This spring, we began a new awards program—the “Best of Microwaves.” While most of the categories reflected how well companies communicate with engineers and the resources and intellectual property (IP) they make available, we also named our first “Living Legend.” This award was inspired by our online hall of fame, dubbed “Microwave Legends.” It occurred to us that we could give an in-person award of the same caliber to someone who has greatly impacted the microwave industry. Our first “Living Legend” award went to retired Agilent Technologies employee Blake Peterson.

Shortly before his discharge from the US Navy in November 1956, Blake accepted a job offer from Hewlett-Packard Co. (HP). He spent his first year in production, followed by a stint as a technical writer, where he authored the first-ever HP schematic diagrams using transistors in place of vacuum tubes. He soon realized that his specialty was technical support and spent his career churning out application notes, training programs, and teaching courses. Herewith, an interview with Blake, who is as nice a person as one could ever hope to meet:

NF: How long was your career at HP/Agilent?

BP: Almost 45 years. HP/Agilent could not have been a better employer. I received more than one call from head hunters, and I told them that I had absolutely no interest in leaving HP/Agilent.  

NF: You created renowned Application Note 150, “Spectrum Analysis Basics.” Did you have any idea that it would make you famous in the microwave industry? Are you surprised that people have used that application note as a reference for so long?

BP: I had no idea what impact the app note would have, and I am indeed surprised that it is still so popular. However, basics—real basics—are needed as building blocks for understanding more advanced technology.

NF: You also helped to create the support materials for the 853A/8553A spectrum analyzer, which heralded the beginning of modern spectrum analysis. What was it like to work on that project? Did you find that engineers needed extra training or had to think in a new way about using that instrument?

BP: I was managing the technical writers who created the operating and service manuals when this analyzer was introduced. Not only was this analyzer a new product, it was HP’s entry into a new area of test and measurement—an area in which the tech writers had no experience to draw on. So I wrote a good part of the manual and laid out the schematics for it. Producing that manual was indeed a challenge. HP would not ship a product without a manual, so we had to get it done on time.

NF: When personal computers (PCs) came along, you recognized how computer-based tools could help the design process and quickly began working on them. How many did you create in total? Which do you think had the most impact?

BP: When interviewing new EE graduates, I found that most could not explain technical processes in simple terms. For example, most were familiar with how an amplitude-modulation (AM) radio works, but had no idea how a basic spectrum analyzer works. I would ask: What do you get as you continuously tune your radio? Loudness versus frequency. Then how do we convert an AM radio into a basic spectrum analyzer? Replace the speaker with a calibrated screen, get rid of automatic gain control, put in a sweep generator to tune the local oscillator, and replace the antenna with an input connector. Now we have a visual picture of loudness (amplitude) versus frequency and can easily examine a signal’s bandwidth, distortion, etc.

I wrote some 50 animated programs to provide insight into spectrum analysis, network analysis, and various communication schemes. I included a program that shows the relationship between rotating vectors and sine waves. It could be used to show the impact of non-uniform group delay. You can set the phase relationships among the frequency components at the input of a device and see what waveform results. Since non-uniform group delay means different transit times for the various frequency components, their phase relationships are different at the output of the device. By setting the new phase relationships and rerunning the program, you can see the impact of the non-uniform group delay.