How a Semi-Anechoic Chamber Stacks Up With an OATS

With the recent increase in open area test site (OATS) testing required by European and U.S. markets and the Bellcore telecommunications testing standards, the problems associated with OATS once again are a major topic of conversation among EMC professionals. These problems are well documented and include but are not limited to:

High ambients in a metropolitan area where most customers are based.

Field uniformity.

Test time lost in dismantling setups in the chamber to take equipment under test (EUT) to the OATS.

The requirement for prescans in a chamber.

The construction cost of the site.

There are many advantages for testing in a semi-anechoic chamber as well as methods for correlating emissions measurements at an OATS with those in a 3-meter semi-anechoic chamber. The OATS must meet the 10-meter site attenuation, and the chamber must meet the site attenuation for a 3-meter alternate test site per ANSI C63.4. There are situations, however, where a 10-meter OATS is required.

Customer Requirements

National Technical Systems (NTS) recently decided to expand its existing EMC/EMI test capabilities as a result of customer requirements and the downturn in traditional military/aerospace testing. We wanted to replace existing OATS in suburban Boston, Los Angeles, and New York City, which are convenient for clients but located in metropolitan electromagnetic ambient environments.

It was determined that a 3-meter semi-anechoic chamber and a simple, but functional, 10-meter OATS was the best solution. The cost for constructing and maintaining a fully climate-controlled and automated OATS while addressing the ambient issue is about the same as constructing a 3-meter semi-anechoic chamber with a functional OATS. The advantages are numerous, and the response from our client base was positive.

Many times, customers complained about ambient problems and the time-consuming task of tearing down the equipment to reassemble it at the OATS. And, with the European Notified Body placing more faith in the semi-anechoic measurement than the OATS and the telecommunications requirements specified in both 3- and 10-meter limits, the semi-anechoic chamber was definitely the most cost-effective facility.

Since radiated immunity also would be performed in the chamber, a small penalty had to be paid: The floor of the chamber needed ferrite material to meet the EN 61000-4-3 field uniformity requirement. This problem was solved by mounting the ferrite on plywood panels with steel backing that were placed on the floor between the transmit antenna and the turntable. This change-out of panels can be performed in less than 15 minutes.

Customers prefer the test technician removing a few ferrite panels rather than tearing down the EUT and moving it to an outside location. This is especially true for telecommunications customers who typically have special test-support equipment. Most of these test sets are built to exercise the telecom unit. They are not designed to be moved around and do not travel well.

We had considered placing the OATS in a secluded area far from a client’s base. This option was rejected because of travel costs and weather-related concerns. Instead, construction consisted of an OATS with a climate-controlled turntable and support equipment complementary to the semi-anechoic chamber for special cases where the OATS was necessary.

Exceptions

At this point, you might conclude that the OATS is an obsolete test facility that soon will follow the dinosaurs into history. Although we would not mourn the death of the OATS, its demise is far from eminent.

Until the cost of 10-meter chambers becomes cost-effective, OATS will be a necessary evil. In particular, when a 3-meter measurement is very close to the limit (1 to 4 dB), the OATS will be the definitive measurement. This does not mean that the chamber measurement is not correct. In fact, the chamber measurement will be identifiable and accurate. But the OATS measurement may require the antenna to be moved closer than 10 meters (3 meters) if the signal is lost in the ambient.

Many years ago, international committees such as IEC, ANSI, and CISPR dictated that the OATS measurement was the standard. However, recent developments in ferrite technology have made 3-meter chambers quite cost-effective. Even the 10-meter chamber now is viable for companies with deep pockets, and a few are presently in operation.

As the technology moves further in this direction, the OATS eventually will be retired. Until that time, an OATS is necessary for larger equipment where the antenna distance of 3 meters is not practical and the units may be too large to fit in a 3-meter chamber.

Obviously, this argument can be taken further for units that are too large for OATS. This is when engineering judgment and technical analysis and reports such as Technical Construction Files are required.

With the signing of the Mutual Recognition Agreement between the United States and the European Union in December 1998, U.S. laboratories are allowed status of Competent Bodies to make these judgments. The assignment of U.S. laboratories as Conformity Assessment Bodies (CAB) was received in late March, and most major laboratories are applying for recognition.

Correlation

Measurements in NTS semi-anechoic chambers have been performed at Boxborough, MA, Tinton Falls, NJ, and Fullerton, CA. Testing was performed in accordance with ANSI C63.4 from 30 MHz to 1 GHz for alternative test sites.

This requires the Normalized Site Attenuation (NSA) to be performed at five positions around the turntable (essentially the four corners and the standard center position). In addition, the measurement must be performed with the antenna placed at 1.5- and 2-meter heights. As a result, 20 measurement positions are required: five positions for vertical and horizontal with the antenna at 1 meter, five positions with the antenna at 1.5 meters in the vertical position, and five measurements with the antenna at 2 meters in the horizontal position.

The chamber shown in Figure 1 is a hybrid containing both anechoic absorbers for high-frequency performance and ferrite tiles for low-frequency performance. The ferrite tiles are not visible since the anechoic materials cover them.

Once the NSA is performed in the chamber and complies with the requirements of ANSI C63.4, testing to the telecom and FCC requirements can begin immediately. The chamber also can be used for emissions tests in compliance with MIL-STD-462D and RTCA DO-160.

The only issue to be considered is testing to the CE Marking in accordance with EN 55022 or other product specifications for the consumer market. The European requirements specify a 10-meter emissions measurement. For the Notified Body to accept data at 3 meters for the 10-meter requirement, a correlation with an OATS that has demonstrated compliance to the ANSI C63.4 NSA is required. All three NTS EMC/EMI test sites have OATS listed with the FCC, which requires the site to demonstrate compliance to ANSI C63.4.

To satisfy the concern of the Notified Body, a calibrated broadband noise generator was used to identify emissions. The generator was placed in the OATS at a distance of 10 meters from the receiving antenna as described in EN 55022. Measurements were performed at the bandwidths using a procedure that complied with EN 55022. This data was stored and compared to the class A and B limits.

The generator was taken into the chamber, and the emissions measurement was repeated at 3 meters. Then the data from the chamber was compared to the class A and B limits adjusted by the 10/3-meter ratio (approximately 10 dB).

The generator can be adjusted to emphasize any specific frequency in the 30-MHz to 1-GHz frequency range if the data shows any possible trouble frequencies. Of course, the OATS measurement near high-ambient frequencies requires special consideration. The chamber ambient is well below the EN 55022 limits.

The results are shown in Figure 2. The OATS and the chamber correlate since the uncertainty in the measurement is above 2 dB. The data did not need correction due to perturbation of the ground plane. This was somewhat surprising since we expected a larger difference in the data. More studies such as antenna height, polarization, and off-axis measurements will be performed to determine when or if this perturbation becomes significant.

Conclusion

After reviewing customer requirements and state-of-the-art technology, we concluded that a 3-meter semi-anechoic chamber along with a functional OATS is the most cost-effective means of meeting client needs. The majority of consumer products can be tested in the chamber for all tests without either dismantling the unit or taking the device outdoors. This also will be true for telecom equipment contained in one rack.

The semi-anechoic chamber is correlated to an approved 10-meter OATS and meets the requirement of ANSI C63.4. The measurements in the chamber are repeatable and made without ambient difficulties. We believe the trend to indoor measurements will continue.

About the Authors

James Press is the chief scientist responsible for NTS EMC technical issues and developments. He holds a B.A. from Temple University and an M.S. from Iowa State University, both in physics. Mr. Press, who has been working in the EMC field for more than 15 years, has published many articles and a handbook on EMC test and design. National Technical Systems, 1146 Massachusetts Ave., Boxborough, MA 01719, (978) 266-1001, e-mail: [email protected].

Mark Betts is an EMC engineer at the NTS New Jersey facility. Previously, he was employed by the Netherlands Measurement Institute, a Competent Body in Holland, and held certification as an EMC test witness engineer and Technical Construction File assessor. National Technical Systems, 36 Gilbert St. South, Tinton Falls, NJ 07701, (732) 936-0800, e-mail: [email protected].

John Ngo is the compliance manager at NTS-Fullerton. He received a B.S.E.E. from Rochester Institute of Technology. Mr. Ngo has more than seven years experience in EMC design and testing and three years experience in the field of product safety. National Technical Systems, 1536 E. Valencia Dr., Fullerton, CA 92831, (714) 879-6110, e-mail: [email protected].

FIGURE 2

Frequency (MHz)	OATS (dBµV)	Chamber Adjusted Data (dBµV)	Deviation
30	47.85	46.58	1.3
40	39.44	40.6	1.2
50	39.4	40.3	0.9
100	28.47	29.96	1.5
200	25.44	26.13	0.7
300	46.39	46.84	0.45
400	47.37	46.25	1.1
500	47.42	46.84	0.6
600	46.79	47.27	0.5
700	42.5	43.08	0.6
800	46.84	46.39	0.5
900	41.68	43.33	1.7
1,000	42.61	43.0	0.4