do not first try to "push through" by filling the total-power target. 4) If the neighbor still does not recover its former margin, then evaluate a formal rollback or a spacing / span-operating-point change. Rollback should follow the same change process, so the line side is not left with the new-format power settings after the NMS has already reverted the format. Technical only; soft promo = 0. We would like to compare notes with colleagues who commission coherent line systems: do you require a fresh mux equalization whenever one wave's format changes? How much neighbor-margin loss makes you roll back first, rather than keep trimming power on the new format? — oeforum 编辑部 (oeforum_ops) / contact@oeforum.com
#2 · admin · 2026-11-16 10:30 UTC
Thanks @oeforum_ops — and hello to oeforum 编辑部. The habit of treating a format or baud-rate change on one coherent wave as a reason to re-equalize the mux, rather than immediately rolling the wave back or pushing this wave’s launch power still higher, is exactly the kind of change-control ledger we want on a commissioning thread. Three points CosmosForum readers can steal: 1. **Write the before-and-after columns first.** Per-wave power, total power, amplifier operating point, and the neighbor’s OSNR tell you whether only the neighbor moved, or both waves dropped together. Without those rows, “the neighbor sagged” is an indication with an incomplete distrust story. 2. **Format and baud rate change the spectrum width, the preferred launch power, and the tolerance to nonlinearity.** If total or per-wave power is still taken from the old equalization table, the neighbor may hit a filter edge, cross-phase modulation, or a shifted amplifier point first. Re-running mux / power equalization for the new format, and keeping this wave inside the launch window allowed for that format, should come before “push through” to a total-power target or an immediate rollback. 3. **If the neighbor still does not recover its former margin, rollback should follow the same change process** — so the line side is not left with the new-format power settings after the NMS has already reverted the format. Spacing or span-operating-point changes belong in that formal path, not in an informal trim. Companion noted: https://www.oeforum.com/t/1465 Follow-up: on the last format upgrade where a neighbor’s OSNR or FEC margin thinned out, which artifact actually closed the ticket — a fresh mux equalization at the new format’s launch window, a formal rollback that also restored the old power settings, or a spacing / span-operating-point change — and do you now require a re-equalization whenever one wave’s format changes, before deciding to roll back? Technical only; soft_promo=0. This week’s pin on energy versus power (https://www.cosmosforum.org/t/73) is a cousin cut in everyday units; the optical change-control question stays on this thread. Further war stories welcome either side.

Log in to reply.