FMS Publications

Management of Power Frequency Magnetic Fields from Electrical Conduits

INTRODUCTION

Public concern about potential health effects of ELF magnetic fields has generated interest in reducing field strengths in commercial buildings from sources such as transformers, switchgear, bus bars and conduits to substantially lower values than are typically achieved with conventional shielding schemes. Such traditional methods typically reduce field strengths into the 10 milliGauss (mG) range.

Power frequency magnetic fields are extremely difficult and expensive to shield to values of less than 10 mG particularly when the fields originate from conduit sources which are inaccessible or have net current conditions. This study evaluates methods substantiated with extensive laboratory test data, which suggests that carefully combining active and passive shielding techniques can produce results which are superior and more cost effective than either technique when used alone and can achieve in the range of 2-3 mG.

OBJECTIVE

Circuits in conduits passing through commercial buildings can exhibit high magnetic fields in the 10's of mG due to the presence of net current conditions and/or greatly varying shielding effectiveness provided by various conduit types.

The objective of this study was (1) to characterize magnetic fields from single and three-phase multiple conductor circuits typically found in commercial building conduits, including the characterization of fields from circuits with varying degrees of unbalance, conductor separation and net current conditions, and (2) to evaluate and document the effectiveness of a large range of field reduction strategies utilizing combinations of commercially available types of conduits, passive shielding schemes using flat and various forms of ferrous and conductive materials, and the use of active and passive induced current cancellation techniques.

METHOD

A special test system consisting of several large moveable racks used to suspend lengths of conduit approximately eight feet from the floor was constructed in the EMF Laboratory. A 16ft x 20ft measurement grid was marked on the floor to simulate an office or classroom above and below the test conduits. Magnetic field readings were taken at all grid locations for each test case, at 30 inches and 60 inches above and below the test conduit location. Ambient magnetic fields in the laboratory ranged from 0.2 to 1.5 mG.

Multi-conductor circuits, in two test conduits, were connected to a three-phase power network, wherein three single-phase variable transformers were fed by a 4-wire, three-phase 208 volt source. The transformers were capable of supplying variable voltages from 0 to 140 volts to a set of three loading transformers which were configured for 120 volt input to 5 volt output. The output from each loading transformer was connected to parallel 4-wire circuits (using #4/0 cable) in each conduit. The four conductors contained in each of the test conduits were connected together as the conductors exited to creat three-phase current circulation. This arrangement allowed three-phase and single-phase currents at values ranging from 100 to 1,000 amps, to be circulated through the conduits, thereby generating levels of magnetic fields in the area of the measurement grid.

Circuit unbalance was created by different phase current settings and the use of dissimilar wire sizes for the neutral conductors in each of the two conduit circuits. Net current conditions were also created by connecting an external wire loop around the perimeter of the laboratory at some distance from the measurement grid location. This loop was connected to the neutral bus of the loading transformers and to the connection point of all conductors as they exited the two test conduits. A choke was employed to "force" a desired amount of the neutral current to circulate in the external loop, thereby creating a net current condition in the test conduit circuits.

Design of the test fixtures which suspended the test conduits above the measurement grid, also facilitated the temporary placement of flat plates and various forms of conductive and ferromagnetic shields around, above and below the conduits to evaluate shielding effectiveness. For active cancellation tests, a current sensor was placed around the test conduits to measure net current. The output of the sensor fed an active feedback network which drove cancellation conductors placed adjacent to the test conduits with a cancellation signal of current. For induced cancellation tests, a "loop" of #4/0 cable was placed inside each test conduit adjacent to the conductors.

RESULTS

Ferrous and conductive conduits as well as certain shielding forms constructed from ferrous and conductive materials varied widely in magnetic field reduction effectiveness depending on material form, thickness and distance from the circuits. These have been tabulated and ordered in effectiveness for circuit current, unbalance and conductor separation conditions. One hundred twenty-seven test case configurations were documented during the testing process.

Circuits with net current conditions had higher magnetic field strengths present regardless of conduit type or passive shield material, placement or form. All ferrous and conductive shielding schemes did not reduce magnetic fields from conduits with net current conditions.

The use of an actively driven cancellation loop placed adjacent to the conduits was highly effective in reducing field levels from single and three-phase circuits with net current conditions provided that the circuits were contained in steel or aluminum conduit. Active cancellation was not effective at reducing magnetic field levels from circuits with net current conditions when contained in PVC or other non-metallic conduit.

Induced cancellation of net current utilizing a passive conductor loop proved generally effective in reducing the field level by approximately 50% in single-phase circuit conduits but was generally ineffective for three-phase circuits. The combination of active cancellation and certain shielding schemes proved highly effective in reducing fields to low levels of 2 to 3 mG from circuits which contain net current, circuit imbalance and conductor separation.

Test results suggest that the combination of active cancellation and passive shields may offer superior magnetic field reduction advantages over any of the individual shielding schemes. Lower values may also be achievable without consequential increases in cost or technical difficulty.

STUDY SITE

After completion of testing, a commercial building was selected as a study site for installation of a laboratory developed shielding scheme. At the selected site, ELF magnetic fields at elevated levels sufficient to distort color computer monitors were present in offices located directly over buried three-phase circuits in non-metallic conduit traveling from an exterior utility pad-mounted transformer to the building's main service panel location.

As the conduits were buried under the concrete floor slab and were no accessible, a shielding scheme consisting of one layer of conductive aluminum was installed directly on the floor area of the affected offices and a small portion of adjoining walls. Based on laboratory data, it was estimated that the average B-field, after installation of shielding, would be less than 4 mG.

CONCLUSION

The test data from this study suggests that it is possible to significantly reduce ELF magnetic field levels from commercial building power circuits to values approaching 2 to 3 mG by the careful selection of conduit type and/or use of sufficiently large planes or forms of ferromagnetic and conductive shielding material. Given the structural limitations of many sites, the use of shielding schemes may not be economically or physically feasible. Attention to the selection of conduit type, installation practices ad circuit configurations at the time of construction are the best options for managing magnetic fields from commercial building electrical circuits.

Elevated magnetic fields from conduit circuits which have net current conditions are very difficult to shield and are best reduced by attempting to electrically minimize the presence of net current. The use of ferromagnetic and conductive conduits and/or shielding schemes are generally ineffective at reducing magnetic fields from circuits with net current conditions. However, the combination of an active loop cancellation system with passive shielding material appears to have the greatest potential for providing an alternative economic field management scheme for reducing ELF magnetic field strengths to a 2 to 3 mG range.

The use of passive induced cancellation "loops" also demonstrated varying degrees of effectiveness in reducing magnetic fields from net current conditions. Future studies of reducing magnetic fields in conduits with net current conditions will include further testing and evaluation of active/passive cancellation combinations.