feature article
Subscribe Now

The Semiconductor Company from Another World: The Siliconix Story, Part 4 – Siliconix and Stewart-Warner Microcircuits

In their 15 years of working together since their marriage in 1947, Frances and Bill Hugle accumulated all the technical know-how they needed to manufacture transistors and ICs. From their efforts to produce gemstones at Stuart Labs, they developed techniques for growing crystal ingots, annealing those ingots, and doping them with impurities. From their years developing optical encoder technology at the DH Baldwin Piano Company, they developed the technology for depositing semiconductor thin films and using photolithography to pattern images on circuit boards and optical disks. At Westinghouse, during the late 1950s and early 1960s, they participated in clean room development and 1950s-style semiconductor manufacturing. At this point, they had everything they needed to start a semiconductor manufacturing company, and that’s precisely what they did. In fact, in Silicon Valley style, they started two or three of them.

The Hugles had moved to Thousand Oaks, California in late 1960 to help Westinghouse set up its Astroelectronics Lab. That lab was part of Westinghouse’s Molecular Electronics program for the US Air Force and was dedicated to developing miniaturized electronic components called FEBs (functional electronic blocks) that were based on semiconductors, but weren’t quite ICs. The skills and knowledge that the Hugles gained in this phase of their careers prompted them to start planning the creation of their own semiconductor venture by the end of 1961 or the beginning of 1962.

In March 1962, Frances and Bill Hugle founded Siliconix in Sunnyvale, California. Investment capital came from an old family friend, the DH Baldwin Piano Company, the Electronic Engineering Company of California, and W. van Allen Clark, Jr, of the Sippican Corporation. Although Siliconix did not spin out of Fairchild Semiconductor, it did draw employees from there and from several other early semiconductor makers including Fairchild, Motorola, Pacific Semiconductor, Rheem, US Semcor, Texas Instrument, and Westinghouse. Dick Lee from Texas Instruments joined at the invitation of the DH Baldwin Piano Company and became the company’s President. As was usual with joint ventures between Frances and Bill Hugle, Frances Hugle became the company’s Director of R&D.

Notably, Siliconix focused on FET manufacturing at the very beginning. This focus sets Siliconix apart from other semiconductor manufacturers, which were all producing bipolar products at the time. Yet the connection with the Hugles’ former employer, Westinghouse, was clearly there in a description of the company published in a book titled “Research and Development: A List of Small Business Concerns Interested in Performing Research and Development,” published in 1963 by the US Small business Administration. The book’s description of Siliconix’s activities reads:

“Integrated circuits, molecular & functional electronic blocks, including thin-film circuits, multi-element & electric-connected transistors, & equipment for processing, unipolar field-effect transistors, thin-film transistors, & tunneling effects. Miniaturized multi-stage thermos-electric coolers & photo-detectors.”

The terms molecular & functional electronic blocks” recall Westinghouse’s unique Molecular Electronics program, as discussed in Part 3 of this article series. Only Westinghouse and the US Air Force used that terminology.

There were many technical hurdles to overcome before FETs could be manufactured reliably. An article titled “MOS bandwagon starts rolling,” written by Walter Barney and published in the March 18, 1968 issue of Electronics Magazine, quotes Arthur Evans, who was the Chief Engineer at Siliconix by then:

“During a study for the Lockheed Electronics Co., Siliconix Inc. disposed of three traditional difficulties in building reliable MOS devices. They were ion migration in the oxide, leading to large swings in threshold voltages, sneak leakage paths between source and drain caused by ions in the oxide, and oxide puncture caused by electrostatic voltages.

“Siliconix found that using phosphorus-impregnated oxides, and covering the gate oxide with metal immediately after it is formed, a step also taken by GI [General Instrument, another early MOS semiconductor vendor], stabilized threshold voltages. Changing geometries, in one case, so that the source is a dot, completely encircled by a circular gate, eliminated sneak leakage; and building Zener diodes in parallel with each gate, so that charges are shunted to the substrate before the voltage is large enough to puncture the oxide, reduced electrostatic puncture. National met the latter problem by switching to 100 silicon, which has a lower pn junction breakdown between gate and base material than does 111; the pn junction can thus protect the gate oxide.”

To see what the company was developing during its early days, it’s helpful to look at the patents granted to Frances Hugle during her time at Siliconix:

Method of ultra-fine semiconductor manufacture, US Patent Number US3165430A, Filed January 21, 1963, Granted January 12, 1965

“I accomplish the precision mentioned by departing entirely from the photographic method in processing the device where minimum dimensions are required. A metal coating is employed to accomplish the resist function and this is mechanically scribed instead of being photographically exposed and developed.

“An object of my invention is to provide a degree of precision in semiconductor manufacture which has heretofore been unobtainable.”

Semiconductor etch and oxidation process, Patent Number US3258359A, Filed April 8, 1963, Granted June 28, 1966

“After repeated attempts to accomplish etching and oxidizing successively in the same system, I discovered that if the water vapor is not chemically formed until the components thereof are at the semiconductor material being processed, all contamination problems are eliminated I accomplish this requirement by employing a form of the water gas reaction in reverse equilibrium. Instead of steam being one of the components, as it is in producing hydrogen at the known artificial gas works, steam is formed by combining hydrogen and carbon dioxide gases at the surface of the semiconductor, which surface is maintained at a high temperature in the furnace.

“An object of my invention is to accomplish two heretofore incompatible processing steps within. one enclosure in semiconductor manufacturing.”

Horizontally aligned junction transistor structure, Patent Number US3246214A, Filed April 22, 1963, Granted April 12, 1966

“I have been able to form a new structure in which the interfaces are between elements which are horizontally disposed. This is accomplished by forming the base of two concentrations of doping; lightly doped in a small and annular shaped part of the element which carries the working interfaces and highly doped in the large supporting part of the element which also acts as the major body of the semiconductor structure. In effect, the working part of the structure is disposed horizontally, one element in relation to the other, atop the highly doped part of the base element.”

Process for manufacturing horizontal transistor structure, Patent Number US3328214A, Filed April 22, 1963, Granted June 27, 1967

“A process for forming a transistor having horizontally disposed elements.

“A body of high conductivity forms a supporting base. An epitaxial layer of opposite conductivity is formed atop that body, the outer portion of which layer, horizontally, forms the collector. A mask is next formed over the epitaxial layer and it is provided with an aperture. A working base is conjoined to the supporting base by diffusing a first vapor impurity through the aperture. An emitter is formed within the volume of the working base by diffusing a second vapor impurity through the same aperture and by stopping the second diffusion so as to retain a ring-shaped working base.”

These patents suggest that Frances Hugle was focused on developing individual transistor structures and manufacturing processes at Siliconix. The first devices produced and sold by Siliconix included P-channel junction FETs, N-channel junction FETs, MOSFETs, FET arrays, and power MOSFETs. The company also introduced bipolar DTL logic chips in 1963.

Vishay started purchasing big chunks of Siliconix stock in 1998 and had essentially acquired the entire company by 2005. Today, the detailed early history of the company has largely vanished from the current Vishay Siliconix Web site, but you can still find bits and pieces of that history with some diligent Web searching.

Although it gets little attention, the Hugles started another semiconductor company around the same time that they founded Siliconix. That company, Opto-Electronic Devices (OED), made CdS, CdSe, and CdSSe photocells. These products clearly echo the Hugles’s photoconductor and optoelectronic work for optical encoders at the DH Baldwin Piano Company during the 1950s. As an independent company, OED was short-lived. Gene Weckler, who worked for the company in 1962 and 1963 as a Product Development Engineer, was responsible for the development, evaluation and application of II-VI compound photoconductor devices. Weckler worked for Shockley Transistor Corporation before joining OED. He eventually became a pioneer in CMOS image sensor design. In 2013, while receiving an Exceptional Lifetime Achievement Award from the International Image Sensor Society, Weckler wrote that OED was “under financed and not well managed.” Sigma Instruments, a manufacturer specializing in thin-film processing, purchased Opto-Electronic Devices in November 1963, made it a subsidiary, and absorbed its photocell product line.

Having started the company in March, 1962, the Hugles departed Siliconix by the end of 1963. They immediately founded another semiconductor company in Sunnyvale – Stewart-Warner Microcircuits – with Frances Hugle again serving as the Director of Research and Engineering. If the name Stewart-Warner Microcircuits seems unfamiliar, the parent company Stewart-Warner is familiar to automobile enthusiasts. Stewart-Warner of Chicago was an old-line maker of dashboard gauges for cars. Its first product was the speedometer for the Ford Model T. The company had started an electronics division by the 1950s, which was a military and aerospace equipment contractor. The company also built radios and televisions. Stuart-Warner’s South Wind Division supplied heat exchangers for the Lunar Module that landed on the moon during the Apollo space missions. There’s scant evidence about how the Hugles in Silicon Valley connected with Stewart-Warner, but Bill Hugle was a persistent and energetic networker and dealmaker, so he seems to have somehow connected with the parent company.

Stewart-Warner Microcircuits was altogether different from Siliconix. Instead of individual transistors, Stewart-Warner focused on manufacturing ICs: first DTL, then TTL and ECL. However, the Hugles’ stay at Stewart-Warner Microcircuits was only a bit longer than their stay at Siliconix. The January 26, 1966 issue of Electronics Magazine reported:

“The Stewart-Warner Corp. is changing the executive lineup at one of its subsidiaries, Stewart-Warner Microcircuits, Inc. Jack Coffey, formerly with the Stewart-Warner Electronics division, has taken over as the subsidiary’s general manager and executive vice president, and John P. Gates will become manager of engineering and manufacturing. Gates formerly was manager of digital integrated circuits at Fairchild Camera & Instrument Corp.’s Semiconductor division. Being replaced is William Hugle, executive vice president. The status of his wife, Frances Hugle, director of engineering and research, is still in doubt.”

Of course, Frances Hugle left quickly after her husband was replaced at the company. If you’re wondering why the Hugles left Stewart-Warner, it would appear to have been a disagreement about the company’s fundamental direction, as described in an article titled “The Worrisome IC,” which appeared on the news pages of the September 4, 1967 edition of Electronics Magazine, without a byline credit:

“[Stewart-Warner Microcircuits], on the other hand, thought there was a larger potential in DTL, and dropped its TTL line during a management reshuffle months ago.”

That was a very poor decision in hindsight. TTL chips rapidly eclipsed DTL devices in the market. It’s little wonder that the Hugles left the company when they did, given that kind of decision-making by the parent company.

A year after their departure from Stewart-Warner Microcircuits, Bill Hugle published a comprehensive article about digital ICs in Computer Design Magazine’s January 1967 issue, which lists him as a Contributing Editor on the masthead. The article, titled “Integrated Logic Circuits: A Comparative Evaluation” included information about RTL, DTL, TTL, ECL, and MOS digital ICs from the deep perspective of someone who had been intimately involved in the manufacture of these devices. The article discusses many of the challenges in developing each type of IC at the individual device level, and, when reading it, I got the strong feeling it was adapted from an internal Stewart-Warner Microcircuits document that Bill Hugle might have written during the mid-1960s to explain the reasons for manufacturing a product line that included several logic families. One of the article’s conclusions was:

“No one integrated logic circuit form is superior to all others for all applications. For each application, the system designer should analyze the basic trade-offs of each logic line.”

It’s clear that Bill Hugle was an advocate for a broad digital IC product line and possibly wanted nothing to do with a company that focused exclusively on DTL parts. Frances Hugle left Stewart-Warner Microcircuits at the same time, as suggested by the installation of John P. (Jack) Gates as the engineering and manufacturing manager when Jack Coffey took over the reins of the company. Stewart-Warner Microcircuits would later reverse the DTL decision and would offer several bipolar and MOS logic families as well as mask-programmable transistor and gate arrays in the 1970s, but the Hugles were long gone by then.

According to Jack Coffey’s obituary, Philips Electronics acquired Stewart-Warner Microcircuits in 1978, and that name quickly faded from the list of semiconductor manufacturers, becoming almost lost in the mists of time. You can still find ICs branded as Stewart-Warner chips for sale on eBay. These relics prove that the company once existed.

The Hugles’ exit from Stewart-Warner towards the end of 1965 or early in 1966 also marked the end of their direct involvement with semiconductor manufacturing companies. Although they founded no other semiconductor makers, their departure from Stewart-Warner Microcircuits did not mark the end of their work in the semiconductor industry. The couple decided to stop making chips and to start using their accumulated expertise to help other companies manufacture them, as will be discussed in the final article in this series along with death, politics, and allegations of espionage.

Note: This history of Frances and Bill Hugle is sparsely documented on the Internet, and this series of articles would not have been possible without the aid and assistance of the Hugles’ grandson, Jake Loomis, and the founder of TechSearch, Jan Vardaman, who was instrumental in creating an IEEE Scholarship program in the name of Frances Hugle, funded in part by Jake Loomis’s mother and Frances Hugle’s daughter, Linda Hugle.

References

Survey of Power Semiconductor Manufacturers for Nippon Kokan KK, Semiconductor Industry Group. Dataquest, December 1986.

“MOS bandwagon starts rolling,” Walter Barney, Electronics Magazine, March 18, 1968, pp. 173-187.

“The worrisome IC,” Electronics Magazine, September 4, 1967, p. 23.

“Integrated Logic Circuits: A Comparative Evaluation,” William B Hugle, Computer Design Magazine, January 1967, pp. 36-47.

“Research and Development: A List of Small Business Concerns Interested in Performing Research and Development,” US Small business Administration, 1963.

Jack O’Neal Coffey Obituary

Leave a Reply

featured blogs
Apr 23, 2024
The automotive industry's transformation from a primarily mechanical domain to a highly technological one is remarkable. Once considered mere vehicles, cars are now advanced computers on wheels, embodying the shift from roaring engines to the quiet hum of processors due ...
Apr 22, 2024
Learn what gate-all-around (GAA) transistors are, explore the switch from fin field-effect transistors (FinFETs), and see the impact on SoC design & EDA tools.The post What You Need to Know About Gate-All-Around Designs appeared first on Chip Design....
Apr 18, 2024
Are you ready for a revolution in robotic technology (as opposed to a robotic revolution, of course)?...

featured video

MaxLinear Integrates Analog & Digital Design in One Chip with Cadence 3D Solvers

Sponsored by Cadence Design Systems

MaxLinear has the unique capability of integrating analog and digital design on the same chip. Because of this, the team developed some interesting technology in the communication space. In the optical infrastructure domain, they created the first fully integrated 5nm CMOS PAM4 DSP. All their products solve critical communication and high-frequency analysis challenges.

Learn more about how MaxLinear is using Cadence’s Clarity 3D Solver and EMX Planar 3D Solver in their design process.

featured chalk talk

Advanced Gate Drive for Motor Control
Sponsored by Infineon
Passing EMC testing, reducing power dissipation, and mitigating supply chain issues are crucial design concerns to keep in mind when it comes to motor control applications. In this episode of Chalk Talk, Amelia Dalton and Rick Browarski from Infineon explore the role that MOSFETs play in motor control design, the value that adaptive MOSFET control can have for motor control designs, and how Infineon can help you jump start your next motor control design.
Feb 6, 2024
10,841 views