Construction of a 10-m Semianechoic Chamber

Unlike a 3-m chamber, a 10-m chamber provides the far-field measurements that are required by many regulatory agencies.

As electromagnetic pollution worsens, measuring electromagnetic interference at open-area test sites (OATS) has become more difficult, especially in metropolitan areas. To avoid high ambient noise levels, using a shielded room has provided the solution to this problem for many test facilities.

When the European Union began enforcing electromagnetic compatibility (EMC) requirements for electric and electronic products in January 1996, it included requirements for both emissions and immunity testing. For emissions measurements, a shielded room must meet the normalized site attenuation (NSA) requirements per ANSI C63.4:1992, CISPR 16-1:1999, or EN 50147-2:1996.^1–3 Immunity testing must be performed in a shielded room that meets the uniform field requirement specified in EN 61000-4-3.⁴

A simple shielded room, however, cannot meet both NSA and uniform field requirements. A semianechoic chamber is the most popular commercially available chamber for performing both emissions and immunity testing. Anechoic chambers are shielded rooms fully lined with radio frequency (RF) absorber on the ceiling, the walls, and the floor.

The size of a shielded room influences its performance. For immunity testing, EN 61000-4-3 defines a preferred measurement distance of 3 m between the antenna and the equipment under test (EUT). For emissions testing, the Federal Communications Commission specifies the measurement distance at 3 m for Part 15 Class B devices and 10 m for Part 15 Class A devices. CISPR 22, which is the basic information technology equipment (ITE) standard referenced by most regulatory agencies, specifies a measurement distance at 10 m for both Class B and Class A devices.⁵

Commercially available 3-m anechoic chambers are typically available in two different sizes: 8 x 4 x 3 m (24 x 12 x 10 ft) and 9 x 6 x 6 m (30 x 20 x 20 ft). An 8 x 4 x 3-m semianechoic chamber is used primarily for immunity testing. The internal dimensions of the chamber are too small to raise an antenna to the 4-m height required for emissions testing. A well-designed 9 x 6 x 6-m semianechoic chamber meets both NSA and uniform field requirements and can perform emissions testing at 3 m as well as RF immunity testing. However, some regulatory agencies do not accept test data taken at a 3-m test site because it fails to meet the far-field requirement. A 10-m semianechoic chamber satisfies this requirement.

This article describes the construction of a 10-m semianechoic chamber located in Duluth, GA. The dimensions mentioned or listed in this article are for this particular chamber. Actual dimensions depend on the design. The construction method described here is one of many methods used. This article should help those involved with EMC regulatory requirements understand the method and benefits of constructing this type of 10-m semianechoic chamber. Chamber construction materials chosen for one installation may not be appropriate for a chamber in another location. For example, Masonite and particleboard are not adequate for a high-humidity area. And, in an earthquake zone, the structure may have particular reinforcement design requirements.

Site Preparation

Finding a building site for a 10-m chamber is the first crucial step. Existing office space often lacks the necessary height (about 30 ft or 9 m). Care must also be taken when constructing a 10-m chamber on newly developed or filled-in land because the land may settle, which can change the shape and characteristics of the chamber.

The size of this 10-m chamber in feet is 48W x 60L x 27H (14.4 x 18 x 8.1 m). The size of the building housing is 60W x 80L x 34H (18 x 24 x 10.2 m), which accommodates the chamber, the control room, and the surrounding walkway (see Figure 1). It is important to determine the size of the turntable before pouring the concrete. The construction of the foundation (slab) and the structural steel installed to support the chamber followed the local building code.

Figure 1. The building housing (60W x 80L) accommodated the chamber, the control room, and the surrounding walkway.

Steel Enclosure

A modular construction method was used for the steel enclosure because modules are easy to transport, assemble, and disassemble. The 200-lb RF panels (4 x 10 ft x 13/16 in.) were constructed of particleboard and then sandwiched between 26-gauge galvanized steel plates. A metal frame strip connected each of the RF panels. The electrical continuity was assured by placing screws every 4 in., which were tightened with a torque wrench.

Floor. To prevent moisture, a sheet of 6-mil plastic was placed on top of the concrete slab, and 1/8-in.-(0.32-cm)-thick Masonite was placed over the plastic. To construct the chamber floor, two pieces of 1-ft-wide Masonite were placed beneath each RF panel to act as spacers (see Figure 2).

Figure 2. Masonite was placed beneath the RF panels to act as spacers.

The finished raised floor was 16 in. above the foundation, which included the RF shielded floor. This dimension varies based on the height of the turntable. The raised floor is supported by pedestals spaced 1 ft apart in heavy traffic areas (from the double door to the turntable), and then spaced every 2 ft in light traffic areas. The raised floor was leveled by using a rotary laser and adjusting the mounting flanges of the pedestals so that all panels would be flush with the turntable. The height difference between the metal frame strip and RF panel of the raised floor was covered by 1/8-in.-thick vinyl tiles.

Walls and Ceiling. The walls and ceiling were constructed with RF panels and frame strips. Inside the chamber, Masonite was installed on top of the RF panels to provide a smooth surface for the RF absorber.

Doors. The door plays an important role in chamber characteristics. Common door design and construction uses a recessed contact mechanism. A raised edge (edge extrusion) on the door rests in a slot (female extrusion) on the frame. The slot is lined with a gasket and copper finger stock material to maintain electrical continuity.

Two swing doors provided the access to the chamber. The dimensions of a single door were 4 x 7 ft (1.2 x 2.1 m), and the dimensions of a double door were 12 x 12 ft (3.6 x 3.6 m). The single door, which used three hinges, was designed so that test personnel could access the chamber from the control room. The double door was designed for placing the EUT in the chamber. Eight 5-in. hinges connected the double door to the frame. Each of the two double doors weighed approximately 800 lb (362 kg). Careful packaging of the doors is important to avoid shipping damage.

Chamber Grounding. Grounding a chamber is much simpler than grounding an OATS. For this chamber, a single-point ground was used to avoid ground loops, and a number-6– gauge copper wire was attached to a ground rod. To maintain the single-point ground, a nonconductive conduit (polyvinyl chloride) was used for routing the electrical wires outside the chamber.

RF Absorber

The chamber was lined with a combination of ferrite grid and foam absorber to provide maximum absorption. The ferrite grid absorbs electromagnetic energy at low frequencies up to 2 GHz.⁶ The standard size of the ferrite grid is 10 x 10 x 1.9 cm (4 x 4 x 0.75 in.). Nine pieces of ferrite grid were glued to a 30 x 30-cm (1 x 1-ft) single metal plate for easy installation.

These photos illustrate various stages of the chamber construction (clockwise from top left): a raised RF shielded floor; a dielectric boot to isolate the structural ground; ferrite grids to prevent reflection; and a turntable supported by a table ring.

High carbon–loaded foam absorber absorbs electromagnetic energy at higher frequencies. The absorbed frequency range of the pyramidal-shaped absorber depends on its length. The length of the foam absorber is typically the quarter wavelength of the absorbed frequency. At a frequency of 30 MHz, the wavelength is 10 m, which would require the foam absorber to be 2.5 m. This length, however, consumes much of the working space otherwise available in the chamber. Therefore, the pyramidal absorbers are often truncated in length with minimal degradation of performance.

Combining ferrite grid and foam absorber provides both low and high absorption. RF absorber was installed between the EUT and the source or measurement antennas to minimize reflection. Areas of likely reflection were identified using computer software simulation.

Using screws through the metal backing plate of the ferrite panels, installers mounted ferrite grid on all inner surfaces except the floor. Using Velcro, they attached foam absorber on top of the ferrite grid on the walls only. A safety net was secured near the ceiling to prevent injury to test personnel in case tiles fall.

No RF absorber was placed on the floor, so that the chamber would meet the conducting ground plane requirement for radiated emissions testing. Ferrites were placed between the turntable and antenna to maintain the uniform field required for radiated immunity testing.

Turntable

The size of the turntable affects the performance of the chamber. The larger the turntable, the more difficult it becomes to maintain the electrical continuity of the chamber. The diameter of the turntable for this chamber was 3 m.

The turntable, which was supported by a table ring and 12 wheels with a radius of 10 cm (4 in.), was driven by a three-phase inverter-duty ac induction motor (see photos above). The turntable can support up to 4000 kg (8800 lb). To maintain electrical continuity, a gasket was installed on the edge of the turntable, beneath the tabletop.

Electric and control wires were fed beneath the raised floor from the electrical control panel and control room, respectively. Cables for the EUT support equipment were fed under the turntable to the chamber exterior.

Power System

The chamber was provided with the following power sources:

• 120 V (1-phase, 60 Hz).
• 208 V (3-phase, 60 Hz).
• 230 V (1-phase, 50 Hz).
• 240 V (3-phase, 60 Hz).
• 277 V (1-phase, 60 Hz).
• 400 V (3-phase, 50 Hz).
• 480 V (3-phase, 60 Hz).

Two 60-Hz power sources (277 V and 480 V) were provided by the power company; 120, 240, and 208 V were stepped down from 277 or 480 V. The 50-Hz power sources were supplied from a generator (480 V, 60-Hz input). A power line filter was fitted onto each power line at its entrance to the chamber. A control panel housed the power sources for the turntable.

Lighting and Air Conditioning

Because fluorescent lighting produces electromagnetic noise from switching and aging, 10 high-hat metal halide lamps (400 W) were mounted in the ceiling of the chamber. Hi-bay lights (mercury vapor lamps) rated at 277 V were installed outside the chamber.

Ten ducts provided air conditioning to the room, and a valve was associated with each opening to control the return airflow. The ducts were attached to honeycomb vents mounted in the high-hats. A dielectric boot on each duct isolated the structural ground from the chamber ground to maintain single-point grounding for the chamber.

Fire Protection System

A fire protection system not only protects life and property, but it also reduces the insurance premium on a chamber. Smoke sensors and detectors for this chamber send a signal to a control box (located outside the chamber, control room, and amplifier room) to open a magnetic-operated solenoid device. Cylinder canister bottles then open to put out a fire. An operator can also push a red fire button to open the solenoid. The fire protection system used a cross-zone design, which means it takes at least two alarm signals (from two smoke or duct detectors) before discharging the canister. The fire protection system was divided into six groups: four for the chamber, one for under the floor, and one for both the control room and amplifier room. In case of a false alarm, an abort switch was provided so that the discharge could be terminated.

To extinguish fire quickly, four gooseneck nozzles were installed beneath the raised floor, as well as six on two walls and four in the ceiling. Sixteen nozzles were installed outside the chamber. To handle the compressed gas (Inergen) pressure, all piping was grade A-53B or A-106 (seamless). Nine smoke sensors were installed beneath the raised floor, and two duct detectors were located in the ceiling of the chamber. Two ceiling-mount smoke detectors were installed in the control rooms, and two in the amplifier room.

Chamber Validation

Normalized site attenuation was used to evaluate the quality of the chamber as a radiated emissions test site. Although the measured NSA must be within ±4 dB of the theoretical NSA for an ideal site, ±3.5 dB is typically the specification required by most test laboratories and therefore imposed on manufacturers. The larger the chamber, the easier it is to meet the ±4-dB requirement.

For a radiated emissions test site other than an OATS, 20 separate site attenuation measurements are required (horizontal and vertical antenna polarizations; center, left, right, front, and rear positions; and 1-, 1.5-, and 2-m antenna heights).^1–3

Normalized Site Attenuation. For this chamber, NSA measurement began at 1 m from the tip of the foam absorber located on the rear wall to the transmitting antenna located on the turntable. To minimize reflections, the receiving antenna was located 12°–22.5° off the centerline. One advantage of such placement is that a second antenna or antenna mast can be located on the other side of the centerline to minimize antenna switching.

Field Calibration. Field calibration was performed to ensure a uniform field for RF-radiated immunity testing per EN 61000-4-3. Calibration was performed on a 1.5 x 1.5-m hypothetical vertical plane at 0.5-m intervals (16 points to- tal). Ferrite grids were placed on the floor between the field sensor and the field-generating antenna to prevent reflection (see photos). The minimum required distance between the field sensor and the antenna is 1 m, although a 3-m distance is preferred. The tolerance for a uniform field was within –0 to +6 dB of the nominal value to ensure that the field strength did not fall below the required level.

Monitoring System

Two closed-circuit TV cameras were installed in the upper corners of the chamber. The cameras provide 480-line horizontal resolution, and the pan-tilt mechanism allows 360° panning. The video signal travels through fiber-optic cables to the fiber-optic transceiver located in the control room, then to the system controller and monitor. The system controller controls the camera's panning, zooming, focusing, and exposure control.

Conclusion

EMC testing for emissions and immunity requirements is complicated and requires that the test site provide certain characteristics in order to obtain the desired measurements. Commercially available 3-m semianechoic chambers are often used for immunity testing. For emissions testing, certain regulatory agencies do not accept test data derived from 3-m semianechoic chambers. Accurate measurements are difficult to obtain from OATS because of their high ambient noise levels. A 10-m chamber, such as the one described in this article, is designed to meet the far-field measurements required by many regulatory agencies. Understanding the construction should help those involved with EMC regulatory requirements determine when such a chamber is appropriate for testing.

Acknowledgment

The authors wish to thank the following people: David Dennis, Gary Flom, and David Schramm of Intertek Testing Services NA Inc.; Ron Harrison of EWM Electric Co.; Jerry Hock of Braden Shielding Systems; Randy Sizemore of Columbus Fire and Safety (F&S) Equipment Company, Inc.; and Ed Snow of Grinnell Fire Protection Systems. Special thanks to Roland Gubisch, PhD, of Intertek Testing Services and Tim Walters of Arris Interactive.

References

1. ANSI C63.4:1992, "Methods of Measurement of Radio-Noise Emissions from Low-Voltage Electrical and Electronic Equipment in the Range of 9 kHz to 40 GHz," American National Standards Institute, New York.

2. CISPR 16-1:1999, "Specification for Radio Disturbance and Immunity Measuring Apparatus and Methods, Part 1: Radio Disturbance and Immunity Measuring Apparatus," International Electrotechnical Commission (IEC), Geneva.

3. EN 50147-2:1996, "Anechoic Chambers, Part 2. Alternative Test Site Suitability with Respect to Site Attenuation," European Committee for Electrotechnical Standardization (CENELEC), Brussels.

4. EN 61000-4-3:1996, "Electromagnetic Compatibility (EMC), Part 4. Testing and Measurement Techniques, Section 3. Radiated, Radio-Frequency, Electromagnetic Field Immunity Test," CENELEC.

5. CISPR 22:1997, "Information Technology Equipment—Radio Disturbance Characteristics—Limits and Methods of Measurement," IEC.

6. Jens Haala and Werner Wiesbeck, "Upgrade of Foam Equipped Semi Anechoic Chambers to Fully Anechoic Chambers by the Use of Ferrite Tiles," 1998 IEEE International Symposium on EMC, Symposium Record, 14–19.

Grace Lin and Michael J. Alvarado are senior project engineers at Intertek Testing Services NA Inc. (Duluth, GA). They can be reached at glin@itsqs.com and malvarado@itsqs.com, respectively.

Back to September/October Table of Contents