Noise Mitigation is one of the many tasks a quality cable maintenance technician will occasionally encounter. Sadly, few techs receive any training on noise mitigation and yet related problems can cause expensive troubleshooting and repair and involve long hours. But there is good news: First, many noise mitigation problems are really bonding and grounding problems. Second, while experienced, qualified, noise mitigation techs have many years of experience under their belt, there is a process that isn’t difficult and can help you locate many noise or bonding problems quickly.

To begin with, if power influence (PI) is over 80dBrnC, it will have an adverse effect on the end user and his equipment. The audible hum on the line is the first indication that there is a problem. The following step-by-step testing will help you locate the trouble, indicating if the problem is on one pair or the complete count of the cable.

Inactive (Vacant Pair Testing)
Remove the battery and ground from the cable pair at the central office (CO) or the remote and test the pair for:

AC Volts Tip and Ring to Ground
.2VAC to 10VAC is common on non-loaded circuits. Beyond 18,000 feet (5,550 meters) AC voltage tip and ring to ground increases. More than 50VAC presents a hazard so use safety precautions.

Longitudinal Balance
Longitudinal Balance should show greater than 60dB and if it fails to identify any balance problem and repair the root cause, look for:

Crossed battery, a tip or ring ground, series resistance, or a capacitive unbalance such as open one side on a lateral, open one side beyond the customer’s terminal, or crossed with a non working conductor. This should solve the noise problem if only that pair is affected.

Active (Working Pair Testing)
Restore dial tone and test for:
Loop Current...................>23Ma
Station Ground................0 to 25 ohms
Circuit Loss.....................< 8.5dBm
Circuit Noise...................<20dBrnC
Power Influence (PI).......<80dBrnC

Noise Mitigation
When PI is >80dBrnC, noise mitigation is in order. The root cause could be bonding and grounding, distribution power associated problems, or a combination of both. With distribution plant in disorder, there is an 80% likelihood that unacceptable bonding and grounding is the issue.

I trained a group of telephone technicians on the basics of noise mitigation the first week in May. We had the opportunity to go to the field on an unacceptable PI problem. The PI measured >110dBrnC at the customer located 4.5 miles from the remote. The cable in question was a 19-gauge filled buried cable.

Two quality field technicians had repaired the bonds in 10 pedestals from the customer to the cross-connect box 2 miles from the remote. Using their clamp-on ammeter to identify bad bonds, they managed to raise the current flow on the shield from an average of 0 Ma to 4 Ma to an average of 1.4 amps. PI was 98dBrnC with the pair open at the cross-connect box, indicating more bonding and grounding issues toward the remote.

We identified bonding issues using the clamp on ammeter in 14 more pedestals. These bonds looked good until they were taken apart. Some of the bonding clamps had series resistance, other bonds were improperly made, and lightning opened others.

When ALL of the pedestal bonds were repaired, current flow increased to 4 amps on the shield and now it needs a route back to the power substation other than on the cable pairs. That is the job of the distribution power neutral.

In the 4.5 miles the telephone shield needs to be tied to the power neutral at a maximum distance of every 1,000 feet (300 meters) and there are only 6 power neutral grounds on the 4.5 mile lead. Telco engineering was to look at the route on May 16 and interact with the power company to have the power neutrals strategically placed near pedestals for telco shield grounding to the distribution power neutrals.

The telephone cable is on the west side of the road and the distribution power is on the other side of the road, therefore when the power distribution neutrals are placed the telco has a choice of going overhead or boring the road and placing #6 ground wire between the pedestals and the power distribution neutral.

Conclusion (So Far)
Until the telco shield is correctly bonded to the distribution power, we will not know if PI will be reduced to <80dBrnC. We do know that with continuous bonding all frequencies that are greater than 1 kHz have been mitigated. We still could have a malfunctioning capacitor bank, an undersized distribution power neutral, a saturated transformer, over voltage, or a myriad of other distribution power associated problems that need to be addressed.

If all else fails, we can properly place an induction neutralizing transformer in the correct spot and that will bring PI down to less than 80dBrnC. That will make the customer happy, especially if it is a DSL INT.

That’s the downside of noise mitigation. All too often the copper plant has been abused and not proactively maintained for the last several years. If your company plans on being in the Triple Play business now is the time to go proactive in the provisioning and maintaining that Last Mile of copper for Fiber-to-the-Node (FTTN) Triple Play service.

Signing off
How important is noise mitigation to you in your day-to-day troubleshooting? I'd love to hear about your experiences because I'm hearing that problems are increasing. Let me know if you have questions about this or any other issues and I will address them. Thank you for reading. Contact me at dmccarty@mccartyinc.com or call me at 831.818.3930.

It's not too early to plan for OSP EXPO 2011, the leader in OSP education. Tell your boss this is a must-attend event: September 12-14 in Cincinnati, Ohio. I'll be presenting and you can challenge me with any questions -- I love being stumped! Get the details at http://www.ospmag.com/expo.