Cable Fault Locating and Repair

0

Part 2. Identifying a Faulted Pair —

In my previous column (Part 1 in the June 2020 issue), I discussed the evils of the fast, easier “cut to clear” solution when repairing a faulted cable. This is a short-term fix that will cost more in time and resources eventually as you run out of pairs. I also provided a bulleted process for qualifying a cable pair as “good.”

In this month’s column, we look at using your test set to explore longitudinal balance as one of the elements that help identify a faulted pair. Additionally, I will provide the process I employ and teach to help students understand and integrate advanced fault locating techniques.

Longitudinal Balance

After connecting a multi-function test set to any cable pair working or not, that pair first should be tested with the test set’s digital multimeter (DMM) tip to ring, tip to ground, and ring to ground, for unwanted AC voltage.
Expect to see from .2VAC to 10VAC.

Ignore any AC voltage that measures less than 50VAC. It will not have any effect on service.

Voltage greater than 50VAC may be a safety hazard. Therefore, if the voltage is greater than 50VAC, follow your company’s safety practices that deal with hazardous voltage.

Longitudinal balance indicates the ability of a cable pair to reject induced noise on the line. It is a relative unit of measurement used to show the ratio of one power to another logarithmically, and it is based on the signal to noise ratio on a cable pair in decibels (dB).

With this in mind let’s consider what we can learn from longitudinal balance:
If the longitudinal balance shows greater than 60dB on a vacant cable pair, that pair shows adequate longitudinal balance, so go ahead and use it.

If the longitudinal balance shows less than 60dB, the longitudinal balance is unacceptable.

A longitudinally balanced cable pair has the same impedance tip to shield and ring to shield.

Impedance is the resistance of a cable pair to AC current flow.

If any AC current flow tip to shield and ring to shield are equal, the pair is longitudinally balanced, noise is reduced, the noise floor is lowered, and interference from other circuits on adjacent cable pairs is also reduced.

If any AC current flow tip to shield and ring to shield are not equal, the pair is not balanced longitudinally. The unbalance is created by a fault on the tip or ring of the cable pair.

An Example of Unacceptable Longitudinal Balance

To understand what is happening with your test set when the test set is calculating the longitudinal balance, let’s look at an example of a noise mitigation complaint that could have been due to unacceptable longitudinal balance.

On a POTS circuit, the customer complaint is a hum on the line or static on the line. Typically, audible circuit “hum” problems are caused by bonding and grounding issues or harmonics of the base 60Hz AC from the power line because of distribution power company issues. The combined harmonics are always there and are coupled to the tip to ring and tip and ring to the shield and or ground of a cable pair. The customer will not hear the induced hum on the line unless the amplitude of the combined harmonics is too high or the cable pair is unbalanced. On the other hand, static on the line is caused by an unbalanced cable pair.

Let’s look at several manual longitudinal balance calculations and their probable root cause.

The power influence (PI) and circuit noise (CN) can be measured with any transmission test set. Power influence should never be greater than 80dBrnC and circuit noise should never be greater than 20dBrnC. Then the longitudinal balance can be manually calculated using the following formula. Acceptable longitudinal balance should be greater than 60dB.

The formula for calculating longitudinal balance is power influence (PI) – circuit noise (CN) = longitudinal balance (LB).

Using a transmission test set at the customer’s network interface and measuring:
Example 1. PI 80dBrnC – CN 20dBrnC = LB 60dB indicating a balanced cable pair.

Example 2. PI 70dBrnC – CN 25dBrnC = LB 45dB indicating a faulted cable pair.

Example 3. PI 90dBrnC – CN 30dBrnC = LB 60dB indicating that the cable pair is good. Both power influence and circuit noise failed indicating a noise mitigation issue such as bonding and grounding or an associated distribution power issue.

Example 4. PI 50dBrnC – CN 10dBrnC it would seem that the longitudinal balance is 40dB, but because PI is less than 60dBrnC you cannot do a longitudinal balance calculation.

This is where the multi-function test set shines. No matter what the actual power influence is when you select the longitudinal balance the test set sends an AC signal at 90dB, then measures the noise across the pair, subtracts that amplitude from 90dB and displays the Longitudinal Balance pass/fail. If there is a fail test the pair ring or tip and ring to ground for crossed battery, a shorted cable pair, a tip or ring ground, an open or split cable pair, or series resistance.  Use your test set’s RFL, TDR, and /or open meter functions to locate the fault.

There Are a Couple of Caveats to the Longitudinal Balance Test

When there is a series resistance fault like an old twisted splice or a faulty MR connector or a popped scotchloc, unless there is at least 1,000 feet of cable pair beyond the series resistance, a longitudinal balance will pass when it should have failed.

To circumvent that faulty result, place a grounded short at the far end of the cable pair and then run the longitudinal balance test. If the electrical center is in the middle, then your set will show a balance greater than 60dB. If the balance test fails, then measure tip to ring, then tip and ring to ground in ohms.
An example of a Pass would show tip to ring 100 ohms, tip to ground 50 ohms, ring to ground 50 ohms.

An example of a Fail would show tip to ring 110 ohms, tip to ground 50 ohms, ring to ground 60 ohms, indicating a series resistance of 10 ohms on the ring conductor. To find the series resistance, go halfway and test again.

A split cable pair will show a longitudinal balance test around 60dB, usually around 58dB or 59dB.
Generally, when measuring the distance to a balanced cable pair with an open meter, the tip to ring measurement is subtly longer than the tip and ring to ground measurement.

On a split cable pair the tip and ring to ground measurements are subtly longer.

If the pair is split, then put tone on the pair and run the count. The split pair will have about the same amount of tone as your pair. All other pairs will have less tone. Your TDR function will show the distance to the split pair.

I hope this provided you with a more in-depth look at the importance of longitudinal balance, and how to best use the readings provided and to correct for misinformation.

In Part 3 of the series, we look into the Digital Multimeter and the Resistance Bridge.

Like this Article?

Subscribe to ISE magazine and start receiving your FREE monthly copy today!

Signing off
If this information is helpful, let me know. If I’ve misspoke, I especially want to hear from you. Doing something for 40 years doesn’t mean I get it right all the time. According to my wife, not even half the time! Reach out to me at 831.818.3930 or dmccarty@mccartyinc.com.

To read Part 1 of this 3-part series, please see “Cable Fault Locating and Repair, Part 1. Short-Term and Long-Term Approaches” in the June 2020 issue of ISE magazine at https://www.isemag.com/2020/06/telecom-cable-fault-locating-and-repair-part1/.

Another Ongoing Use for Copper Lines

Don,
In an article in the January 2020 issue of ISE magazine, the author mentions the use of copper pairs to supply remote power (48Volt DC) to small cell sites required by nationwide 5G build-out.

The other alternative is to use power run over coax cables which require up converters to 190 volts dc and down converters back down to Telecom standards of -48 volts DC. (Source: “Small Cells Unplugged”, ISE magazine, January 2020, pages 42-47)

Maybe that’s another reason to keep the copper plant in good shape for the short-term future.

From my little experience, it’s still costly to interface to a fiber pair for anything but 10/100 Ethernet. Those interfaces are cranked out by the thousands in China or other island manufacturing centers, and are very inexpensive.

Anything else like T-1 circuit interfaces come at a premium price to light up the fiber.

Of course you have to justify the start-up costs with fiber but it’s a steep leap to get a single circuit going versus copper lines. It easily makes sense when you can multiplex which drives the per-connection cost down.

Good luck in your efforts with maintaining the Copper Age.

I do read and enjoy your columns monthly.

Bye.

Farrell Silverman
State of Maryland, Department of Natural Resources
Hollofield Radio Shop
Ellicott City, Maryland

 

Related

About Author

Don McCarty is the OSP EXPERT columnist for ISE magazine, discussing the issues around provisioning, testing, and maintaining copper for all services from POTs to IPTV. Don is also president of and the lead trainer for McCarty Products, a technical training and products company training field technicians, cable maintenance, installation repair, and Central Office technicians and managers. For more information, email dmccarty@mccartyinc.com or visit www.mccartyinc.com.

Comments are closed.