Who This Checklist Is For
If you're working on XGS-PON deployments and using an EXFO FTB-400 (or just inherited one from a retiring engineer), this list is for you. I've been handling fibre testing orders for six years, and in that time I've personally made — and documented — 14 significant mistakes, totaling roughly $8,700 in wasted budget. The worst one was in February 2024: I set the wrong wavelength for a 10G PON line test and had to redo an entire 48‑port splitter cabinet. That one cost us $1,100 in overtime and a 3‑day schedule delay. So now I maintain our team's checklist to keep others from repeating my errors.
This guide assumes you already have an EXFO FTB-400 equipped with the 1577nm and 1270nm modules. If you're using an older unit (like the FTB-200), some steps will differ — I'll note where they don't apply.
Step 1: Verify Your Test Module and Wavelength
Before anything else, confirm which module is installed in your FTB-400. The EXFO FTB-400 can accept multiple modules; for XGS-PON you need either the IQS-2400 or the OTDR-4800 (or similar). The critical point: XGS-PON uses 1577nm downstream and 1270nm upstream. If you accidentally test at 1310nm (common for GPON), your results will be meaningless.
What I mean is that the 1577nm module is for the downstream direction from the OLT to the ONT, and the 1270nm module is for upstream testing. Mixing them up is the #1 mistake I saw in my first year. I once ordered a 2‑piece FTB-400 rental with the 1577nm module when the job actually required 1270nm upstream — $680 wasted on shipping and a week delay. So glad I now have a laminated checklist taped to the unit.
Checklist item: Look at the module label. If it says "1577nm" and you're testing upstream, swap to the 1270nm module. (The EXFO FTB-400's hot‑swappable design makes this quick.)
Step 2: Configure the OTDR Settings
Once the module is correct, power on the FTB-400. The touchscreen interface is fairly intuitive — relatively speaking. I'll walk you through the key parameters you should set:
- Distance Range: Set to at least 1.5x your expected span. For a typical XGS-PON with 20 km reach, choose 30 km.
- Pulse Width: Use a shorter pulse (like 10–30 ns) for near‑end event resolution, or a wider pulse (100–300 ns) for longer distances. I usually start with 50 ns and adjust.
- Wavelength: Match to your module. If you're testing the downstream path, select 1577 nm.
- Duration: 15–30 seconds is typical. The longer you average, the more accurate the trace.
Why does this matter? Because a wrong pulse width can mask splice losses or make an event look like a reflection. In Q3 2024, I used a 300 ns pulse on a 10 km run and missed a bad fusion splice that had 0.25 dB loss — it looked completely flat. The contractor had to send a crew back out. $450 additional cost.
Step 3: Clean All Connectors First (Seriously)
This is the step that 70% of engineers skip when they're in a hurry. Dirty connectors create high reflectance that will throw off your entire trace. I've seen results where the OTDR read a −30 dB event that was actually just a speck of dust on the launch cable.
Take the launch cable (the short fibre you connect between the OTDR and the network), examine it with a fibre microscope — EXFO sells a nice FIP‑400, but any 400x scope works. If you see any contamination, use a click‑type cleaner (I prefer the 2.5 mm Cletop). It's tempting to think you can blow it off with canned air. But that just redistributes the dirt. Clean it properly.
Checklist item: Clean both the launch cable connector and the FTB-400's port before every test. I learned this the hard way after three consecutive rejections on a $3,200 inspection job in June 2023.
Step 4: Run the Test and Capture the Trace
Press the start button. While the OTDR acquires the trace, don't wiggle the fibre! Even a small movement can cause measurement errors. I once had a technician leaning on the launch cable — the resulting trace looked like a roller coaster. We caught the error when we compared it to the reference trace from the previous day. Lesson: use a stable launch cable tray or at least tape the cable down.
After the trace completes, save the file with a meaningful name — not "Test 001". Include the date, location, and wavelength. For example: "2025-01-15_CentralOffice_XGS_1577nm.sor". EXFO's FTB-400 software (FastReporter) can export to PDF, CSV, or directly to your cloud platform.
Step 5: Interpret the Results
Now look at the events table. You'll want to check:
- Total loss — should be under your link budget (typically 28–32 dB for XGS-PON).
- Reflectance peaks — high peaks might indicate a connector issue.
- Non‑reflective loss — a gradual drop suggests a bend or splice.
The 1577nm wavelength is more sensitive to bends than 1310nm, so if you see a sudden loss at 1577nm that wasn't there at 1310nm, you likely have a micro‑bend. This is where the industry evolution matters: five years ago, we mostly tested at 1550nm for GPON. With XGS-PON's 1577nm, the bend sensitivity increased by about 30%. Old best practices (like assuming a 2 cm bend radius is safe) may not apply in 2025. But the fundamentals — clean connectors, correct pulse width — haven't changed.
Common Mistakes & How to Avoid Them
Mistake #1: Forgetting to Check the Module Compatibility with Your EXFO FTB-400
Not all modules work with all FTB-400 chassis. Some older units (pre‑2017) may not fully support the 1270nm module's power requirements. If you get a "Module not recognized" error, update the firmware via EXFO's support site. I learned this when I tried to use a brand‑new 1577nm module on a 2015 FTB-400 — it loaded but gave inaccurate results. A $1,200 order had to be re‑tested.
Mistake #2: Misinterpreting the 1270nm Upstream Test
You can't directly test the 1270nm upstream from the OLT side because the OTDR transmits on the same wavelength — you'd interfere with live traffic. You need a separate test access point (like a WDM coupler at the OLT). Alternatively, test from the ONT side using a loopback. I had a colleague ask, “How do I turn on my Verizon 2660 flip phone to make a test call?” — we all laughed, but it reminded me that even with advanced gear like the EXFO FTB-400, sometimes the basics matter. The answer: you don't need a phone; you need an optical test plan.
Mistake #3: Not Creating a Reference Trace
Before you start, take a reference trace of the launch cable alone. This helps you subtract the launch effects from the actual measurement. If you skip this, you'll overestimate loss by 0.5–1.5 dB. In December 2024, that mistake caused us to reject a perfectly good splice because the loss reading was 0.9 dB higher than the threshold. We only discovered the issue when the contractor showed us his reference trace — we had used a different launch cable. Embarrassing.
Final Notes
This was accurate as of January 2025. EXFO's firmware updates and new module releases happen fairly often, so verify current specs at exfo.com. I learned most of this in 2020–2023, and things have evolved — especially with XGS-PON's growth.
Part of me wishes I had found this checklist earlier. Another part knows that making those mistakes taught me more than any manual could. I compromise by updating our team's checklist quarterly. If you have other tips for testing XGS-PON with the EXFO FTB-400, I'd love to hear them.
