W2DU’s Son Read My Article—Then Corrected the Historical Record
W2DU’s Son Read My Article—Then Corrected the Historical Record
Rick Maxwell read my article about his father’s work, appreciated the analysis—and supplied the laboratory history I had left incomplete. His letter deserves to be read as part of the story.
Every technical writer hopes to hear from a reader who can add something the published record missed. In this case, the reader was the son of one of amateur radio’s most widely read transmission-line authors.
After reading my article Reflections Revisited: What’s Still True—and What’s Been Refined—Since W2DU, Richard A. “Rick” Maxwell, W8KHK, contacted me. Rick is the second son of M. Walter Maxwell, W2DU. He explained that his callsign had been his father’s original call in 1933, when Walt was 14. ARRL’s biography of Walt independently records the original W8KHK call and Rick’s later adoption of it.
Rick was not asking me to stop examining his father’s work critically. He appreciated the comparison between earlier and modern measurement methods. His correction was more specific: my list of couplers, bridges and oscilloscopes described Walt’s early work, but did not describe the laboratory he developed over the decades that followed.
The missing part of the history: Walt’s later bench included vector instruments capable of measuring magnitude, phase and impedance. Describing his entire body of work through the equipment of its earliest period gives readers an incomplete picture.
Rick gave me explicit permission to publish his message and invited minor corrections for clarity and repetition. The letter below has been lightly edited for spelling and flow. Its first-hand laboratory and family details are Rick’s account; the technical qualifications following it are mine.
The Letter That Added the Missing Years
Hello Joeri,
I enjoyed reading your analysis of the Reflections publications by my father, M. Walter Maxwell, W2DU.
I am his second son, Richard A. Maxwell, now W8KHK—which was Dad’s original callsign in 1933, when he was 14 years old.
I very much appreciate your analysis and review of his findings, and your comparison of what can be observed now with state-of-the-art equipment with what he employed in his earlier work.
You wrote that Maxwell’s tools included couplers, SWR bridges, directional couplers and oscilloscopes rather than network analysers. That is an accurate list for his earlier work, but his laboratory resources were upgraded over the long period that followed the start of this work in the mid-1960s.
Early on, he used rudimentary directional couplers and bridges. For better accuracy, he soon used the General Radio 1604 bridge; HP 606, 608 and 612 generators; a slotted line; the PRD 219; and an HP 415 meter.
When they became more affordable, during retirement soon after 1980, he acquired an HP 8640A generator, an HP 8405A vector voltmeter and its associated directional couplers, along with an HP 4815A vector impedance meter. It was with these instruments that he was able to validate the conjugate-match theorem with linear generators used in amateur-radio applications. He also acquired a Tektronix spectrum analyser with a tracking generator.
It is true that non-linear generators change the landscape and the results of matching calculations. However, when most of his work was done, vacuum-tube final amplifiers with traditional tank circuits or π-network coupling networks were commonplace, while solid-state finals with broadband filter networks were beginning to displace the earlier tube equipment. Class-D and class-E finals and SDRs were future innovations, as you mentioned.
Although HP vector network analysers were in use at the RCA Astro Electronics Division, he had no access to those resources after retirement. I believe it is reasonably safe to assume that the many sequential measurements made with the vector voltmeter and vector impedance meter could achieve results equivalent to those obtained with a vector network analyser, although with considerably more time and effort. The resulting Smith Chart plot would still be essentially identical.
I also appreciate that you mentioned the fact that many of the calculations assume “lossless networks”. These are, of course, purely theoretical and do not exist in practice. Unfortunately, many readers gloss over this and then question the power relationships when applying the conjugate-matching theorems referenced in the manuscript.
Joeri, I believe your document is clear and concise, complimentary to the work my father did, and helpful in allowing readers to view his work from the perspective of the tools available then and now. To make it more historically accurate, you might consider expanding the list of instruments he used over time. The later HP and Tektronix equipment provided a much more accurate and professional validation of his earlier work.
It appears to me that his intended guidance to amateurs, concerning practical coupling and matching techniques for this class of communications, retains lasting value for most readers.
Thank you again for taking the time to write your analysis. I am quite sure Dad would be pleased if he were able to read it today.
Warm regards,
Rick Maxwell, W8KHK
ex-WB4GNR, WB2HKX, AFC2HKX
What Rick’s Correction Changes
The important correction is mine. Walt’s early bridges and couplers were not the end of his laboratory story. According to Rick, the later HP vector voltmeter, vector impedance meter, generators and Tektronix equipment allowed a much more capable investigation than my short instrument list suggested.
That does not require treating family recollection as an independently audited inventory. It requires naming the source and giving the recollection its proper place. Rick’s account supplies the chronology; contemporary instrument documentation helps explain what those instruments could actually do.
It also changes the tone of a fair comparison. “Then versus now” must not quietly become “crude equipment versus real measurement.” Careful vector measurements were possible before an automated VNA put an entire sweep on a screen. Automation, convenience and uncertainty are related questions, but they are not the same question.
What the Named Instruments Could Measure
An HP 8405A vector voltmeter with suitable directional couplers can compare RF voltage magnitude and phase. An HP 4815A vector impedance meter can measure complex impedance. Hewlett-Packard’s January 1967 Journal comparison of impedance-measuring systems documents those different capabilities. Sequential readings can reconstruct impedance and reflection relationships when calibration, loading and reference planes are controlled.
That does not make every sequence automatically equivalent to a calibrated vector-network-analyser result. Directivity, source match, tracking, connector repeatability, fixture behaviour and error correction determine the uncertainty. A similar Smith-chart trace can be assembled from sequential data, but equivalence belongs to a declared measurement method and uncertainty budget—not to the instrument names alone.
When Rick describes Walt validating the conjugate-match theorem with linear generators, I read that as experimental support for its predictions in the tested arrangements. Agreement between calculation and observation matters. It does not by itself establish the same behaviour for every transmitter topology, source impedance or nonlinear operating state.
None of those measurement limits changes Rick’s central correction: Walt’s later work was supported by much more than a simple SWR bridge. A fuller instrument history strengthens a fair technical reading of Reflections.
Linear Matching and Later Transmitter Topologies
Rick’s letter also places the work in the era of vacuum-tube finals with tuned output networks and the increasing use of solid-state finals with broadband networks. The timeline needs one distinction: Class-D and Class-E cannot both be treated as inventions that only arrived after Walt’s research period. Nathan and Alan Sokal’s Class-E paper was published in June 1975 and discusses earlier Class-D circuits. The development of a circuit principle is not the same event as its later widespread adoption in amateur equipment.
The source model still matters. Conjugate-match and available-power relationships are precise for the stated linear network assumptions. A practical transmitter may include protection, output filtering, control loops, compression or switching behaviour that changes the relationship between terminal impedance and delivered RF power.
Read Maxwell Again with the Fuller Story
Rick’s letter gives Reflections Revisited a fuller historical foundation. The useful reading path is not to choose between respecting Maxwell and testing his assumptions. It is to do both: understand the instruments and transmitter context, then distinguish a conjugate relationship at a stated plane from a claim that everything is matched everywhere.
The related articles above follow that path through conjugate matching and RCA’s wider engineering history. Rick’s contribution belongs alongside them because it tells us how the work was investigated—not just which equations appeared in print.
A Conversation That Continued Beyond the Letter
My correspondence with Rick did not end with that first message. In the exchange that followed, he shared further links, memories and technical leads: Walt’s later writing, family amateur-radio history, vintage transmitters and the practical laboratory culture surrounding the work.
Rick also kindly offered to read relevant manuscripts before publication. That offer, and his willingness to share what he remembered, gave the conversation a value beyond a single correction. Those leads deserve the same combination of first-hand attribution and careful engineering analysis.
A Record Made More Accurate
The best outcome of technical publishing is not being treated as the final authority. It is creating a record that knowledgeable readers can test, refine and improve.
Rick Maxwell’s message does exactly that. It preserves the central value of W2DU’s work while expanding the picture of how that work was tested over time. For that contribution—and for the trust involved in sharing it publicly—I am deeply grateful.
Mini-FAQ
- Why did Rick Maxwell contact me? After reading Reflections Revisited, Rick appreciated the analysis and supplied the later laboratory history missing from its account of W2DU’s instruments.
- Who is W8KHK? Richard A. “Rick” Maxwell is the second son of M. Walter Maxwell, W2DU, and holds his father’s original callsign.
- Is Rick’s letter published with permission? Yes. Rick gave explicit permission to publish the message and make minor corrections for clarity. The laboratory recollections remain attributed to him; the technical qualifications outside the letter are mine.
- Could Walt’s vector instruments produce useful results without an automated VNA? Yes. Sequential magnitude, phase and impedance measurements can establish vector relationships. Comparison with a VNA requires controlled calibration, reference planes, loading and an uncertainty budget.
- Why qualify the letter’s Class-E timeline? The Sokal Class-E paper appeared in June 1975 and discussed earlier Class-D circuits. Publication of those principles and their widespread adoption in amateur equipment are different historical events.