A practical approach to production test of UWB devices: Part 2
Adam Smith, Verigy -- Test & Measurement World, 12/18/2007 5:38:00 AM
Ed. Note: This article is part 2 of a 3-part series. Read part 1 here.
In order for UWB technology to be commercially viable, it must be very inexpensive. Consumers are not going to be willing to pay significantly more for a product solely because the wire has been removed. In other words, UWB is not going to be perceived as a value-added component to a system—it does not add features, megapixels, megahertz, or megabytes.
Because of this, the total cost for the system must be very inexpensive. The initial target for the four-chip UWB solution is less than $15 to $30 for the entire chip set: RF front-end, RF transceiver, digital baseband, and media access controller (MAC). Look to the selling price of Wi-Fi silicon over the last several years to see the price pressures that UWB will face after entry in the market. For device manufacturers, this means that the production, assembly, and test cost must be extremely low. This article will focus solely on the test cost. There are many challenges involved in testing UWB devices, and unfortunately, they are both economical as well as technical. In many markets, such as the high-end PC or digital TV markets, major technical challenges stem from added performance for the consumer, which allows for a higher average selling price (ASP) of the device—UWB does not have this luxury, and must contend both with the technical challenges in the face of expected low ASP.
Test challenges for UWB
As the name Ultra-Wideband implies, UWB presents a unique challenge with the bandwidth requirement: 528 MHz. Also, in order to find usable unlicensed spectrum to transmit a signal this wide, the channel frequencies range from 3.5 GHz to 10.5 GHz. Figure 1 shows the FCC frequency band allocations for UWB in the US (Ref. 1).
When looking at how these two UWB requirements fit in with other consumer devices, UWB is an obvious outlier. Figure 2 shows how different these two aspects (frequency and bandwidth) are from today’s high-volume consumer wireless devices. Because of this, the production test equipment, known as automatic test equipment (ATE), has generally been optimized to test the requirements shown in the area designated in the blue rectangle. This wireless market snapshot clearly shows that the ATE manufacturer would be wise to focus on the tight grouping of applications and serve the UWB market with an option to extend the capability of the tester.
Design validation vs. production test
To a manufacturer of UWB silicon, it is very important to understand and characterize the performance of the silicon. This is known as design validation or characterization. There are significant differences in focus between production test and design validation. For design-validation purposes, the focus is primarily on the performance of the silicon and not on trying to cost-optimize the test solution. When moving forward with production, the key for UWB manufacturers is to identify defects rather than characterize the silicon. The cheaper that one can identify defects, the cheaper the device becomes.
Since the UWB market is under such cost pressure, the focus on testing these devices must be only two areas:
• the need to distinguish a functional device from a non-functional device, and
• the lowest possible cost.
Any other focus on collecting device parametrics in production, other than random sampling, is just a luxury that cannot be afforded in UWB high-volume production test.
A practical approach to UWB test
In an ideal world, there is always a right tool for the job at hand. Again, as shown in Figure 2, UWB poses a challenge to the ATE manufacturer in both channel frequency and bandwidth that “breaks” the gradual evolution of technology with a very large step function. Even worse, the UWB technology driving the requirements is in its infancy, is suffering from differences of implementation, and has not yet gained any significant acceptance in the marketplace. Also, because of the delays in implementation, project volumes continue to push out. Figure 3 shows the projected production volume of UWB devices, based on Gartner’s February 2007 Ultra-Wideband Wireless Forecast.
Since UWB continues to be in the infancy of its market roll-out, this chart is probably optimistic. We can, though, make some statements regarding the relative volume of device shipments. There are generally three phases in the roll-out of a new technology:
• First phase: Early market penetration looking for an application to gain acceptance. This phase describes where UWB technology is today probably through 2008. In this phase, volume shipments are generally in the hundreds of thousands to a few million units. In this phase, the crucial factor for the device manufacturer is time to market; even more so than cost. Also, since shipment volume is small, “exotic” test solutions (that is, ones involving complicated test-fixture hardware) are acceptable, since only a very small number of the test solutions needs to be maintained. In this phase, meeting shipment commitments takes a slight priority over cost-optimization.
• Second phase: Market acceptance growing into the mainstream. This phase would involve a few key applications starting to drive the growth of UWB to adjacent spaces. An indicator for the entry into this phase would be crossing the mark of 1 million units per month, or roughly greater than 10 million units per year. UWB likely looks to enter this phase in the 2009 timeframe. In this phase, it is still acceptable to maintain the solutions that were created in first phase; however, the device manufacturers are going to start to rely more on the test equipment vendor to provide a more robust test solution while the cost of the solution is also becoming quite important.
• Third phase: Mainstream market acceptance. This phase generally lags the second phase by a year or two since the second phase usually indicates a volume inflection point. This phase will be indicated in the UWB market when shipment volumes cross tens of millions to greater than one hundred million devices. For UWB, this timeframe is still a fluctuating guess, but looks to be in 2010 or 2011. In this phase, it is no longer acceptable to have a complicated test solution, as “copy-exact” is now crucial in this high-volume part of the life cycle. Also, cost is now the driving factor for the device manufacturer more than time to market. In this phase, as in the Wi-Fi market, device ASPs will start to decline dramatically. It is extremely crucial to have a robust and cost-effective test solution in place, and the device manufacturer will require the test vendor to provide a turn-key solution for UWB.
A very rough rule of thumb for ATE is that one production ATE tester can test just over 6 million devices per year (assumptions: three test seconds per device and 60% overall utilization of the test equipment). Using this assumption, this means that the entire projected world-wide demand for production ATE (specifically for UWB) is roughly three testers for 2008, and approximately 18 testers for the volume in 2010.
It would benefit the device manufacturer and the ATE manufacturer to take a very pragmatic approach to testing these devices. Likewise, the UWB device manufacturer needs to its their risk by finding ways to test their devices on readily-available and existing production test solutions, as subcontract test houses will be tentative to commit significant investment in UWB testers until there is momentum behind the technology.
Ed. note: This article is part 2 of a 3-part series. Part 1 describes UWB technology, and part 3 covers test-cost targets, native testing, and DFT strategies.
REFERENCE
1. Kolic, Rafael, “An introduction to Ultra Wideband (UWB) wireless,” deviceforge.com, 2004, www.deviceforge.com/articles/AT8171287040.html.
Adam Smith is a business development engineer at Verigy. He has 10 years of ATE industry experience, focused on RF/microwave device test technology. Adam holds a Bachelor of Science in Electrical Engineering from Cal Poly, San Luis Obispo. He can be reached at adam.smith@verigy.com.
Talkback
Related Content
Related Content
Sponsored Links
TMW Resource Center
Browse Resources by Type:
- Access to EASE (Equipment Appraisal Search Engine)
DispoDog | Product Demo
VIEW NOW
National Instruments | White Paper
VIEW NOW
National Instruments | White Paper
VIEW NOW
View More
