Anechoic stands for 'An (none) + Echoic'. Literally speaking Anechoic Chamber is a type of chamber that does not echo. More specifically, it is a type of chamber that does not reflect any signal back to the measurement sensor placed within the chamber.

Let's see what that takes in practice. A chamber has to absorb the waves that hit its walls, keep outside signals out, and leave enough room between the measurement antenna and the device. I'll go through the absorbers first, then the reason for the chamber, the distance it needs and the types in use.

• How Does a Chamber Stop Reflections?
• Why we need a Anechoic Chamber ?
• How Far Must the Antenna Be From the DUT?
• Types of Anechoic chambers

How Does a Chamber Stop Reflections?

Every wall, floor and ceiling reflects part of a wave that hits it. A chamber cannot remove those surfaces, so it covers them with a material that turns the wave into heat instead of sending it back.

How to make it not reflect any signal to the measurement sensor ?  A typical method is to stick absorbers with a specific shape to everywhere that may reflect the signal as shown below. Those absorver usually has a shape of pyramid and often called as a cone. The cone is made up of the material that absorbs the signal/wave and the shape is designed in such a way that the signal is reflected (if not 100% absorved) in random or specific directions away from the measurement sensor (e.g, measurement antenna).

The material of the cone (abosorber) is selected depending on the nature of the signal (e.g, audio or electro-magnetic signal) and the dimension and shape of the cone is designed differently based on the characteristics and the frequency of the signal.

Image Source : Click here

The pyramid shape has a reason beyond scattering. A flat sheet of absorbing material would itself reflect part of the wave at its front face, because the wave meets an abrupt change of material. The tip of a pyramid presents almost no material to the wave, and the cross-section grows gradually toward the base. So the wave enters the absorber with little reflection and is absorbed on its way in. This taper works well only when the pyramid is long compared with the wavelength. That is why low frequencies need long absorbers, and why the absorber is chosen for the lowest frequency the chamber must handle.

A chamber also has a second, less visible layer. Behind the absorbers, the walls of an RF chamber are metal, so the room is a shielded enclosure. The shielding keeps outside signals, such as live cellular networks, from reaching the device. The absorbers then deal with the waves inside the enclosure.

• The cones in the diagram cover every inner surface : the cut-away view shows absorbers on the walls, the floor and the ceiling, all pointing inward.
• The pyramid is a gradual transition : the wave meets more absorbing material step by step, so little of it reflects at the front.
• Absorber length follows the wavelength : a chamber built for low frequencies needs longer absorbers than one built for mmWave.
• Absorbing and shielding are two separate jobs : the metal shell keeps outside signals out, and the absorbers stop reflections inside.

Why we need a Anechoic Chamber ?

Simply put, we use the Anechoic Chamber for the accuracy and repeatability of the measurement/test. If we measure a signal from a source (i.e, transmitter) in ordinary chamber or in open space with many reflecting objects, the measured value may vary because the signal reaching the DUT is summation of the original signal (i.e, line of sight signal) and many other versions of reflected signal. This kind of situation would make the measurement not repeatable (i.e, different values from every trial) and in some cases the reflected signal would interfere with the line of sight signal in negative way making the signal connection unstable.

It helps to put a number on the problem. Suppose one reflected wave reaches the measurement antenna at -20 dB relative to the direct wave. Depending on the path difference, the two add in phase or out of phase. The measured level then moves between +0.83 dB and -0.92 dB around the true value. A reflection at -10 dB moves it between +2.39 dB and -3.30 dB. A reflection at -30 dB, which a good chamber can reach, moves it only between +0.27 dB and -0.28 dB. Each of these follows from 20log₁₀(1 +/- r), where r is the amplitude ratio of the reflected wave to the direct wave.

The region inside the chamber where the remaining reflections are small enough is called the quiet zone. The device under test is placed there. The quiet zone matters for 5G, because FR2 devices are tested over the air. TS 38.101-2 states that the FR2 UE transmitter characteristics are specified over the air, OTA, unless otherwise stated. It also places the radiated requirements reference point at the centre of the quiet zone. So in FR2 the chamber is part of the definition of the requirement, not only a convenience.

• A reflection adds a ripple to every reading : the error is 20log₁₀(1 +/- r), and it changes with the position of the device.
• A -30 dB reflection keeps the error below about 0.3 dB : a -10 dB reflection can move the reading by more than 3 dB.
• The device sits in the quiet zone : that is the region where the chamber meets its reflection target.
• FR2 requirements are OTA requirements : TS 38.101-2 puts the reference point at the centre of the quiet zone, so an FR2 UE cannot be tested without a chamber.

How Far Must the Antenna Be From the DUT?

A chamber that absorbs every reflection still has one more condition to meet. The measurement antenna has to be far enough from the device that the wave arriving from the device is close to a plane wave. Otherwise the measured pattern depends on the distance, and the result is not the far-field pattern that the requirement means.

The usual rule for this distance is the far-field distance R = 2D²/λ. Here D is the largest dimension of the radiating antenna and λ is the wavelength. The distance grows with the square of the antenna size and in proportion to the frequency. So it is short at low frequencies and long in mmWave, as the table below shows.

The size D in the table has to be chosen with care. If the whole phone can radiate, D is close to the size of the phone. If only a small antenna module radiates, D is the size of that module, and the distance becomes much shorter, as the last row shows. At 28 GHz with D = 15 cm, a direct far-field chamber needs more than 4 m between the device and the measurement antenna, plus the absorbers behind both. That is a large room.

A compact antenna test range, CATR, solves this problem with a reflector. The feed antenna illuminates a curved reflector, and the reflector turns the spherical wave into a plane wave across the quiet zone. The chamber can then be much shorter than the far-field distance. Chamber (C) in the next section shows a reflector of this kind at its left end.

• The far-field distance is 2D²/λ : it grows with the square of the radiating size and with the frequency.
• D is the size of what radiates : a small antenna module needs a much shorter distance than a whole phone.
• mmWave pushes direct far-field chambers to several metres : at 28 GHz and D = 15 cm the distance is about 4.2 m.
• A CATR reflector creates a plane wave in a short chamber : it replaces distance with a precisely shaped reflector.

Types of Anechoic chambers

There are many different types and sizes of anechoic chambers. Some of the examples are shown below. There is a huge chambers like (A) so that you can put a whole vehicle there. Usually chambers for audio testing (B) is usually pretty large because of the wave lengh of audio is pretty long (i.e, frequency is low).  Usually the chambers (C, D) being used for testing mobile phone in RF/mmWave range is relatively small comparing to (A), (B).

• (A) is a vehicle chamber : a car sits on a turntable under a large arch that carries a ring of measurement antennas, and the whole room is lined with absorbers.
• (B) is an acoustic chamber : a head and torso simulator faces a loudspeaker, and the walls carry large absorbing wedges.
• (C) is a cut-away of an RF chamber : it shows a curved reflector at the left end and a device positioner inside the absorber lining.
• (D) is a compact RF chamber with its door open : absorbers cover the inside of the door and the walls, and a positioner holds the device in the middle.

The wavelength explains most of the size differences. Sound in air travels at about 343 m/s, so a 100 Hz tone has a wavelength of 3.43 m and a 1 kHz tone has 0.343 m. An RF wave at 700 MHz has a wavelength of 0.43 m, and at 28 GHz it is only 10.7 mm. So an acoustic chamber like (B) needs very deep wedges to work at low audio frequencies. An RF chamber for mmWave can use short absorbers, and its size is set more by the far-field distance and the positioner than by the absorbers.

Chambers also differ in which surfaces are covered. In a fully anechoic chamber, all six surfaces carry absorbers, so the device sees free space in every direction. In a semi-anechoic chamber, the floor stays reflective. That setup is common for vehicle and EMC testing, where a ground plane is part of the real situation.

• Chamber size follows the wavelength and the measurement distance : low frequencies need deep absorbers, and large devices need long distances.
• Acoustic chambers are large because audio wavelengths are long : 100 Hz in air is 3.43 m, while 28 GHz RF is 10.7 mm.
• A semi-anechoic chamber keeps a reflective floor : it reproduces a ground plane, for example under a vehicle.