ATE in the Age of Convergence and Exascale Computing

By Matthias Stahl, Business Development Manager, Advantest.

We are currently in the midst of the age of convergence – that is, the convergence of data from a range of applications and data sources. These sources constitute anything that creates data – ranging from human-created data, such as voice and video, through automotive, mobile, and wireless/IoT devices. This also includes edge computing and servers storing the massive amounts of data needed for high-performance computing (HPC), AI, machine learning and many other applications. 

This data must be processed, and that is where the age of convergence also becomes the age of exascale computing. The term refers to a supercomputer capable of calculating at least 1018 floating point operations per second. Currently, no single exascale computer exists, but the combined compute power available certainly exceeds this number. Figure 1 illustrates the parallels between data source and processing convergence that we are witnessing, with the chips and technologies that are being made for mobile, performance and next-generation computing – all of which have unique testing requirements.

Figure 1. The age of data convergence has given rise to the age of exascale computing

The V93000 platform has been used successfully for HPC test since its introduction in 1999. It became part of Advantest with the acquisition of Verigy in 2011, and we’ve continually added new capabilities that have enabled us to keep pace with customers’ HPC needs. Figure 2 shows that many diverse drivers are creating the need for these new capabilities.

Figure 2. Exascale computing creates new ATE requirements.

As transistor count increases with smaller nodes, scan data volume goes up as well. This creates the need for deeper memory, faster scan, and new methodologies. At the same time, as device nodes continue to shrink, power-supply requirements escalate – not just power as such but also power dynamics. For example, devices require power supplies that can accommodate fast switching with no glitches, providing stable and consistent performance.

Concurrently, multisite testing demand will increase, with the industry looking to keep test costs in line. Another trend is the integration of RF capabilities into many devices, requiring a test platform that can accommodate the full range of RF and digital test. Today, we already have power management ICs (PMICs) close to the CPU, and we will see more uses going forward for data center applications, creating high-voltage test requirements. 

Bringing test into the exascale age

The new EXA Scale™ Generation of the V93000 (Figure 3) addresses these challenges with advancements to the proven V93000 architecture, designed to enable new test methodologies. Initially targeted at advanced digital ICs up to the exascale performance class and RF devices, more applications like MCU or automotive device test will be added. The system is designed to provide superior processing power for massive test data, as well as the highest possible currents and up to 256 power channels per card. The tester and handler can be tightly integrated, which, when combined with the tester’s active thermal control, allows the test cell to offer superior thermal control overall. 

Figure 3. The V93000 EXA Scale features powerful processing capabilities in a compact footprint.

Four key innovations enable the newest V93000 tester to deliver exascale-level performance: Xtreme Link; new test heads; and a new universal digital and power supply card (Pin Scale 5000 and DC Scale XPS256 DPS, respectively) for lower cost of test and faster time to market.

Xtreme Link

The V93000 EXA Scale incorporates Xtreme Link, our specialized ATE network with edge computing capabilities. Rather than using off-the-shelf technology like Gigabit Ethernet, we created dedicated technology with an optimized protocol focused on test needs and requirements for high throughput and large test data handling. Figure 4 illustrates the structure and benefits of this network.

Figure 4. Xtreme Link technology enables massive scalability and flexibility for the V39000 EXA Scale.

Pin Scale 5000

The Pin Scale 5000 is a new digital instrument created to set a new ATE standard for scan test. The fastest general-purpose ATE pin on the market, it offers 256 pins running at 5000Mbps maximum speed, scan result capture at up to 5000Mbps, and <1.5ps RMS jitter for accurate reference clocks. The Pin Scale 5000 also features the deepest vector memory available, with 3.5G gigavector (GVec) scan per pin, or 28GVec scan per 8 pins using pooling and fan-out technology. It’s designed to enable all scan implementations, including parallel, multiplexed and HSIO scan, and its configuration flexibility supports high site count – this allows customers to speed their overall test time by performing parallel core test. 

Figures 5 and 6 provide examples of the superior measurement and performance capabilities that the card enables.

Figure 5. Pin Scale 5000 phase noise measurement example. The RMS jitter is just 0.9ps, well below the specified 1.5ps.

Figure 6. Pin Scale 5000 receiver performance at 5Gbit differential. Even at top speed still 55% height and 75% width.  

DC Scale XPS256

The XPS256 combines the best capabilities of several predecessor Advantest instruments. It features many pins with small currents (256 pins x 1A), and it can gang those resources to achieve high current as needed. Combining these capabilities in one DPS allows the XPS256 to offer optimal flexibility and utilization of resources, in a common configuration well suited for 5G, mobile and HPC/AI applications. In addition, its improved accuracy and dynamic response enable achievement of higher yields.

The XPS256 power supply covers wide-ranging current requirements, implementing unlimited ganging to scale from milliamps (mA) to thousands of Amps with no performance degradation. Combining three instruments in a single power supply, the DPS pin delivers best-in-class flexibility, accuracy (±150µV) and dynamic response, with full four-quadrant voltage-current (VI) capabilities and very small overshoot/undershoot, and provides zero-overhead simultaneous voltage and current monitoring. 

A unique feature of the XPS256 is its built-in probe needle protection. With individual needles connected to separate power supply pins, currents can be limited to 1A or less per needle. An ultra-fast (<1 µs) hardware clamp allows current to be limited and shut down almost immediately if needed.

Both the Pin Scale 5000 and the XPS256 utilize Advantest’s unique next-generation multicore test processor. The processor is packaged using 2.5D integration, with two 8 core die, and two memory chips, providing 16 fully independent pins in a very small form factor.

Figure 7.  Test processor and memory for 16 channels. 

New test heads

Three new extended test heads developed for the EXA Scale generation tester offer superior configuration scaling from engineering to high multisite applications: the V93000-CX with 9 universal slots, the V93000-SX with 18 universal slots, and the V93000-LX with 27 universal slots. All feature a “zero footprint” design – all electronics are integrated into the test head, eliminating the need for a separate rack. Together, the new test heads cover all application segments and a wide price range, while their enhanced infrastructure helps contribute to lower cost of ownership for customers.

Platform compatibility facilitates transition

The EXA Scale generation of the V93000 platform is compatible with the Smart Scale generation. Smart Scale cards will work with EXA Scale, the load board dimensions are compatible for ease of migration between the systems, and the EXA Scale system can run both the SmarTest 7 and SmarTest 8 versions of our ATE programming environment. This will allow customers the ability to select the V93000 configuration that best meets their product and application requirements.

To date, we have already shipped a significant number of V93000 EXA Scale systems, both for engineering and high-volume production, to multiple customers. We look forward to sharing further successes as the age of exascale computing speeds forward.

Testing AIP Modules in High-Volume Production for 5G Applications: Part 2

This article is a condensed version of an article published in the November-December 2020 issue of Chip Scale Review, p. 31. Adapted with permission. Read the original article at https://www.chipscalereview.com/issues/ChipScale_Nov-Dec_2020-digital.pdf#page=33.

Jose Moreira, Advantest Senior Staff Engineer, SoC R&D.

This article is a follow-up to an article [1] where we described options for high-volume over-the-air (OTA) testing of antenna-in-package (AIP) modules with automated test equipment (ATE). In this follow-up article we present measurement results for two OTA testing approaches: far-field and radiating near-field OTA. But before we go to the measurement results, we need to first define an AiP device under test (DUT) that can be used for the measurements.

Creating an AiP Evaluation Vehicle
Proper evaluation of an ATE OTA measurement setup requires an AiP module. Using, for example, a reference antenna instead of an AiP DUT (for instance, a reference horn antenna) would not take into account all the components specific to an ATE implementation like the DUT test-fixture PCB or the DUT socket. Using a commercial AiP module is also not currently feasible, since few available commercial AiP modules are would come with IP restrictions on using them to publicly show OTA measurement results.
Therefore, we decided to create the simple AiP module shown in Figure 1. It was manufactured in a multilayer PCB with a Rogers 4350B top layer and a BGA ball grid array on the bottom. The antenna array is composed of a 2 by 2 array of dual polarized patch antennas [2]. They are microstrip-feed with two quarter-wavelength transformers for impedance matching. This antenna design is narrow-band, was tuned for 28 GHz, and can support our 100-MHz measurement modulation range. We used a 0.4-mm pitch for the BGA array on the bottom of the PCB. Note that AiP modules come in a multitude of package types [3]. We chose this one because it was the simplest to design and manufacture.

Figure 1: Simple antenna in package module demonstration vehicle for OTA measurements.

Because we designed the antenna array using a microstrip feed (for simplicity), we needed to supply a waveform with a 180-degree phase difference on both sides of the array for each polarization to obtain an antenna beam that is in the horizontal direction. By selecting the appropriate phase difference, we could move the beam direction as expected from an AiP phased-array antenna. One critical point on OTA testing of AiP modules with ATE is that a DUT socket is always a must. The challenge is that the socket lid will have an impact on the antenna array beam, as shown in Figure 2. . One can only try to minimize its impact by the proper design of the socket and its material selection (especially the lid), but at the same time there are other conflicting requirements in a high-volume production test cell. These include, for example, supporting hot and cold testing as well as guaranteeing a proper electrical contact into the electrical side of the socket even in the presence of a small package warpage.

Figure 2: Socket lid impact in the AiP module antenna array beam.

An additional requirement to achieve a complete AiP module emulation is the active part—that is, the silicon die. To emulate that part, we used an external evaluation board (Anokiwave 0151-DK), which provides four dual-polarized RF channels with independent phase and gain control of each channel. With this complete setup shown in Figure 3, we can fully emulate the OTA testing of an AiP module. In the ATE system used for the presented measurements, only two ATE mmWave measurement channels were available. Therefore, we used a solid-state switch to switch between polarizations.

Figure 3: Block diagram of the used ATE OTA measurement setup.

Figure 4 shows the DUT test fixture (or loadboard) with the far-field socket installed but without a DUT. For the electrical side of the AiP module we used an elastomer socket because we needed to support 28-GHz signaling. The DUT test fixture is a simple mmWave design with the signals from the Anokiwave evaluation board connected to the DUT AiP via a microstrip trace. We also implemented some auxiliary test and calibration structures. The Anokiwave evaluation board resides on a garage space bellow the DUT test fixture. It is powered by the ATE power supplies, and it is programmed with ATE digital channels using an SPI interface. Figure 5 shows the ATE system configured for far-field and radiating near-field OTA measurements. All the measurements presented were performed using an Advantest V93000 Wave Scale mmWave ATE system.

Figure 4: DUT test fixture.

Figure 5: ATE measurement setup showing the far-field setup (left) and radiating near-field setup (right).

Before we proceed, we need some reference numbers for the far-field distance from the antenna array. Figure 6 shows the AiP antenna-array dimensions. It also shows a computation of the far-field distance using the Fraunhofer distance equation [4]. The computed far-field starting distance is 32 mm for this case.

Figure 6: Far-field distance computation for the antenna array.

Results with a Far-Field OTA Measurement Setup
In the far-field setup (Figure 5, left), the measurement antenna is a dual-polarized horn antenna (Ainfo LB-SJ-180400) located 10 cm from the DUT AiP (clearly within the far-field zone). We know the measurement antenna gain, and since the measurement antenna to DUT distance is fixed, we also know the air-loss. We can use these values to calibrate the measured results. Figure 7 shows the measured error vector magnitude (EVM) and corresponding constellation diagram measurement of the AiP DUT using a 28-GHz 5G QAM64 waveform with 100-MHz modulation bandwidth. This measurement is performed with the entire antenna array transmitting and pointing in the horizontal direction to the measurement antenna. Only the H-polarization field is measured (see Figure 3).

As previously mentioned, this AiP device is intended to be a demonstration vehicle; due to its simplistic design one cannot expect good performance. This measurement setup is straightforward and provides an easy way to correlate with a 5G-compliant bench measurement setup. Although the far-field OTA measurement setup is excellent for characterization, as discussed in [1], it presents major challenges for test-cell integration in a multisite high-volume testing setup due to its mechanical requirements.

Figure 7: Measured transmitted far-field EVM and constellation diagram of a 28-GHz 5G QAM64 waveform (100-MHz modulation bandwidth) for H-polarization.

Results with a Radiating Near-Field OTA Measurement Setup

In the radiating near-field setup (Figure 5, right), a dual-polarized patch measurement antenna is used on the socket (as shown in Figure 4 of [1]). This measurement antenna is set 11 mm from the AiP DUT, so it is within the near-field region as shown in the Figure 6 computations. The 11-mm distance was selected based on the standing-wave effect that is present on any radiating near-field OTA setup as described in [1,5]. Figure 8 shows the EVM and constellation diagram measurement of the AiP DUT in the exact same conditions as for the far-field measurements shown in Figure 7.

Figure 8: Measured transmitted radiating near-field EVM and constellation diagram of a 28-GHz 5G QAM64 waveform (100-MHz modulation bandwidth) for H-polarization.

Figure 8 shows a measured 2.76% EVM value for the radiating near-field, while the far-field EVM measured result (Figure 7) was 2.74%. Although in this example the EVM results correlate, for a different AiP module with a different antenna array or a different design of the measurement antenna and its distance to the DUT, the difference between a far-field and radiating near-field OTA measurement setup might be significant. Other issues relate to calibration, and other possible measurements include phase linearity and ACLR, as described in [7].

Summary
OTA testing with ATE is possible in different configurations: in the far-field, radiating near-field, and reactive near-field as described in [1]. An OTA loopback configuration can also support OTA testing in some circumstances. We have shown that parametric measurements can be done in the radiating near-field if careful attention is placed on the measurement antenna design and also on the choice of the distance between the DUT and the measurement antenna. In the radiating near-field case a straightforward value correlation is not always possible with the far-field. But in a production setup, the important task is to be able to differentiate good from bad devices, and that is achievable with a radiating near-field OTA configuration. As shown in [1], the radiating near-field has significant advantages for high-volume production in terms of complexity and cost.

Acknowledgements
We would like to thank Sui-Xia Yang and Frank Goh from Advantest for the test-program development and also for the measurements execution. We would also like to thank Natsuki Shiota, Aritomo Kikuchi, Hiromitsu Takasu, Hiroyuki Mineo, and Yasuyuki Kato from Advantest for their technical contribution for this project. We would like also to thank Prof. Jan Hesselbarth from the University of Stuttgart for his continuing collaboration on OTA testing.

References
1. J. Moreira, “Testing AiP Modules in High-Volume Production for 5G Applications,” Chip Scale Review, May/June 2020.
2. Kim-Lu Wong, “Compact and BroadBand Microstrip Antennas,” Wiley, 2002.
3. Curtis Zwenger, Vik Chaudhry, “Antenna in package (AiP) technology for 5G growth,“ Chip Scale Review March/April, 2020.
4. Meik Kottkamp, et al., ”5G New Radio Fundamentals, Procedures, Testing Aspects,” Rohde & Schwarz.
5. J. Moreira, J. Hesselbarth, K. Dabrowiecki, “Challenges of over the air (OTA) testing with ATE,” TestConX China, Shanghai, Oct. 29, 2019.
6. C. Parini et al., Theory and Practice of Modern Antenna Range Measurements, IET 2014.
7. J. Moreira, “Testing AiP modules in high-volume production for 5G applications,” Chip Scale Review, November-December 2020, p. 31.

T2000 with Multiple Interface Unit Supports RF SiP OTA Test

By Koji Miyauchi, T2000 Business Unit Solution Department, Advantest Corp.

Introduction
In recent years, the proliferation of the IoT has focused attention on low-power-wireless applications. IoT modules incorporating functions such as Bluetooth Low Energy (BLE) transceivers, MCUs, and power-management circuitry are becoming system-in-package (SiP) and even one-chip devices. Such devices increase the demand for a mass-production test environment that can measure them in a short time. To meet this demand, we at Advantest have focused on combining conventional ATE test and system-level test (SLT). Specifically, we have developed a hybrid SLT system that combines ATE test in the form of a T2000 system with SLT implemented using a multiple-interface-unit (MIU) box attached to the T2000. Advantest customers are using such systems today. In this article, we describe the hybrid SLT using a BLE SiP module as an example.

IoT and low-power wireless
The IoT represents a rapidly growing market for which high-speed wireless connections suitable for various usage systems are indispensable. The arrival of digital transportation is accelerating this trend. In addition to high speed, low power is crucial for IoT wireless communications, leading to the adoption of low-power wireless standards such as BLE and ZigBee for short-range communication and LoRa and Sigfox for long-range communication (Table 1).

Table 1. Low-Power Wireless Standards and Features

As shown in Figure 1, the ratio of BLE usage is high compared with usage of other communications standards for low-power wireless applications, and the average annual growth rate of BLE is 24.5%. IoT modules that can implement these standards can take various forms, from module-level devices that combine discrete components to SiP and single-chip devices. Due to the increasing demand for higher functionality, power saving, and miniaturization in recent years, IoT module functions increasingly will be integrated into one device. With the current acceleration of the IoT, the ratio of such SiPs and single chips to modules with discrete components is increasing, thereby driving an increasing demand for a production test environment that can quickly measure a large number of highly integrated devices at once . In addition, IoT modules built with SiP devices or single chips enable a wide variety of products optimized for each application.

Figure 1. BLE finds significant use in low-power wireless applications.

Technical challenges for SiP
An IoT module is a composite device that packages processing, power-supply, wired-communications, and wireless-communications functions. To test such a composite device, the tester must have the conventional capabilities required for testing multiple devices such as an MPU, a transceiver IC, and a PMIC. At the same time, a trend in the semiconductor manufacturing process is emerging that drives an increase in the ratio of SLT to conventional final test. Currently, there is a demand for system upgrades that incorporate the trimming process using conventional ATE, rather than system upgrades that combine actual end-application-level machines such as motherboards with measuring instruments. As shown in Table 2, the SiP presents technical challenges as well as benefits. The increasing demand for SiPs and the technical issues caused by their introduction are the reasons why the ATE environment is required instead of an upgraded SLT system using jigs such as motherboards.

Table 2. SiP Benefits and Technical Challenges for Testing

Asynchronous testing with ATE
One of two methods can help to solve the technical issues of IoT module SiPs, depending on the production volume:

1. A high-productivity measurement environment using multiple simultaneous-measurement SLT systems.
2. A hybrid environment that realizes SLT with conventional ATE test.

Here, we focus on the second: a hybrid environment that realizes SLT with a conventional ATE tester. This method is particularly suitable for high-mix production.
In conventional device tests, ATE takes the initiative in controlling the DUT to perform tests efficiently. However, SLT provides test in actual application-level use case, in which the application runs on the OS inside the CPU of the DUT.　In this application environment, with processing timing unique to the DUT, test operations are performed in accordance with the internal clock in the DUT’s CPU, so the output timing is uncertain. In other words, asynchronous testing becomes an issue when performing SLT on an ATE system. The testing must be performed mainly by DUT, since ATE is not good at test operations that are out of sync with itself.

To solve the problems related to asynchronous testing and SiP technical challenges, Advantest has developed hybrid SLT, which is a combination of SLT using an MIU box and ATE test. The MIU box is a unit that performs the SLT of IoT modules using tester resources and a processor that can operate asynchronously (Figure 2). The MIU box is controlled by the T2000 via Ethernet. Asynchronous test is realized while interlocking with the tester OS by letting the MIU execute the test script for SLT using the TSS (T2000 System Software) application software that has been used conventionally in the T2000.

Figure 2. The Hybrid SLT solution combines a T2000 tester and MIU box.

Implementation example of 16 parallel tests by OTA

Figure 3 shows an example of the testing of 16 BLE-equipped IoT modules using the T2000 MIU solution. Four DUTs can be measured for each MIU board. In addition, the MIU box is equipped with four MIU boards and can measure 16 devices. Moreover, because BLE has 40 channels, simultaneous measurement can be performed while avoiding channel interference. A performance board (PB) or load board with a shield function prevents radio-wave interference with a dedicated socket for OTA and the adjacent tester. Before and after the RF OTA test, digital tests and DC tests using conventional tester resources are performed.

Figure 3. This hybrid SLT example shows the testing of 16 BLE-equipped IoT modules.

The mass-production test steps in hybrid SLT (excluding the conventional chip-test items) are as follows; these tests are conducted asynchronously between the DUTs using OTA:

• Tx power. The DUT uses the received signal-strength indicator (RSSI) function of the RF front end to measure power while continuously transmitting Tx signals with direct test mode (DTM). [Is this correct: DTM stands for “direct test mode”?]
• RSSI. As with the Tx power measurement, the DUT receives the Tx signal from the RF front end with DTM, and the RSSI value is read out with the MIU.
• Packet error rate (PER). This test reads the PER while the DUT and RF front end communicate in the actual usage environment.
• Data transfer. After performing the advertising scan, this step connects the DUT and RF front end, sends and receives data at the application layer, compares the data with the expected value, and makes a judgment.

A test system can employ three types of OTA test methods—radio wave, electrostatic induction, and electromagnetic induction—depending on the application requirements. For this example, we have adopted electrostatic induction and have developed a socket and antenna board to implement this method. The Figure 4 block diagram shows the connection from the MIU box to the DUT as well as a photo of the MIU box and test head. Figure 5 shows results from the BLE PER test case.

Figure 4. The block diagram (left) shows the DUT-to-MIU box connection, while the photo (right) shows the MIU box and test head.

Figure 5. OTA test of the RF front end to DUT connection (left) yields the BLE PER test results Shmoo Plot (right).

Conclusion

Hybrid SLT is ideal for testing the ever increasing number of IoT modules. Since the traditional SLT communicates in an actual application-level use case, the overhead of test time tends to increase. However, the hybrid SLT can realize the concurrent conventional-ATE test and SLT, thereby shortening the total test time. Hybrid SLT can support wireless communication standards other than BLE, so customers can take advantage of Advantest’s wide range of the SLT solutions.