Please note that JavaScript and style sheet are used in this website,
Due to unadaptability of the style sheet with the browser used in your computer, pages may not look as original.
Even in such a case, however, the contents can be used safely.
Volume 66 (June 08, 2007)
Aspiring to attain a high current output: A power MOSFET development story
The story behind the development of automotive power MOSFETs began back in 2002 as a challenge to become the first in the industry to develop an ultra-low resistance power MOSFET capable of "breaking the 2 mΩ barrier." Then, in May 2004, use of a 0.25 µm process (UMOS4) led to the realization of an on-resistance value of 1.4 mΩ (maximum guaranteed value of 1.8 mΩ) at a withstand voltage of 40V, which in turn brought about the advent of the NP Series with its high current capacity of 110A. At that time, a new development story had already begun in which developers aimed to attain even higher current output.
Aiming toward the development of a power MOSFET with even higher current output
The advent of the NP Series received a great deal of attention in the industry because experts believed that the product line would spur the wave of computerization of automobiles through the diffusion of electronic power steering—which was already being employed in compact cars—in large-size vehicles. However, knowing that, within five years, further advancements in electric power steering would be made, primarily in large-size vehicles, a power MOSFET with even higher output capacity was desired. One day, we were approached by a customer who said, "We're looking for a 160A-compatible high-quality power MOSFET." Apparently, the customer wanted to know whether or not a high current could be realized using our NP Series of MOSFETs for electrical equipment. We immediately began conducting a benchmark survey of competitor products. Although there was fierce competition among companies in terms of high current products, no company outside of Japan had taken the lead in 160A-compatible power MOSFETs and Japanese semiconductor manufacturers had yet to release a product with 160A output. Seeing this as a "chance to enter the market," NEC Electronics embarked upon full-blown development of a high current output product.
At the time, an output of 110A was the limit for the 3-pin TO-263 package used in the new NP Series of products. The Production Engineering Group's Toshinori Kiyohara, who received the request to develop a 160A-compatible package, set out to achieve compatibility with high currents by utilizing a 7-pin lead, which had never before been used in Japan. By adjusting wire positioning to enable well-balanced flow of electric current and using 400-µm-diameter aluminum wire for internal connections, NEC Electronics engineers soon realized that the development of a 160A-compatible package was actually feasible. To accommodate this package, a 250-nanometer (nm) UMOS4 process was adopted for chip design, which led to even greater improvement. Development had gone smoothly up until this point, but no one at the time knew the rocky road that lay ahead.
Test production and evaluation using hand-made samples
Test production of samples for evaluation began, but at that time in 2005 there was no manufacturing equipment available for sample test production of the package that had been designed. Manual equipment was therefore used at the evaluation facilities and manufacturing plants of equipment manufacturers to create hand-made samples. It became a sort of sample trial production "relay" encompassing places ranging from Fukui to Tokyo, Malaysia, Kyoto and Kanagawa. At first, since there were differences in manufacturing precision, process conditions and so forth at each facility, things did not necessarily go according to plan. As evaluations of the finished samples were carried out over and over again, Kiyohara's enthusiasm became contagious and the vectors of those involved became one. Everyone banded together, and day after day passed as they worked to produce samples and conduct evaluations. It was these efforts that resulted in the completion of a hand-made sample. Kiyohara says with deep emotion, "If we had not had a cooperative framework in place for the chip design department, package design department, plants and so on, there is no way we could have accomplished this."
In January 2006, the baton was passed from Kiyohara to Kensuke Yamamoto of the MOSFET Development Group. Yamamoto carried out quality checks using the sample created with Kiyohara's own hands. Among these checks, the one for limit evaluations using a pulse current turned out to be rather harrowing. After it was confirmed that the sample could sustain up to 1000A of current, the team heard a loud bang and the sample exploded. Yamamoto recalls, "It was scary because although we predicted it to happen, we had no idea at which point the sample would explode until we actually conducted the test." In the aluminum-wire fusing-current test conducted next, the sample once again exploded. Although everyone was surprised by the explosive sound of the samples each time, Yamamoto was relieved to learn that the current value was identical to what had been designated in the original design.
Ensuring quality through trial and error
The next task was to determine how to ensure quality. In April 2006, shipment of samples began. Koichiro Oku of the Product Planning Group looks back on that time, saying, "Electric power steering is a field that is expected to see a great deal of growth in terms of automotive systems in the future. Since this was a major project in which we were aiming to attain the highest level of current output in the industry, we were prepared to forge ahead at any cost to see this product through to mass production, and were determined to meet our customers' expectations." A request was soon received from a customer to conduct a high current power cycle test. Although a test jig for high current tests had not been prepared and an evaluation method was yet to be established, the evaluation was carried out using a process of trial and error. However, it was by no means easy to achieve the level requested by the customer.
Oku's heart nearly broke with each test as he wondered why things weren't working. Determined to meet customer expectation, he held discussions time and time again with related departments in an effort to find a solution for the defects. Kiyohara, who had been involved in package development, was also a part of this, as was Yoshimasa Uchinuma of the MOSFET Development Group, who had taken over Yamamoto's duties. Moreover, Tsuyoshi Tachiyama, a quality assurance professional from the System Reliability Group, was involved in the advancement of quality evaluations. Kiyohara says, "To investigate the causes, we spent day after day going through the process of analyzing, investigating, experimenting, evaluating and altering the design. It was a time of many hardships for us." Tachiyama, reflecting on the days of trial and error, adds, "We were discovering defects through tests we had never conducted before, and were unable to provide quantitative explanations. At times, we were even scolded by the customer. In addition to visiting the customer numerous times to discuss the causes of the defects, the mechanism behind the breakdowns and corrective strategies, all of the departments worked together to carry out test production to improve the product. The production line was improved, selection requirements were reconsidered, and test production and evaluations were repeated. It was after all this that the quality problems were finally resolved.
Mass production finally begins
It was now October 2006. To establish the diffusion and assembly processes for mass production, an engineer was stationed full time at the plant. Also at this time, all of the related departments united to establish a mass production system. Uchinuma explains, "The power MOSFET development team was composed of many talented people, each of whom contributed his or her opinions freely, regardless of department or position . We were able to carry out development as a result of frequent questions, consultations and discussions." Then, at the end of February 2007, development of a 160A output product was completed. Everyone on the development team speaks of the sense of relief they felt afterward and says, "We were just extremely relieved."
Working towards the realization of the industry's first "180A output" power MOSFET
Currently, the power MOSFET development team is moving ahead with the development of what is to become the industry's first "180A output" power MOSFET. To achieve 180A compatibility, package development has been reconsidered, and a 500-µm-diameter aluminum wire will be used to further increase electric current flow. Development is advancing full steam ahead, thanks to the technology and experience cultivated during the development of the 160A output model. Uchinuma states, "Being involved in the development of a product that has had such a great impact on the market has been a very valuable experience for me. Since this product will be used for applications such as automotive products, there are many new tasks we must take on even within our own departments, and that is highly motivating." Oku states with confidence, "We will continue to trace and follow development as we aim toward the commencement of sales of this top-performance 180A product."
"Working towards the development of unique products!"
It is because of such great motivation that the development of power MOSFETs will continue.
269KB)