A ‘chief skunk’ looks back
By Jonathan Coopersmith|March 2023
Sherman Mullin, former president of Lockheed Skunk Works
Given the big jobs Sherm Mullin held over his 56-year career, one might expect to find a plethora of advanced degrees on his resume. But Mullin spent only a semester at Princeton University. He attributes his rise through the ranks of the Lockheed Aircraft Corp., as it was called before the 1995 merger with Martin Marietta, to self-discipline and the company’s willingness to judge employees by their accomplishments, not their formal credentials. He’s also quick to note the role that good fortune played in him becoming the second program manager of the F-117 stealth fighter and his subsequent appointment as head of the famous Skunk Works division in the 1990s — two of the three “professional surprises” of his career. I reached Mullin in his California home by phone to discuss these surprises and the lessons he’s learned about program management along the way. Here’s our conversation, condensed and edited.
Q: In a 2009 letter on innovation, you wrote that “individuals count, process does not.” Is that how you would describe your life philosophy?
A: I had a lot of good fortune along the way, and I tried to treat other people the same way. I didn’t realize this explicitly at the time, but in my 35 years with Lockheed, my lack of a degree never was a factor. I was judged on the basis of the work I accomplished. Later, as a manager and executive, I was judged on my ability to successfully lead ever increasingly large groups. Another factor that was very important in my career, which I had no control over, were external events in the electronics area. When I started out, I happened to hit when the transistor was being introduced in the late ’50s and vacuum tubes were going down the drain. Second, in one of my early jobs, I gained a really good knowledge of digital computing at a time when analog computers were going down the drain. Third, in the early years of my career as an electronics engineer, my seniors were people — and we used to joke about this — who had electronic engineering degrees and some had master’s degrees, but their knowledge had become irrelevant. In other words, they had learned a lot about classical power, electronics, generators, motors, transmission lines, vacuum tubes and all of that.
Q: When you were growing up, were there family expectations or values that shaped what career path you would take?
A: My father was from New Brunswick. He immigrated down in the 1930s, and he had the standard eighth grade education. His influence on me was “work hard, be honest, don’t screw up, don’t destroy your integrity. But do what you think you need to do because I am not in a position to give you any expert advice.” I managed to screw up. In high school, my self-discipline was pretty zero. I never did much homework. I had four years of math from an outstanding teacher. We also had an outstanding English teacher, and in my junior and senior year, we wrote more stuff than you can imagine. As a result, I wanted to go to college. That’s the good news. The bad news is you can’t go through Princeton in a completely undisciplined way. At Princeton, the high point was getting admitted. The low point was I was put on academic probation after my first semester. I decided to drop out. Most everybody of my age served in the military, so I went home to the local recruiting office. In another life-changing event, the recruiting sergeant told me, “With your background, you can qualify for guaranteed technical training. I’ve got a bunch of brochures.” He picked one and said, “This, I don’t know what it’s all about.” The brochure was called “guided missile electronic maintenance.” This is where my career really started, without me knowing it. I ended up going to Fort Bliss, Texas. I went through a four-month cram course on electronics, which was mostly vacuum tubes, including good labs, and another four-month course on the Nike Ajax missile system. I finished third in my class. In April 1955, I was appointed to the faculty of the Guided Missile School as a private first class. This wasn’t as exotic as you may think because the school was expanding rapidly and, out of each class, the top people would be grabbed to stay. I taught there for the next 26 months under excellent officers. Rank didn’t count for much. If you were the instructor, you were in charge. I also got promoted to corporal and then staff sergeant.
Q: It’s impressive that you realized you and Princeton were not a good match and acted on that.
A: I had some luck along the way, but I acted on it. This lack of self-discipline really registered on me, and I vowed to myself it would never happen again. I was a diligent and self-disciplined guy in the Army and lucky enough to be working for some really first-rate officers who appreciated it. The big lesson was self-discipline. But in 1957, I made a big mistake of going back to Princeton when my enlistment expired. At the time, I really thought I would like to do writing. I was thinking I would become a great novelist, but reality caught up with me pretty quickly. So I went to work for Burroughs, the contractor on a radar signal processor for the SAGE [Semi-Automatic Ground Environment] air defense system.
This U.S. Air Force network of computer centers produced images of U.S. airspace in real time by combining data from multiple ground-based radars. It was created in response to the Soviet Union’s first test of an atomic bomb in 1949, with the intent of increasing U.S. detection of approaching Soviet bombers. — JC
I got that job because I had technical teaching experience. It changed my life again because I rapidly picked up digital design and a working knowledge of this massive hardwired digital processor and taught classes to field technicians and engineers. In 1958, I decided I am going to have a career as an electronics engineer and find a job where that possibility exists without a degree. So I went to work in New Jersey for Stavid Engineering. I wanted to work for a small company. Six months later, they’re bought by Lockheed and become Lockheed Electronics. I spent nine years there, starting as a field engineer. In 1968, I made the big move to California.
Q: To the P-3C surveillance aircraft program, where you experienced what you call the first of three professional surprises.
A: That was the biggest good move of my career. Lockheed California had a long history with the Navy on patrol aircraft and ASW [anti-submarine warfare] programs. The analog vacuum avionics system in the P-3B aircraft was hopelessly out of date and not operationally effective. The P-3C was a new digital avionics system in which the keys to the kingdom were actually the mission software. There are two aspects on major programs to system integration. One is the technical aspects of driving the test program, identifying the technical problems and getting them fixed. Number two, equally and maybe more important, you are integrating organizations. In the P-3C, I was integrating three organizations at Lockheed, which boggled my mind when I first arrived. There was a project design organization with avionics engineering people. There was an avionics laboratory organization in another building with a totally different group. And in the flight test organization, there was a separate group. When you take that mix along with approximately 20 major avionics subcontractors, it was 50% technical integration and 50% getting that team glued together and making it stick until you got the job done. I became chief engineer [of the program] in 1974. That was the biggest surprise of my life, frankly. I spent two years as chief engineer. I got promoted to program manager, and I loved it. That got me involved in a lot of general management work in terms of responsibility for production, for financial performance, for supporting and participating in marketing activities that included winning two international competitions. I was getting a broad range of new experience both in the U.S. and these international programs.
Q: What was your second surprise at Lockheed?
A: I had not ever worked in the Skunk Works and knew essentially nothing about it. But unbeknownst to me, in late 1981 the Air Force decided that they were going to change F-117 program managers. It was in flight testing but had to transition into production and, more important, to transfer into operational use. So in February 1982, cold turkey, I became the vice president and program manager of the F-117 stealth fighter program. A lot of people wonder, “Is the Skunk Works really different?” I’ll tell you, if you go like I did — where Friday of one week, you’re in the white world and you have been for the last several years, and you wake up next Monday morning, you are in the black world and you are sitting in a conference room with [Air Force program director] Col. Richard M. Scofield — it was wonderful. It was hard work and there were a lot of problems, but we got the goddamn problems fixed. The real job was to get the production line straightened out and make deliveries. A lot of the work was dog work when you’re in a concurrent program, where you’re still in flight testing just as the initial production airplanes are being delivered and the operational unit is coming up to speed. There are a lot of changes flowing through the system. The dog work is running one of the highest speed retrofit programs you can possibly imagine. Meeting the low observable requirements, the radar cross section primarily, was not the driver. The real driver for IOC [initial operational capability] was the infrared-based passive fire control system for laser-guided bombing. We had plenty of problems with this, and the problems were not digital. The problems were the analog servos in the infrared units.
Q: Was it understood that the infrared passive targeting system was the real challenge?
A: When I arrived, it was recognized, but the source of the problem was not fully understood. Everything looks obvious in retrospect, but it ain’t obvious at the time. We figured out that the servos that did the pointing of the two units had deficiencies. The supplier was Texas Instruments, so all of this equipment out of every single operational airplane and in every airplane on the production line, all of these units had to be recycled through TI in Dallas to get these modifications installed. Most of the other problems we could fix ourselves. We could get the changes into the production line. And we set up mod capability both onsite at Tonopah [a classified test range in Nevada] and in Palmdale [California, a Lockheed facility] in a classified manner to get other changes into the airplanes. It was an intense operation. We had a very experienced workforce, and the key to a workforce is the first level of supervision. And it was excellent. I don’t think they ever got the credit they deserved. The Skunk Works had a long tradition of knowing how to manage retrofit programs because the U-2 is the most retrofitted airplane in the history of aviation. The quality of the people on that program was unmatched on any other program I worked on. These were absolutely astonishingly competent guys. I am a blunt, frank person, but I like working with blunt, frank people. And, boy, in the 117 program, you had this in spades. A lot of details worked into this. At Tonopah, everybody ate in the same mess hall. You don’t find anybody all revved up about whether your shoes are shined or crap like that. You’ve got doers. I think when people see that, it propagates, so that type of condition is self-improving.
Q: Going back to the low observables technology, tell me about the challenges of that.
A: If you go back to the ’70s when low observable technology became important, there were a considerable number of theoretical low observable technology people. The problem is they didn’t know how to build it, or the organizations they were in didn’t know how to build it. The big important factor with a Skunk Works is the ability to build stuff. That manufacturing capability is where most of the key components of the F-22 and currently the F-35 grew out of. For example, the radome technology used on the F-22 and now used on the F-35 was developed as experimental work we did on the F-117. The software was endlessly improved to compute radar cross section of airplanes. The F-117 had flat facets because the Skunk Works had developed software that could predict the radar cross section of flat facets. But by the F-22, we could deal with a fighter-shaped airplane, a real airplane. The F-117 was a real airplane, but the F-22 is a vastly different airplane and vastly more of a problem to make the computational stuff. The evolution was continuous and pretty major.
Q: Tell me how you got involved with Lockheed’s proposal for the Advanced Tactical Fighter, as the F-22 was originally known.
A: In 1985, I was asked to start working part time on the proposal. After a few months, I had a big run-in with the president of Lockheed California, Dick Heppe. I told him we were going to lose “because this program is contaminated with part-timers, including me and you.” The next day, I was program manager. In the first round of competition, we and Northrop won.
Under the $691 million contract, Lockheed built, and flew two YF-22 demonstrators. Northrop did the same for its YF-23 design. — JC
Q: For the second round, Lockheed teamed up with two other companies. What was that partnership like?
A: The toughest job of my career probably was running that three-company team with Lockheed as prime. Our partners were Boeing and General Dynamics. The essence of my job was to build a cohesive three-company team and make it stick together. And then when we got some really big, ferocious subcontractors, build them into the team and make it stick. You had three very different cultures, and they really didn’t mix very well. So the essence of the fix is letting each company realize you can’t change the other two. You got to learn to live with them. The Boeing program manager, Dick Hardy, and the GD program manager, Randy Kent, the three of us decided that if we want to win, the three of us have to stay absolutely glued together, not just at the hip, but at the brain, at the feet, everywhere. Which we did.
The U.S. Air Force awarded the joint team a $12.1 billion contract in 1991 to manufacture some 650 of the aircraft. — JC
Q: Tell me about taking the helm at Skunk Works.
A: Lockheed CEO Dan Tellep and I had developed a close working relationship on the ATF. He picked me to run the Skunk Works when Ben [Rich] retired in 1990. Even that had some interesting aspects: It’s like, “Mullin, you got the Skunk Works. Oh, by the way, you’re responsible for shutting down the Burbank plant and getting the Skunk Works relocated to Palmdale.” We couldn’t fit it into Palmdale, and it took investing $40 million in a new classified-type office building at Palmdale. By the time I retired, the Skunk Works obviously is still there, and Burbank was shut down. The unwritten policy of Lockheed, in my case and many other cases, was academic credentials didn’t count as much. What counted is how were you performing and leading as a manager and leading the organization. I didn’t know this at the time because I never really thought about it, but the Skunk Works was full of guys like me with no degree. We were kind of the campground of non-degree managers mixed in with some first-rate guys, most of whom had master’s degrees.
Q: The X-33 was discontinued after you left Skunk Works. Why do you think it didn’t work out?
A: I was involved in the early days of conceiving it and making sure we invested in it. That was probably another manufacturing step too far. In order to meet the weight objectives, you had to be able to build these lightweight composite fuel and oxidizer tanks. It turned out that was not achievable.
Mullin is referring to a 1999 test in which the outer and inner skin of the composite hydrogen fuel tank for the X-33 separated and cracked. In a 2001 report about the program’s cancellation, the U.S. Government Accountability Office wrote that “NASA did not develop realistic cost estimates, timely acquisition and risk management plans,” and therefore could not “effectively deal with the technical problems that cutting -edge projects invariably face.” — JC
Building nonmetallic tanks that will handle liquid hydrogen and liquid oxygen wasn’t successful. The question was one of weight.
Q: Do SpaceX’s reusable rocket stages reduce the potential economic advantage of single-stage to-orbit craft?
A: I’m not well informed on that. To me, SpaceX is not getting credit for one of the great accomplishments in aerospace of recent times. The idea that you can successfully reland these rocket boosters is phenomenal. The other thing that doesn’t get proper credit is the GPS and the other precision electronics in this vehicle. It amazes me the reliability of this system, where you can have the data and the computational ability on board to have a guidance and control system to do this precision landing. Without GPS, it would have been hopeless.
Q: How should recent engineering graduates learn in the real world?
A: There’s no capability in the time that’s available for getting a bachelor’s or a master’s degree. The only way you gain this knowledge — and this is true of basically any engineering discipline — is by doing it. It cannot be conveyed. There are some things you only learn by doing it. It’s that simple.
Q: Do companies do a good job training new engineers?
A: I think they do. I can only speak from my own experience hiring new graduates. If you’re doing it right, you accept that before a new graduate engineer becomes really effective, he needs to work on two or three projects and needs to be moved around. This was particularly true in the Skunk Works. They understood that. When you have a new grad in their first job, nobody is going to last very long. I don’t mean in terms of employment. I mean, you’re going to come to work there and some new opening is going to come up. And you’ve now impressed a few people and they figure, “Give this guy a shot at this more challenging job.” And you go from there indefinitely, you go from there. There’s a big problem teaching program management. I really believe that the only way you learn program management is to work for a good program. Management is very hard to convey, and I happened to work for two really outstanding ones at a certain point in my career. I also worked for some crummy ones.
Q: Another theme you have emphasized is the importance of writing well and being able to communicate. It sounds like it’s been a continuing problem with a lot of engineers.
A: There is no question that oral and written communications are really important from a career viewpoint. This is an egotistical statement: I was considered an A-plus writer mainly because I could write. Also, I was involved in a lot of proposal efforts where finally people figure out, “Hey, this guy, he actually knows how to write or edit a proposal.” Or conversely, “He also knows how to communicate with customers and is bi-directional: He knows how to listen and he knows how to convey information.” One of the things we did in my years in the Skunk Works both on the 117 program and elsewhere, but particularly on the ATF competition, is we focused to make sure our people who are capable get customer interface experience. You get them started doing dry runs and stuff like that. Some of them turn out to be absolutely fantastic.