Cyclotron Info
On this page

How to Learn to Build a Cyclotron

The useful literature of small-cyclotron building is compact. A handful of construction-era reports, one standard textbook, four thoroughly documented amateur-scale machines, and two adjacent communities cover most of what a builder draws on; across the documented machines in the builds census, the same short list of sources recurs in build after build. That makes a study path practical to write down. This page gives one, in three tracks. Start the first; open the second after the first two stages; the third's safety and legal reading comes before any purchase. And if you have not yet read the case against building one at all, start there — this page assumes you have and are still here. It is the external companion to the reading order on the project-planning page, which routes through this site; this page routes through everything else.

Looking for a career path rather than a machine? That is a different sequence; see self-study for accelerator careers.

Track 1: the reading sequence

Four stages, each of which changes what the next one means. Skipping ahead works badly here: the Design Guide reads as a wall of disconnected facts until the physics and at least one worked design are in place.

  1. How a Cyclotron Works, then Choosing Your Machine: the resonance condition and what breaks it, then what each energy class costs and buys. An evening each.
  2. Wouters, General Recommendations for Design of Small Cyclotrons (UCRL-476, 1949): the nearest thing this shelf has to a design cookbook for small machines, written at Berkeley to pass on what the early labs had learned. Forty pages; the physics has held up, the 1949 component engineering has not. Hosted here with its public-domain rationale and a reading guide keyed to its page numbers.
  3. Livingston & Blewett, Particle Accelerators (1962), chapters 5 and 6: particle motion in fields, then the cyclotron itself — the standard engineering reference, by authors who built the early machines. Still in copyright, so it is cite-only in the library; used copies are common. Read those two chapters, then return to the rest as reference; cover-to-cover is a poor first pass.
  4. The Design Guide, by subsystem, while designing: not a book to read through. It is the cross-index into the two sources above and a hundred more: when the magnet is on the bench, read the magnet rules; each rule carries its verbatim quote and page citation, so it also serves as a guide to which pages of the primary sources matter for the decision in front of you.

Track 2: four working machines

Theory tells you what should happen; build records tell you what happened. Among the census machines with documented beam, four carry records deep enough to study rather than merely admire, and the four records complement each other: a video course, a chain of theses, a photographed build log, and a published book. Read at least two of them end to end — not skimmed for the parts that look relevant — before committing to a design of your own. They stall and recover in different places, and the differences are the lesson; for each build, note what stalled, what the fix cost, and what you would copy or avoid.

  • The Rutgers 12-inch and its lineage: begun as an undergraduate project and continued at the University of Maryland, this is the longest-running amateur-origin effort in the census, and the best taught: the Cyclotrons! video series (Koeth Research Group with USPAS, nine episodes) walks the real machine subsystem by subsystem, and the project site documents the hardware. Census entries: Rutgers 12-inch, Maryland 19-inch.
  • The Houghton College cyclotron: a working machine built and rebuilt by successive student generations under one program, and its record has no counterpart in the census: each stage is written up as a thesis in an open repository, so the reasoning behind every change survives, including the changes that failed. The deepest prose record of the three. Census entry: Houghton College.
  • The Mullins cyclotron: a father-and-son garage build carried from magnet design to beam, with a run log spanning 2018 to 2025, documented in a build log with less text than the Houghton theses and more photographs: what each subsystem looks like as actually fabricated, at the scale of a home shop and budget. Census entry: Mullins Cyclotron.
  • The COLUMBUS school cyclotron: built 2012–2014 as a school project in Coburg, Germany, with a donated laboratory magnet and expertise from Forschungszentrum Jülich, then run for a decade as a teaching instrument at Hochschule Coburg. Its record documents teaching with a cyclotron as much as building one: an open conference-paper series runs through the project, from the 2013 design paper and the ten-year status paper (the library indexes the full series), and the builders published a short book on the machine as operated (Prechtl & Wolf, Das Lehr-Zyklotron COLUMBUS, Springer 2020). One asterisk: the book is in German and commercially published, so the freely readable layer is the paper series. Study it for what it chooses to measure — the standing experiment identifies which particles circulate from their specific charge and deliberately evaluates no beam current or energy — and for the builders’ stated reason for not extracting the beam, that no radiation can leave the experiment, a position they have since begun to revisit with an extraction system. Census entry: COLUMBUS Coburg.

The builds census holds the wider field, including the instructive failures; the What the census teaches section is the digest.

Track 3: practice and people

A cyclotron is a vacuum and high-voltage project before it is an accelerator, and a radiation project from the first time high voltage or RF is energized: X-rays do not wait for beam. Each of those skills can be practiced on cheaper hardware first.

  • Safety and legal pages before any purchase: the classic reports predate modern radiation regulation and community threads on these topics are anecdote, so this layer has to come from current sources — start with those pages, then your state radiation-control program, which is the actual authority.
  • Fusor.net: the amateur fusion community, with deep hands-on vacuum, high-voltage, and radiation-measurement craft, and a history of cyclotron build threads in its advanced-projects section.
  • The Guild of Cyclotron Builders: a private Facebook group of people who have built or are building machines — the one dedicated venue, but closed and unarchived; fusor.net is the open room.
  • The calculators and free modeling tools: as each subsystem gets real. Running FEMM on a published magnet before designing your own is study and practice at once.

Going deeper: the professional layer

Everything above stays at amateur scale. When a question outgrows it, the professional teaching material is free:

The full annotated list, including everything this page left out and why each entry is there, is on the external resources page.