Link This |
Email this |
Blog This |
Comments (0)
System Gives Accurate Flicker-Noise Measurement
May 19, 2008
It’s a well-established fact in our industry that the development costs for each succeeding semiconductor manufacturing process node continue to rise and that time-to-market pressures won’t be letting up anytime soon. This means that accuracy in measuring critical parameters such as flicker noise is of increasing importance. For the development of today’s advanced processes and devices, flicker-noise characterization has moved from being a marginal factor to becoming a major concern. Lower operating voltages, smaller geometries, increased device complexity, and new materials have all contributed to a situation in process and device development where flicker noise has a profound effect on the useful range of today’s devices and compromises the performance of advanced ICs. Specifically, flicker noise can cause jitter or phase noise in communication devices, resulting in high bit-error rates. It can also cause random retention errors in flash memory or soft errors in SRAM.
Adding more urgency to the accurate characterization of flicker noise, is the fact that while it was largely held that operating voltages would get lower as device geometries got smaller, the ITRS actually amended this observation in 2003 when it reported increased operating voltages as device geometries shrank.
As device voltages decrease, noise increases and negatively affects performance. Source: Cascade Microtech.
Performing the actual measurement and extraction with high precision has proven to be much easier in theory than in practice. Accurately measuring flicker noise is very difficult due to the multiple sources of extrinsic background systematic or environmental noise that can obscure the ultra-low-level noise corner frequency. Typical measurement systems are subject to disturbances from the 60-Hz input power radiation, switching power supplies from multiple instruments at 30 kHz or higher, radio transmissions from 1 to 100 MHz, and from handheld wireless devices at 1 GHz or higher. Also, at smaller geometries, flicker noise has become a key contributor to overall noise in high-performance devices.
Cascade Microtech (Beaverton, Oregon) has announced its EDGE Flicker Noise Measurement System, which the company touts as the only flicker-noise measurement system certified to provide accurate measurements from 1 Hz to 30 MHz. The company is referring to the new system as the industry’s only fully integrated measurement system, bringing together the wafer probe station, instruments, software, and accessories. Cascade calls traditional flicker-noise measurement solutions “bolted-together systems,” comprising system elements from up to five different vendors. R&D engineers, technicians, and lab managers who need to measure flicker noise for device modeling or process development and evaluation can have simple access to flicker-noise data over what the company describes as widest frequency range, with the lowest background noise.
The system is centrally controlled from a software GUI. The platform's software controls all instruments, executes DC testing based on user selection of bias conditions, recommends series resistances for flicker noise testing, dynamically calculates the system roll-off frequency, and calculates and sets required SMU voltages. Additionally, the tool provides unattended measurement execution—single die or full wafer—and graphically displays results and archives data files.
Source: Cascade Microtech.
Cascade claims that EDGE provides simple access to flicker-noise data over the widest frequency range—up to 30 MHz—with the lowest background noise, typically less than 1.2 nV/rtHz at 100 kHz and above. The wide frequency range and low background noise lets users increase device performance, command higher margins, and reduce time to market.
Because it is integrated, the platform eliminates the need for custom fixturing or instrumentation. In addition, it has been designed to switch between flicker and DC measurements with push-button automation, providing both sets of measurements, over temperature, in one system. This avoids potentially risky transfer of the wafer from one measurement station to another, and removes the risk of error associated with reconfiguring existing piecemeal set-ups to change their functionality.
Posted by Alexander E. Braun on May 19, 2008 | Comments (0)
