For years, Raychem Corp. has been one of Silicon Valley's premier high-tech pioneers -- even though it has virtually nothing to do with silicon, computers or software.

Instead, the $700-million-a-year firm has made its fortune applying a different brand of high technology to basics such as electrical connectors, soldering sleeves, coaxial cable splicers and literally thousands of other similarly mundane industrial paraphernalia.

While its products may be dull, many analysts and observers believe that the technologies underlying them put Raychem in a superb position to be part of what should be one of the most exciting multibillion-dollar growth industries of the next decade: new materials.

Everything is made out of something; and Raychem has been developing proprietary technologies in some of the highest-margin segments of the materials science business.

"They have a dominant position in an exciting and alive technology," said Mark Hassenberg, an analyst with Donaldson, Lufkin & Jenrette. "They're so good . . . they have so many opportunities."

"We specialize in 'transition' materials," said Raychem's founder and chairman, Paul M. Cook. Transition materials sense and respond to environmental changes -- to light, to heat, to sound, to chemicals, to electricity. In essence, Raychem tries to make its materials "smart."

The company got its start by giving materials a memory in 1957.

Cook, who in the grand Silicon Valley tradition started Raychem in a garage, had been a Stanford Research Institute engineer conducting experiments in "radiation chemistry" (from which Raychem derives its name) and its effects on plastic.

Normally, plastic simply melts when it's heated. That's because plastic consists of separate and independent molecules. However, when plastic is bombarded by high-energy electrons, the molecules may fuse and interlock into "crosslinks" that will withstand tremendous amounts of heat.

Consequently, those crosslinked molecules create a skeletal structure that "remembers" the original shape of the plastic no matter how much it has been expanded or contracted.

What Raychem does is form a plastic into its end-use shape, use the electron-bombardment technique to create the shape memory, and then heat the product to shrink it and cool it to allow it to expand again. It's shrunk again, retaining its desired shape, when the Raychem customer installs it.

One example of this "heat-shrinkable plastic" is a splicing case for coaxial telephone cable. When the cable pieces are aligned, the repairman wraps them in plastic and uses an acetylene torch to shrink the plastic into a snug, waterproof fit. A more primitive version of the same technology, familiar to most people, is the heat-shrinkable plastic wrapping used for supermarket meats.

Raychems TraceTek water-sensing technology was developed when company scientists evaluating its system to help stop corrosion in bridges and buildings wondered what could be developed to detect water, the source of construction corrosion.

"This is part of Raychem's efforts to supply complete systems and not just components," Halperin said.

Raychem's heat-shrinkable-plastics technology has been the nucleus of the company's business. Other core technologies include "heat recoverable" metals, the metallic counterpart to the company's plastics technology; and "conductive polymers," plastics that can conduct electricity.

However, Raychem's materials apparently do a better job of managing transition than does Raychem itself. For the last three years, the company has failed to meet its self-described "very conservative" sales estimates -- in large part because of the unexpected strength of the dollar. Raychem derives more than 55 percent of its revenue from overseas. Moreover, the company's fiscal 1985 sales grew by only 7 percent to $675 million, while net income dropped 24 percent and earnings per share fell by 26 percent.

The bottom line is that, after 25 years of 25-percent-a-year growth and a reputation for enlightened management, Raychem has been encountering the same growing pains that other entrepreneurial Silicon Valley firms endure. "We weren't putting enough dough into the technology," said Cook, the company's chief strategist and guiding force, pointing out that Raychem's research and development expenditures had dropped to 6.3 percent of sales five years ago. "The problem was foraging for new technologies. In 1981, we decided to crank up R&D," he said. Now Raychem is spending more than 10 percent of sales on research.

But Raychem's problems run far deeper than technology. They strike at the company's tradition as an innovator that will give its people the chance to take risks and make mistakes -- even expensive mistakes. "They're not as conscientious about the cost side of things as other well-run companies," Donaldson, Lufkin's Hassenberg maintained. "That's what's really been hurting them."

Raychem recently took painful steps to cut its costs, including laying off more than 500 employes since March. That sort of surgery is especially painful at Raychem where literally dozens of Raychem employes have been with the company for more than 20 years.

Recognizing that it faces competition from Japan, the company is broadening its market definition and efforts to secure technological advantages.

Raychem's current financial disappointments in the context of its past success have prompted Cook to wonder if the company isn't "tough enough . . . maybe we had it too easy for too long.

"This is the key moment in Raychem's new history," he said.

The document for that new history is a 10-year "very risky" plan that calls for Raychem to resume its historic 25-percent-a-year growth, to stress profitability as well as sales, to increase the number of product introductions dramatically, and to turn Japan into Raychem's second-largest market after the United States.

In effect, the plan is to transform Raychem into the most profitable multinational materials-science company in the world.

The company now has extensive research under way in about 20 new areas of materials science ranging from ceramics to liquid crystal displays to "thin film" technologies that would allow electronics companies to fabricate physically flexible circuits.

Now Raychem has expensive programs under way in semiconductor "packaging" -- figuring out how to provide materials that enable even tinier and denser silicon chips to be linked physically to computer systems. The company also is looking at better ways to mount chips on circuit boards.

This thrust into electronics is a vital part of Raychem's efforts to win a place in Japan's electronics industry -- which is indispensable to the company's goal of making Japan its second-largest market.

The company also is trying to establish itself as a materials supplier to the automobile industry, a market it formerly ignored and one that could add hundreds of millions of dollars in revenue.

Indeed, a large part of Raychem's future hinges upon its ability to become a key supplier in areas expected to enjoy significant new capital investment -- such as electronics, infrastructure repairs and automobiles.