Caltech's Role in NASTRAN

by Howell N. Tyson Jr. (BS 50 ME)

THE CALTECH DIRECT ANALOG COMPUTER AND THE DEVELOPMENT OF NASTRAN

Description of NASTRAN

NASTRAN is a powerful general purpose digital computer program for the finite-element structural analysis of small to large and complex physical devices and systems. NASTRAN has been a proven standard tool in the field of structural analysis for decades. It provides a wide range of modeling and analysis capabilities, including linear statics, displacement, strain, stress, vibration, heat transfer, and more. NASTRAN can handle any material type from plastic and metal to composites and hyperelastic materials.

NASTRAN is written primarily in FORTRAN, and contains over one million lines of code. NASTRAN is compatible with a large variety of computers and operating systems, ranging from small workstations to the largest supercomputers.

Applications of NASTRAN

Development History: The Direct Analogy Electric Analog Computer

During World War II, Gilbert D. McCann, who had studied electrical engineering at Caltech (BS, 1934, MS, 1935, and PhD, 1939), was a research engineer at Westinghouse. At that time, Westinghouse research was largely focused on supporting the military. McCann set to work to devise a way of doing complex engineering calculations using electrical circuits to simulate mechanical forces, velocities, and displacements. These electric circuits formed the basis of a machine that McCann developed, called the Direct Analogy Electric Analog Computer (or Direct Analog Computer).

Persuaded to return to Caltech after the war, McCann started as an associate professor of electrical engineering in 1946, rising to professor a year later. He immediately established the Institute Analysis Laboratory to make a larger, improved version of his direct analog computer. Westinghouse was already at work on one when he left, and McCann negotiated a deal whereby they would make two of everything and ship the second set to Caltech for a very good price. Assembled in the Norman Bridge Laboratory of Physics with the aid of Charles H. Wilts (BSEE '40, MSEE '41, and PhD Electrical Engineering, 1948), and Bart Locanthi (BS Physics '47), the calculator weighed 33,000 pounds.

The principal uses of this new computer were the solution of design analysis problems in the fields of solid mechanics, fluid mechanics, and heat transfer. In particular, methods of structural stress analysis, vibration analysis, and aeroelastic analysis of airframes were developed. This latter research was carried out primarily by Charles Wilts, Richard H. MacNeal (MSEE '47, and PhD Electrical Engineering, 1949), and Gilbert McCann.

Caltech's direct analog computer was soon providing an invaluable service for JPL, the military, and most of the Southern California aircraft industry, all of whom lined up to have their engineering problems solved, particularly those related to missile guidance systems and aircraft structural design. At one stage, McCann recalled, most of the aircraft companies in the United States, and some in Europe, were customers; and in 1950, Engineering and Science reported that the laboratory was "too busy to take on all the problems which have been submitted." "It was a world-class instrument at the time," said Carver Mead, Professor of Engineering and Applied Science. "It could do things that nobody else could do anywhere." And by the early 1950's, the fact that the Institute Analysis Laboratory was becoming a computer bureau to service the needs of industry got to be too much for Caltech, particularly for Clark Millikan, director of the Guggenheim Aeronautical Laboratory. He suggested spinning off a commercial company, and Computer Engineering Associates was formed on Halstead Street in Pasadena.

Computer Engineering Associates

McCann was the largest shareholder, but could not run the company because he would have had to resign his faculty position. Richard MacNeal and William J. Dixon (BSEE '48, MSEE '49, and PhD Electrical Engineering, 1952), became senior managers at Computer Engineering Associates (CEA). CEA's service bureau customers were most of the aircraft firms, not only in Southern California, but also throughout the United States. CEA also built direct analogy electric analog computers for many of these same aircraft firms. Bart Locanthi was in charge of constructing these computers.

Digital Computers

But, by the early 1960s, the digital computer was powerful enough and easy enough to use by design engineers, even though they weren't programming professionals. Consequently, CEA could no longer compete with the large digital computers that were installed at most of CEA's client locations. Hence, by 1966, CEA was forced to close up shop, and by the end of the decade the direct analogy electric analog computer became a computer of the past.

NASTRAN

In 1963, Dr. Richard H. MacNeal, together with a colleague, Robert Schwendler, formed the MacNeal-Schwendler Corporation (MSC) for the purpose of developing engineering analysis software for the digital computer. And in that first year, MSC developed the engineering analysis program, SADSAM (Structural Analysis by Digital Simulation of Analog Methods).

Then in 1965, MSC participated in a NASA-sponsored project to develop a unified approach to computerized structural analysis. The resulting finite-element analysis program became known as NASTRAN (NAsa STRuctural ANalysis program); it was one of the first comprehensive efforts to consolidate structural mechanics into a single digital computer program.

In 1971, MSC announced a proprietary version of NASTRAN.

In 1973, MSC expanded its operations from the United States to Europe and to the Asia-Pacific region; in 1993, to Russia; and in 1996, to South America.

Today, NASTRAN is widely used throughout the world in the aerospace, automotive, and maritime industries. It is considered the industry standard for the analysis of aerospace structures.

Last updated 2/16/07.

Christopher E. Brennen