From: Induction Neutralizing Transformer Can Reduce Power Line Disturbances. Part 2 of a Two-part Series on the Effects of Power Line Disturbances on Nearby Metallic Telecommunications Facilities, Looks at Some of the Solutions to the Problem, by Russ Gundrum, Telephony, September 8, 1980
“Although an INT can be located anywhere in the cable plant and achieve a substantial voltage and power influence reduction, this may not achieve the maximum metallic noise reduction. To maximize the INT’s effectiveness in reducing metallic noise, it must be so located to reduce the power influence ahead of any unbalance, whether that unbalance is in the equipment or on a cable pair. Otherwise, after the longitudinal-to-metallic noise conversion has taken place, the noise will merely pass through an INT. To reiterate, the INT will NOT reduce noise if it already has been converted metallically.” (Otherwise you would never be able to have it pass ringing voltages!)
“The application of an INT in solving low-frequency induced voltage problems is not new. As early as 1914, units were being successfully applied to open wire telephone lines. The principle of neutralizing transformers is more commonly thought of in applications at power substations where metallic facilities are exposed to high fault induced longitudinal voltages and ground potential rise (GPR).”
“However, removed from the worst-case location (at a substation) and applied in the telephone plant, the INT is effective in suppressing induced voltage and current from faults of lower magnitude and their resulting impulse noise aspects, because the predominant 60 Hz component on which the other frequencies are superimposed during the initial surge is reduced substantially. Also, when the 60 Hz fundamental frequency is reduced, the harmonics of 60 Hz are proportionately reduced.”
“The INT is similar to the energy-absorbing transient suppression devices (zener diodes and varistors) in this regard, for it is permanently connected to the circuit and begins acting as the voltage rises. The INT does not have the delay limitations of spark gaps, which must wait for the voltage level to exceed a certain threshold before the device can effectively break down and provide protection. Thus, certain electronic components could have been degraded or damaged by a transient before the protectors operated.” (Some people think they have to protect solid-state equipment with additional solid-state equipment!)
“In the early 1970’s, the U.S. Navy began a controversial project in upper Wisconsin (selected because of its remote location and extremely high earth resistivities) that has been known by many names: Project Sanguine, Project Seafarer and Project ELF for Extremely Low Frequency. The project involved the construction of several 14-mile long single-wire antennas (with the far end grounded) that could transmit up to 760 amperes of 45 Hz power to communicate to underwater submarines as a secure communications facility. Because the antenna project would act like a continuously faulted power line, serious interference problems were anticipated on the telephone cables in the affected areas. Bell Labs successfully experimented with 20 INTs and effectively demonstrated that compatible operation of mitigated telephone plant exposed to a severe electromagnetic environment could be achieved.”
“Since the INT became commercially available in the early 70’s, many Bell System and Independent telephone operating companies, railroad companies and others, have standardized on the units and have written practices regarding their applications. In fact, I wrote the Bell System Practice (873-505-107) on the INT. The Rural Electrification Administration’s (REA) newly revised Section 451 on “Telephone Noise Measurements and Mitigation” in the Telephone Engineering and Construction Manual includes a discussion on the INT. The REA considers the INT as one of the primary means available for solving power line problems.”
“Another way to understand how an INT works is to establish an analogy to shielding theory. The grounded cable sheath can be considered the primary winding and the cable pairs are the secondary. The object of a perfect shield is to cause all the voltage induced on the shield or primary winding from the induced power line current flow to be equal, but opposite in polarity to the induced voltage on the secondary winding or working pairs generated from an opposite shield current flow. These voltages should cancel or neutralize one another. This works quite effectively for the higher noise frequencies, but it is not effective at 60 Hz due to the lack of a high mutual inductance in the cable. This impedance can be obtained only by using heavy, ferromagnetic materials spread out along the cable sheath or by lumping the iron in one or more locations with an INT.” Refer to paragraph 2.6 on page 11 of Engineering Report No. 26, “Shielding of Ground-Return Circuits at Low Frequencies,” Joint Subcommittee on Development and Research, Edison Electric Institute and Bell Telephone System, March 16, 1934)
“For electrical safety reasons, the recommended location of an INT is at the electrical center of the exposure, or where the steady-state induced voltage is one-half of what is measured at the end of the loop. Since roughly 90% of the total longitudinal voltage will be redistributed across the INT, this would assure that the voltage-to-ground on either side of the INT would be approximately one-half. For example, if 100 volts was measured at the end of the loop and the INT was located where 50 volts was measured, roughly 90 volts would appear across the INT, 10 volts would appear at the end of the cable and only 45 volts-to-ground would be measured on either side of the unit.” (Thus, this would meet the old Bell System personnel safety value of no more than 50 volts-to-ground anywhere along the cable route. Historically, of course, INTs were rarely ever placed for safety reasons! Let’s just hope this location was not in an area of wet cable, otherwise the noise on these pairs could have been increased substantially. This would have really been embarrassing, especially if after spending a lot of time and money on trying to solve a noise problem, one now finds himself with a greater noise problem!)
“It was previously mentioned that according to the 1971 Bell Labs noisy-loop survey, 90% of the electrical centers of exposure would be nearly 90% of the way out from the central office. Thus, it is reasonable to conclude that most INTs would normally be located near the end of the loop. However, power loads and power lines are not static, so it would be unrealistic to assume that the electrical center of exposure would always remain at one location. For this reason, then, it makes more sense to locate an INT for the best noise reduction possible (which, of course, is also subject to change, but may not be as rapid as the power loads) and then locate several INTs if the voltage needs to be brought down to safer levels under worst-case conditions.”
“Depending on which electrical safety criteria one adheres to, it may be necessary to engineer several INTs in tandem (series) with one another. The electrical exposure can be equally divided into separate exposures and INTs placed at the halfway points between the end grounds. The largest multiple INT installation of this type was recently completed for a railroad company (Philadelphia Airport High-Speed Rail), which placed 10 units, with roughly one every mile, to assure that under the worst-case condition there would never be more than 7.5 volts-to-ground (appearing anywhere in the circuit).”
“As seen from the previous discussion, an INT is a simple item of telephone plant that does not have to be regarded as a special or unusual application requiring extensive engineering. In fact, it has been proven time after time to give impressive results regardless of where it is installed.”
“Proper knowledge of the unit’s operation and selecting the right location are the key points to a successful INT installation. This is why it is recommended that a small unit be used as a test tool in searching out the best location for a permanent installation. This technique can be applied to larger cables, as long as it is remembered that the maximum effectiveness will result when the entire cable is fully treated, thereby eliminating the effects of secondary induction on the untreated pairs."
“The current standard sizes of INTs that are commercially available from the SNC Manufacturing Co. are 6, 12, 18, 25, 50 and 100 pair units ranging in weight from 25 to 162 pounds and in size from 7”x 8”x 6” to 20”x 13”x 10”. This is what was earlier referred to as putting the metal back into the plant. The 6 through 50 pair INTs are wound with approximately 500 feet of standard, even count, color-coded, PIC insulated, twisted and transposed 26 gauge cable core.”
“Since the largest individual INT SNC manufactures is a 100-pair unit, larger cables must be treated with multiple units in parallel operation. Successful multiple installations of up to 1200 pair are presently in operation as is a 2300 pair unit in a fiberglass shelter. This project also involves a 900-pair unit that will be part of a tandem installation in a resort area having a limestone foundation (see Telephony, August 11, 1980, page 74). Summer-time power loads have not only created a noisy environment, but there has been over 150 volts-to-ground induced on this particular cable route and an electronic central office conversion is scheduled for the near future.” (This was in Lake of the Ozarks, Missouri and received the attention of SWBT President Zane Barnes.)
“SNC also manufactures three special T-carrier INTs for treating single systems, 25 systems and 50 systems. The two latter INTs are wound with 24 gauge cable cores and have a special shield separating both directions of transmission to prevent any crosstalk or DC offset problems. These units have an insertion loss range from 2.2 dB to 6.0 dB depending on the frequency range from 0.772 MHz to 3.152 MHz and their location in the span line.”