Cyclotron Info
On this page

Amateur & teaching machines

The library classifies its documents by era and kind, which scatters the amateur and teaching-machine record across three of its four sections. This page cuts the other way: everything the small-machine community wrote about its own machines — 50 documents across 8 programs — grouped by machine, each section linked to its builds census entries. Machines with no indexed documents appear in the census only; the technique literature the community leans on (vacuum, RF, magnet engineering) stays in the main library's reference sections.

Documents link to their publishers' or authors' own copies where a verified link exists, with Internet Archive fallbacks shown as “archived” badges once a snapshot is verified — the same preservation practice as the census and library citations. Each document's library entry carries its annotation and scores. One series — the Mullins Mark I notes — is hosted on this site directly; its section states the terms.

The Niell cyclotrons

Census: Niell Cyclotron I (c. 1993–1994) · Niell Cyclotron II (1994–1995)

Two machines built at home by a high-school student, every subsystem — magnet, RF amplifier, vacuum gauges — his own construction. The first demonstrated cyclotron mass resonance on residual air; the second, with a variable-frequency RF system, matched hydrogen and helium resonance predictions to within about 2%. The document indexed here is the builder's later conference talk on equipping a laboratory on a budget; the machines themselves are documented on the builder's site, cited from the census entries.

  • Effective Scientific Equipment Procurement Strategies: Building on a Budget

    Fred Niell (Small Cyclotron Conference 2010) · 2010 · library entry

Rutgers 9-inch, 12-inch, and Maryland 19-inch

Census: Rutgers 9-Inch Cyclotron (prototype) (1995–1999) · Rutgers 12-Inch Cyclotron (1995–present) · University of Maryland 19-Inch Cyclotron (c. 2016–present)

From a basement nine-inch prototype begun in 1995 to the Rutgers teaching machine to the Maryland 19-inch successor: the best-documented amateur-origin cyclotron in this site's census. Its papers — subsystem memos, conference papers, neutron-physics reports — have a dedicated reading guide, ordered as the machine earned them.

The program's 20 indexed documents have a dedicated reading guide, with links to the authors' hosted copies.

Houghton College Cyclotron

Census: Houghton College Cyclotron (2001–present)

A deliberately small cyclotron run as a continuing undergraduate research program since 2001, each student generation documenting its improvements in theses and conference talks — the college repository holds roughly twenty. The indexed set runs from a 2002 permanent-magnet design study through the 2015 focusing thesis.

MIT 2 MeV Cyclotron Project

Census: MIT 2 MeV Cyclotron Project (c. 2006–2007)

An undergraduate thesis on the design and construction of a cyclotron intended to reach 2 MeV around a surplus NMR electromagnet. No demonstrated beam is documented and the project appears not to have been completed, so the thesis's numbers are design targets rather than measurements — read it for the planning approach, not the values.

Cyclotron Kids 2 MeV Cyclotron

Census: Cyclotron Kids 2 MeV Cyclotron (c. 2006–2013)

Two high-school students who conceived a 2 MeV proton machine and were invited to build it at Jefferson Lab. Their master design document covers the full machine, from proton energy requirements to target plans; the conference talks carry the project through 2013. All components were completed, but lab safety rules for minors prevented power testing, and no beam was ever run.

COLUMBUS School Cyclotron

Census: COLUMBUS School Cyclotron (2012–present)

A school cyclotron built at the Gymnasium Ernestinum in Coburg with expertise from Forschungszentrum Jülich, now a teaching instrument at Hochschule Coburg. The indexed papers run from the 2013 design pair through the 2022 status paper, including the short Springer book that documents the machine as operated rather than as designed.

Kwolik St. Cloud Cyclotron

Census: Kwolik St. Cloud Cyclotron (by 2015 (construction years undocumented))

An unfinished 9.25-inch build known from the builder's 2015 photo tour, preserved as a web capture: a 1,080 lb magnet with machined pole tips, an oil-cooled RF matching network, and a pump stack the account claims reached 10⁻⁹ torr — still awaiting its 3 kW RF supply where the record ends.

  • Building A 9.25" OD Cyclotron (LinkedIn Pulse post)

    James Kwolik · 2015 · library entry

Mullins Cyclotron

Census: Mullins Cyclotron (2017–present)

A father-and-son basement build carried from magnet design to a working machine between 2017 and 2019 and operated since, documented on a dedicated project build log covering the magnet, vacuum system, RF chain, and ion source. The technical-note series below, contributed by the builders, extends that log with the as-built record.

Project build log archived Aug 2026

Notes on the Mullins Mark I Cyclotron

8 notes, hosted on this site with the builders' permission; each is licensed CC BY 4.0 and carries its author, revision, and license on its title page.

Simulation model packages

4 downloadable packages (CC BY 4.0, same terms as the notes), documented by the series' simulation-models note; each unpacks to a folder with a README stating what the model is, how to run it, and results against measurement. Solver outputs are excluded and regenerate from the included scripts.

  • FEMM design models, 2018–2020 ZIP · 408 KB

    The builder's original planar 2-D H-frame design models and Lua sweeps at the 1.5 in design gap.

  • FEMM as-built rerun ZIP · 43 KB

    Parametric rebuild at as-built dimensions with a 10–55 A current sweep, field-profile runs, joint-gap variants, and comparison against the measured B(I).

  • FEMM electrostatic dee model ZIP · 1.1 MB

    Electrostatic sections of the dee between the lids: accelerating-gap field, peak edge field, and dee-body capacitance against the machine's 74/100 pF record.

  • Elmer 3-D H-frame model ZIP · 77 KB

    Three-dimensional H-frame magnet model in Elmer 9.0 with nonlinear 1018 steel, validated by ideal-iron and no-iron controls; lands 2–6% above the machine's Hall calibration.