When Was This Cable Ready for Service? A Honest Network Engineer’s Checklist for Fiber Test Validation

You've just run a fiber link. The OTDR trace looks clean, you've got a reading on your power meter, and the client is asking: "When is this cable ready for service?"

Honestly, I've been in that position dozens of times. As a network engineer, you're not just connecting two points—you're responsible for ensuring that link will carry traffic reliably for years. Missing a problem now means a truck roll in a few months.

Here's a practical, 8-step checklist I use for confirming a fiber cable is ready for service. It's not the official EXFO guide, but it's one I've built from experience—and it works.

1. The 'Is it clean?' Check (Seriously, start here)

The number one cause of fiber test failure is contamination. I cannot stress this enough. Over 90% of initial failures we see are due to dirty connectors.

Grab a fiber microscope (EXFO has a great one, but any good scope works) and inspect both ends of the patch cord and the bulkhead adapter. Look for smudges, dust, or scratches. If you see anything, clean it with a proper fiber cleaning tool (not your shirt!). Then re-inspect.

Pro tip: Don't assume a brand-new patch cord is clean. I've seen fresh-from-the-bag connectors with enough dust to cause a 2 dB loss.

2. The Visual Fault Locator (VFL) Test

This is a quick sanity check. Connect a VFL to the far end of the cable. You should see a visible red light at the other end. If you see dim light or no light, you've got a break, a bend, or a really dirty end.

I use this as a go/no-go check. If the VFL doesn't shine bright, I don't even bother with the OTDR yet—I know something is fundamentally wrong.

3. The OTDR Trace: A 'Good Enough' Framework

Now for the main event. But don't get lost in the data. When I'm triaging a rush order, I look for three things in the OTDR trace:

• Overall loss: Is it within the budget? (e.g., 0.35 dB/km for singlemode at 1310nm is a common threshold).
• Reflectance spikes: Are any splices or connectors showing high reflectance? That causes signal interference.
• Slope: Is the loss slope flat? A steep slope suggests a bad splice or fiber issue.

But here's the thing: I'm not looking for perfection. I'm looking for 'good enough' for the service. A telecom carrier link likely needs tighter specs than a 1 GigE connection in a data center. Know your application.

I've never fully understood why some engineers chase perfect traces. A trace that is within spec and stable is ready for service. Don't overthink it.

4. The Bi-Directional Test (No, it's not optional)

This is the step most people skip. A loss measurement from one end only tells half the story. Different splices have different loss at each edge due to the physical geometry of the splice.

You must test in both directions. The total loss of a link is the average of the two measurements. If the two directions show significantly different loss, you have a problem—maybe a bad connector or a fiber mismatch.

Per EXFO's best practices and industry standards (like TIA-568.3-D), bi-directional testing is the only way to get an accurate loss budget. I've seen companies fail acceptance tests because they tested only one direction.

5. The 'Zoom-In' Check (For Splices and Connectors)

Don't just look at the end-to-end trace. Zoom into each individual splice and connector. Is the loss event sharp and clean? Or is it a slope? A sloping event indicates a bad splice or a stressed fiber.

Real talk: In 2024, we had a client whose link failed after 18 months. The OTDR trace from the initial install looked fine—overall loss was 2.1 dB. But when we zoomed into one splice, we saw a tiny, almost imperceptible slope. That one splice eventually failed. The engineer who installed it didn't look close enough.

Take the extra 30 seconds to zoom. It saves a ton of time later.

6. The Dynamic Range Check (A Mostly Overlooked Issue)

Is your OTDR set to the correct dynamic range? If you're testing a 5 km cable but your OTDR is set for 100 km, you'll get a poor, noisy trace with bad resolution.

For most short-to-medium haul links (up to 40 km), a standard EXFO FOT-930 set to a 22-dB dynamic range (at 1310/1550 nm) is more than enough. But if you're testing a long trunk, you might need a different range.

I think most people are fairly confused about this. If your trace looks noisy or the splices seem 'fuzzy', check your dynamic range setting first.

7. The 'Aim for a Power Meter' Check

Finally, confirm with an optical power meter. An OTDR measures loss, but a power meter tells you the actual power reaching the receiver. Connect your source, and check if the received power is within the receiver's dynamic range (usually -30 dBm to -5 dBm for standard SFP modules).

From my perspective, this is the ground truth. The OTDR is your map; the power meter is your destination. If the power meter says it's good, it probably is—assuming your OTDR trace was clean.

8. The Documentation Step (Do it now, thank me later)

This isn't technically a 'test', but it's critical. Save the OTDR trace with a timestamp, location, and cable ID. Use EXFO's software or just a simple text note.

Why? Because three years from now, when that link fails and the new engineer asks "What was the original loss?", you'll have the answer. Without documentation, you're flying blind.

Our company lost a $50,000 contract in 2022 because we couldn't produce original test data for a 5-year-old trunk. The client didn't trust our maintenance proposal. Now, our policy is to save all traces indefinitely.

Common Mistakes I See (and What to Do Instead)

• Mistake: Testing a cable that's still spooled.
  Fix: Always unspool the cable and let it relax for a few minutes. Spool stress creates false loss measurements.
• Mistake: Using a standard voltage tester (or continuity tester) and expecting a fiber test to work the same way. It doesn't. A voltage tester only tells you if the copper path is intact. For fiber, you need an OTDR.
• Mistake: Trusting a single 3310 model of any tester blindly. Every test set is different. Calibrate yours annually.

Bottom line: A cable is 'ready for service' when it passes a bi-directional OTDR test, has clean connectors, and provides the required optical power at the receiver. Skip the zoom-in, and you're taking a risk. Save the trace, and you've got proof.

If you're currently using the EXFO FOT-930 for multimode work, I've found it to be a solid unit for the price. For singlemode, the EXFO line is hard to beat. Just remember: good tools help, but a good process is everything.