I’m often asked technical questions about copper and fiber optic cabling. The majority are about problems troubleshooting cabling systems or networks. Some have simple answers but not all. Here are some questions we’ve been asked; see if you can answer them. Answers and explanations are in red.
1. Which of the following is more likely to cause rather than prevent fiber optic cable plant problems?
A. Periodic maintenance, which includes cleaning and testing
B. Placing patch panels and splices behind closed and locked covers
C. Running cables through high risk areas in innerduct
D. Leaving the cable plant alone
Fiber optic networks do not need maintenance; it’s more likely to cause damage than prevent problems. If possible, lock up all the connections to keep meddlers out and leave everything alone.
2. When troubleshooting a network that is having problems, the first test should be to measure power at the __________.
D. Test source
Always test the receiver power first. If it’s within spec, the network has an equipment problem. If it's too high, you need to add an attenuator to reduce the power. If it's too low, it means you have a cable plant or transmitter problem.
3. If a single-mode system has proper power levels but will not work, the link should be tested for reflectance.
Reflectance can cause distortion in laser transmitters or high background noise called multipath interference on short links.l
4. When testing an installed single-mode cable with a light source and power meter, all the fibers tested showed about a 20-dB loss, but inspection with a microscope showed all the connectors looked OK. The fibers showed continuity with a visual fault locator. The most likely problem is __________.
A. High stress point in the cable
B. Poorly polished connectors
C. Testing single-mode fibers with multimode reference cables
D. Bad test source
A confused installer did not realize the reference cables being used for testing singlemode fibers were actually multimode, and the mismatched losses were what he was measuring. Retesting with singlemode reference cables showed the fibers to be all be good.
5. When testing an installed multimode cable with an automated optical loss test set, all the fibers tested showed about a 3-dB loss, but inspection with a microscope showed all the connectors looked OK. The fibers showed continuity with a visual fault locator. The problem could be __________.
A. Worn-out mating adapters
B. Bad connectors on the reference cables
C. Testing 50/125 micron multimode fibers with 62.5/125 micron reference cables
D. Any of the above
Any of these could be a problem, but since it’s around 3 dB, the typical loss when transmitting from mismatched 50/125 and 62.5/125 fibers, we’d check the reference cables first.
6. The fibers in a cable pulled in conduit tested with about the same loss at 850 and 1,300 nm, instead of much lower at 1,300. The most likely cause is __________.
A. Stress or excess bends on the cable
B. Highly reflective connectors
C. Bad reference cables
D. Problems with the test equipment calibration
When stressed, a fiber will show higher loss from the stress at longer wavelength, so with enough stress, the loss at the longer wavelength, which should be lower, can be equal or higher to the loss at a lower wavelength.
7. Optical time domain reflectometer (OTDR) test results from one fiber in a 300-meter cable showed low loss on the fiber, but the fiber could not transmit any light from a visual fault locator. The most likely problem is __________.
A. Improper setup on the OTDR
B. A bad connector on the far end of the cable
C. Ghosts in the trace
D. Wrong analysis of the trace
Unless you use a receive cable on the far end of the cable being tested, the OTDR will not measure the loss of the connector on the far end.
8. After testing more than 100 fibers at 850 and 1,300 nm, the installer discovered the wavelength on the power meter was set at 850 nm for all measurements. What should he do?
A. Retest all the fibers with the meter set to the wavelength of the source.
B. Change the 1,300 nm readings with the calibration factor from the meter.
C. Change the 850 nm readings with the calibration factor from the meter.
D. Nothing. The data is OK as is.
Loss is a relative measurement, the difference between two readings on the meter, so the dB loss measured will be the same no matter what the calibration wavelength of the meter.
9. Certification testing of a Category 6a link failed for return loss at the far end connector. The first thing to check is __________.
A. Kinks in the cable near the connector
B. Too much untwisting in the pairs at punchdowns
C. The jack is not Cat 6a rated
D. All of the above
Any of these can cause impedance mismatches at the connector. They are all possible, but usually untwisting the pair too much is the problem.
10. In terminating an old 25-pair cable run between two buildings, the cable had larger 19 gauge wires, which are not compatible with modern punchdown blocks (newer category-rated pairs are 22–24 gauge).
The 19 gauge wires are only 0.01 inch larger than the largest wires specified for a punchdown block, 20 gauge, so a punchdown block can accept the wires.
HAYES is a VDV writer and educator and the president of The Fiber Optic Association. Find him at www.jimhayes.com.