If you ask a hundred engineers who created the commercial function generator, at least 99 will tell you it was Wavetek which was not quite the case. It's close, but there are several people who could vie for credit.
Wavetek certainly popularized the instrument and placed it the hands of thousands of engineers and production workers. But the first function generator came from . . . well, nobody is really certain.
Hewlett-Packard Co., which produced almost any test instrument you could name, had a function generator, the 202A, in 1951. Designed by Robert Brunner, it was a specialized, very low-frequency instrument, aimed mainly at servo testing, vibration and geophysical studies. It produced sine, square and triangle waves at frequencies ranging from 0.01 to 1,000 cycles per second in five ranges. (This was before the hertz was invented.)
Bell Labs had announced the invention of the transistor just three years earlier, but transistors couldn't yet provide all the functions that one could get with tubes, so transistors weren't in widespread use. The 202A used vacuum tubes, so it was a large box 19 inches wide x 13 inches deep x 10 1/2 inches high and weighed 38 pounds. It may not have been the earliest function generator. Nobody remembers who made an earlier one, although people vaguely recall that there was a predecessor. But the 202A was the first to reach down into those low frequencies.
The 202A aimed at an important but limited market. So the instrument wasn't one you would find on every engineer's workbench. In fact, you didn't find many function generators around in any frequency range. That changed in the 1960s, when the function generator usually a Wavetek function generator joined the scope, DVM and power supply as a must-have instrument.
And that came about because of work on an oscilloscope an unusual oscilloscope and then a DVM.
The key man, at first, was Joe Deavenport. With a BSEE from Texas Tech in 1949, he joined the Naval Electronics Lab in San Diego; then joined Emerson Electric in St. Louis; then returned to San Diego, working first as an engineer with DVM manufacturer Cubic Corp. and then as vice president of engineering at DVM manufacturer Electro Instruments from 1957 to 1961. (Deavenport describes himself as self-unemployed for 40 years, developing new operations. Most of them failed. Two lasted decades: California Instruments and Wavetek.)
While Deavenport was a key man in developing the commercial function generator, the key area was San Diego, a hotbed of electronics activity in the 1960s, largely because of the Naval Electronics Lab, General Dynamics, Ryan Aeronautical and the burgeoning DVM business.
After leaving Electro Instruments, Deavenport founded California Instruments (CI) in the summer of 1961. CI was one of half a dozen companies he founded during his career.
CI's mission was to develop an automatic oscilloscope, an instrument Deavenport designed and patented, that would select the best ranges for displaying a few cycles of an input signal. The work on the scope captured the attention of engineers at Hughes Electronics, which had already introduced the Mem-O-Scope, the world's first storage scope. The Hughes engineers would visit CI to learn more about its unusual scope. During their visits, they got friendly with Deavenport, learned of his earlier experience at DVM manufacturer Electro Instruments and mentioned their need for a special, peak-to-peak reading 5-digit DVM.
In the fall of 1961, just a few months after Deavenport started California Instruments, he sold his interest, rented a space on Engineer Road (where he remained for 37 years) and started a consulting company, Electro Design, to develop that 5-digit voltmeter for Hughes.
California Instruments continued work on its oscilloscope and hired a consultant, Joel Naive (rhymes with rave), to complete the project. Naive, who had a BA in math from Pomona College and an MSEE from Stanford, had just left DVM pioneer Non-Linear Systems to start a consulting business. At NLS, where he spent five years, he had managed the Specials Department, which at one time was responsible for more than 40 percent of the company's business.
As Naive began to immerse himself in the scope design, he developed a number of questions about what Deavenport had already done. So he would call or visit to discuss the scope project. They became friends, and their discussions went beyond automatic scopes.
Meanwhile, Deavenport needed a slow ramp circuit to test the DVM he was developing. So he designed a simple function generator using circuitry that had been around a long time. Deavenport's circuit produced square waves and triangles at rep rates up to a few thousand hertz. (He couldn't afford the HP function generator.) For a very fast output stage, he used a four-transistor flip-flop that he had designed for the automatic scope in 1961. The circuit, one of 30 or 40 he patented during his career, could provide a 5-ns rise time for a 1-V square wave, which was remarkably snappy at the time.
Deavenport's technician, Lon Severe, put the circuit together with a few switchable frequency ranges. He used germanium transistors (silicons were still too slow), and other components mounted on Vector boards.
When Naive saw Deavenport's function generator, he suggested that an instrument that generated multiple wave shapes over a broad range of frequencies could prove very handy to lots of engineers.
Driving along the freeway one day, Deavenport and Naive shook hands on a deal to go into business producing function generators.
Again using Vector boards, Severe built a prototype that could deliver square waves and triangles at rep rates up to 1 MHz. They never tried to sell that instrument, which they informally dubbed the DN-1 (for Deavenport-Naive).
Then Naive took Deavenport's design and, starting in September 1962, spent months in his garage packaging what was to be the first solid-state function generator: the 101. The box was finished in December for a cash outlay to that date of $426. The first production run was 25 units. A year after Naive finished the first box (with just square waves and triangles), the company moved out of the garage to its first rented space.
Deavenport and Naive each pitched in $5,000 and, in April 1963, the company was incorporated. After they built the first unit, Naive took it on the road to round up sales representatives. But why, the reps asked, would anybody buy an instrument they never heard of from a company they never heard of?
Naive had a ready response. Even if they couldn't sell the instrument, he told them, they could use it as a great tool to demonstrate their scopes, counters and strip-chart recorders. The price of the portable, line- or battery-powered instrument was about $395.
The reps bought the idea and, in time, sold more Wavetek function generators than anybody would have predicted. At one point, Severe recalled, "we were shipping hundreds every week, and orders kept coming in. We wondered if somebody out there was eating them."
Much of the company's success can be laid at the feet of William Zongker, a former product manager at Non-Linear Systems who shared with Naive a love of vintage cars and motorcycles. He joined Wavetek as a consultant, then became marketing manager and, in October 1969, vice president. He stayed seven years. He was replaced by Tom Kurtz, whom he had hired.
Wavetek continued to prosper, offering an instrument with sine waves in addition to square and triangle, then a succession of instruments with voltage control, pushbutton control, crystal control, remote control and program control. There was a move into pulse generators and arbitrary function generators.
Deavenport missed the wave of prosperity, however. He left the company after only seven months and sold most of his interest so he could return to the consultancy business. Today, as Joe Deavenport Electronics, in San Diego, he's developing a lightweight frequency-changer power supply.
In January 1965, Naive suffered a heart attack and decided to abandon day-to-day operations of the company and concentrate on his first love, design engineering. He took the title of vice president of engineering and chairman and, later in the year, hired John Thornton, a former naval officer in Korea and a Harvard B-school graduate, as president and chief operating officer. Thornton had been on Wavetek's board of directors since 1963 and had been a Rohr Corp. executive and Naive's friend for eight years.
Thornton later recalled the problem posed by the battery-operated instruments, which accounted for 5 percent of the units and 80 percent of the headaches. Users kept calling because their instruments stopped working. They kept forgetting to recharge the batteries.
In 1968, Wavetek branched further afield and created its "Dialamatic" Model 201 differential voltmeter. That meter would mechanically switch to the next or previous range as a user reached beyond 9 or below 0 on one range. Thus, for example, if a user advanced the second digit when the meter's knobs were set to 49, the first digit would automatically switch to 5, to provide a dial setting of 50, not 40, protecting the meter from overvoltage. It was a clever design, but it bombed in the marketplace, possibly because tackling the king of the differential-voltmeter market, John M. Fluke Mfg., proved too tough.
In May 1968, Wavetek bought an Indianapolis company, Sweep Systems Inc., which was close to bankruptcy, and renamed it Wavetek Indiana. The company made TV and CATV test equipment.
Now past its prime, Wavetek began a series of management changes in 1978. Based on Naive's recommendations, the board transferred his CEO responsibilities to Thornton. Naive officially became a consultant but retained a position as director. David Goodman, a longtime director, was elected chairman, a post Thornton took over in 1983, when Louis W. Abbott, formerly general manager of Sweep Systems, was appointed president and chief executive. Abbot retained that position until John W. Battin replaced him on April 1, 1985, a year of a large loss, the company's first. Battin, in turn, was replaced by C. Frederick Sehnert in 1988, when the company suffered another loss.
In 1987, Wavetek bought Datron Instruments, a British maker of precision calibration instruments; in 1989, it bought CT Systems, a manufacturer of communications-test equipment.
By this time, Thornton was no longer chairman. He had resigned that position on the advice of the company's lawyers, because, in strong disagreement with others on the board of directors, he urged that the company be sold because it was no longer at a forefront of technology. He was replaced by David Pivan.
In its final year as an independent company, the fiscal year ended September 30, 1990, Wavetek recorded its highest ever level of revenue more than $90 million.
In 1991, Thornton, still on Wavetek's board and the largest shareholder, negotiated the sale of Wavetek to a British venture-capital company, Torrey Investments, headed by Terence Gooding.
Wavetek then acquired several companies, including the Instrumentation Products Division of Beckman Industrial and the Communications Test divisions of Schlumberger.
In October 1993, Naive died in Ireland after an exhilarating run of antique motorcycles at the Irish Motor Cycle Rally. Some porridge he had eaten had gotten stuck, eventually rupturing his esophagus, and he succumbed to infection some weeks later.
In 1998, Torrey merged Wavetek with Wandel Goltermann, a German instrumentation manufacturer, to form WWG. In January 2000, WWG sold Wavetek Precision Measurement (the former Datron) and Wavetek Testing to Fluke Corp., which today continues to offer Wavetek-branded function generators and Datron-labeled calibrators as well as function generators and other instruments with the Fluke brand. And in May, TTC Corp. bought the remainder of WWG and formed a new company, Acterna, to manufacture automatic test systems and a variety of communications products.
Function generators had come a long way.