The rapidly expanding wireless industry is extremely competitive in terms of price, features, and quality. The response to such pressures requires rapid development of new models, steep ramp up to volume production, and a large degree of confidence that the product will be fully functional at the point of use. High-quality final test of completed wireless products is essential to meet these demands.
A typical system to evaluate wireless devices includes signal sources and measurement instruments, system and DUT power systems, resource switching, internal system and system-to-DUT communications, test-system verification and diagnostics, a system controller, test executive software, and a DUT fixture. The fixture comprises an enclosure, a nest, the physical connection from the test system to the DUT, and fixture electronics (Figure 1).
The Enclosure
Isolation
Wireless-device development and production facilities typically are high RF noise environments. RF noise is generated from within the facility as well as from outside sources. The testing of wireless RF devices requires them to be isolated from the local RF environment to ensure that RF noise doesn’t negatively influence the tests. Adhering to regulatory requirements and not wishing to interfere with adjacent test cells mandate limiting RF noise generation into the environment as a testing byproduct.
Adjacent test-cell isolation is additive. For example, if adjacent test cells A and B each offer 60-dB RF isolation, the total isolation from DUT A to DUT B is 120 dB. Isolation of 40 dB in the frequency range of interest is considered minimal, and 60 dB is a more common figure of merit. In the case of RF isolation, more is better since test repeatability and accuracy increase as RF noise decreases.
Enclosure Types
Three common types of enclosures are the drawer, the door, and the clamshell. The drawer enclosure resembles a single-drawer file cabinet where the drawer pulls open for DUT access. The door enclosure is similar to a closet where the door opens for DUT access. The clamshell is like a box that has been cut at some distance from the top so there remains a hollow base with a hollow top hinged to it. The top opens for DUT access.
In each case, the fixture is closed for testing. Enclosure opening and closing and DUT placement can be manual, automated, or some combination of the two.
Isolation Verification
The design and development of an RF isolation enclosure must address the measurement of isolation effectiveness. To have confidence in the testing procedure, consistent and repeatable measurements are essential.
The RF energy-generation and measurement equipment must have sufficient bandwidth for the frequency range of interest. To measure all sides of the fixture, an RF anechoic chamber with a rotating table for the fixture is desirable.
The most common testing method generates RF energy within the enclosure and senses any leakage outside the enclosure. RF energy generation outside the enclosure with internal leakage sensing is much less common because it is more difficult to generate a consistent RF field of sufficient intensity in the larger volume of the anechoic chamber. Testing can be enhanced by comparing the results of both types of measurements.
Enclosure Design
The key design goal of an RF enclosure is to manage the openings in the enclosure itself since any opening can allow RF energy to pass through. These openings include fabrication seams, functional seams, and holes for connectors and hardware.
Some ways to manage openings are RF gaskets, welding, brazing, wave-guide geometries, and electrical filtering. Seams that never open may be managed by brazing or welding. Holes for connector attachment and seams required for access can use RF gaskets (Figure 2).
Figure 2. Typical Installation of Compressible RF Gasket
Since the choice and use of RF gaskets are not necessarily straightforward, the answers to some basic questions can offer guidance.
- Is the gasket under constant compression or repeatedly compressed and relaxed?
- What is the gasket sealing pressure during use?
- Is there an abrasive or wiping action as the gasket-sealed seam opens and closes?
- Is there metallurgical compatibility of the gasket and the materials to be sealed?
- What are the effects of anticipated temperature cycling, humidity cycling, and corrosive atmospheres on the gasket and the seams or holes?
Cables and connectors usually attach the DUT to the rest of the test system and most often enter the enclosure directly. Some methods for increasing inbound and outbound RF isolation for cables and connectors include waveguides, filtered connectors, and RF gaskets around connector shells.
Internal Control of RF and Audio Energies
When testing an RF-generating device in an enclosure, the RF energy can reflect one or more times from the walls within the enclosure. Those reflections, differing in phase from one another, generate RF noise which may be frequency-dependent and can impede testing as much as RF noise leakage into the enclosure. Some methods for controlling RF reflections include the following:
- Using materials inside the enclosure that don’t reflect RF energy.
- Keeping internal geometries simple, such as at 90° or 180° to each other, to allow some predictability of reflection angles.
Electronics internal to the fixture also can add to general RF noise. If internal fixture electronics are anticipated, a double-cavity enclosure can be used where the fixture electronics reside in a secondary cavity and the DUT is tested in the main cavity (Figure 1).
Since DUT testing protocols may include audio performance, the enclosure also should control inbound, outbound, and internally reflected audio noise. Similar to making the enclosure anechoic by using RF absorbing materials, the enclosure also may be made anechoic for audio energy by using sound-absorbing materials.
RF energy-absorbing materials also can absorb audio energy. Good practices for performing audio measurements include placing audio transducers (usually speakers and microphones) close to their targets on the DUT and creating a small sound chamber between a transducer and its target by using a sound-absorbing foam toroid (Figure 3).
Nesting
Wireless RF device testing requires the DUT to be held in position within the enclosure for protection and to maximize test re-peatability because of consistent placement. Many times, this holding apparatus is called a nest. A nest must accommodate DUT geometries and features and is used for positioning the DUT in all three axes in relation to the test-enclosure cavity.
A general-use nest should have a broad range of adjustability for variations among DUT models. Such variations include size, shape, prominent features, accessibility of buttons, connector positions, display position and size, speaker placement, microphone placement, and positioning requirements.
A nest dedicated to a single or similar DUT models doesn’t require such a broad adjustment since geometry, accessibility, and positioning may not need to be changed. A hard-tooled nest is a good approach for a dedicated nest (Figure 1).
Physical Connection
The system-to-DUT interconnect may include power, serial communications, digital I/O, RF signals, and audio signals. Some attributes of this connection to consider are wear-out life, ease of replacement, reliability, the ability to pass the anticipated signal frequencies, and appropriate current ratings.
For a general-use fixture, an easily detachable cable interconnect offers the flexibility to accommodate a variety of DUTs. But for a fixture dedicated to a specific DUT, a more direct connection could be more useful. In either case, wear-out replacement should receive due consideration. Examples of wear-out devices include an easily replaceable, inexpensive, sacrificial cable; a sacrificial connector saver; and replaceable spring pins (Figure 1).
Fixture Electronics
Generally, power and signal conditioning will be required between the fixture and the rest of the system. Consequently, fixture electronics in the form of an active conditioning module or PCB may be part of the fixture. Such a module or PCB could provide the following:
- Voltage level shifting.
- Serial communications voltages.
- Audio signal input and output conditioning.
- Digital I/O control and logic levels.
- RF signal routing.
-
An area for custom circuitry.
Other Considerations
There are many other considerations to keep in mind when designing a fixture. Here are some general questions to consider:
- What are the geometric constraints of the enclosure?
- What is the most useful opening and closing mechanism of the fixture?
- What does the connection to the test system look like?
- Will audio testing be required, and how will it be implemented?
- Will vision systems, such as those to look at displays, be required?
- Will an antenna coupler be required for over-the-air testing?
- What is the ergonomic environment, such as fixture height and operator load and unload movements?
- Will automation and robotics be required?
- Will the fixture be opened or closed manually or automatically?
- What are the operator safety considerations?
-
Is automated button-pushing anticipated?
Conclusion
The design of RF-isolating fixtures is far from trivial. The design, fabrication, and development, including RF isolation testing, ergonomic and usability testing, and durability testing, are all very time and resource intensive.
Typical fixtures usually are custom, expensive, not easily adaptable or reconfigurable to a variety of DUT models, and not necessarily well characterized regarding RF isolation and internal RF energy reflection. The first major decision for those requiring a wireless-device test fixture is whether to make or buy. After that, the questions and considerations presented here as well as any unique requirements must be explored and requirements generated to ensure that the purchased, contracted, or in-house designed and fabricated fixture will meet those needs.
About the Author
Bill Miner is a mechanical engineer in R&D specializing in the mechanical architectures and design of test systems. Before the Agilent Technologies split-off, he was with Hewlett-Packard for 20 years in R&D, production engineering, and materials engineering management. Mr. Miner has a B.S.M.E. from Ohio State University. Agilent Technologies, 815 14th St. SW, Loveland, CO 80539, (970) 679-5000.
Published by EE-Evaluation Engineering
All contents © 2000 Nelson Publishing Inc.
No reprint, distribution, or reuse in any medium is permitted
without the express written consent of the publisher.
May 2000