--- title: "ORNL-2644 — Cyclotrons and High-Energy Accelerators, 1958 · Cyclotron Info" canonical: https://cyclotroninfo.com/library/classics/ornl-2644/ description: "Hosted public-domain copy of Howard's 1958 ORNL-2644 world accelerator compendium — construction parameters for 140 machines — with a guide to using it." generated: 2026-11-12 note: "Markdown rendering of the canonical HTML page, generated at build for AI assistants and text tools. Figures and interactive elements are described, not reproduced. On Design Guide rules the source quote, this site's paraphrase and this site's editorial note are labelled separately and must not be conflated." licence: "Datasets CC BY 4.0, census snapshot CC BY-NC 4.0 (https://cyclotroninfo.com/data/); cite as Cyclotron Info, cyclotroninfo.com" --- [Construction Classics](https://cyclotroninfo.com/library/classics/) # Cyclotrons and High-Energy Accelerators — 1958 Howard, F. T. *Cyclotrons and High-Energy Accelerators — 1958*, ORNL-2644. Oak Ridge National Laboratory (operated by Union Carbide Corporation for the US AEC under contract W-7405-eng-26), November 17, 1958. Record: [OSTI 4286826](https://www.osti.gov/biblio/4286826) [Download PDF — 18.6 MB, 317 pages](https://cyclotroninfo.com/classics/ornl-2644.pdf) ## Why this document can be hosted ORNL-2644 is an Oak Ridge National Laboratory report produced by Union Carbide for the AEC under contract W-7405-eng-26 — both statements are printed on its title page. AEC-issued contractor reports are treated as US Government works, which carry no copyright; pre-1978 distribution without notice is an independent public-domain basis, the same rationale as the hosted ORNL-1196 and ORNL-2648. Verified against the title page and the OSTI record; determination made August 2026. ## How to use this compendium This is not a report to read; it is a database to query — 140 accelerators worldwide, verified questionnaire data for 108 of them, collected by ORNL in mid-1958 and printed as standardized data sheets. For anyone trying to place a design in context, it answers one question no single machine report can: what did everyone else choose? Page references are the report’s printed numbers; the hosted PDF runs seven pages ahead. Entry is through the General Directory (pp. 3–14), organized by country, then institution: each line gives machine type, pole diameter, energy, particle, status, and the page of its data sheet. The sheets themselves are grouped by machine class — fixed-frequency cyclotrons from p. 15, frequency-modulated machines from p. 171, synchrotrons and linacs from p. 209 — and every machine gets the same two-page form. The first page is design data: magnet (pole-tip and core diameter, gap, peak field, iron weight, winding, cooling), RF (dees, frequency, dee voltage, oscillator input), history and cost. The second is measured performance: energy, frequency, field, dee-to-dee kilovolts, stable and peak beam currents internal and external, an “unusual features” note, and — the quiet treasure — the published articles describing the machine, which make each sheet a doorway into that machine’s literature. Three worked uses. First, sizing by precedent: to see every machine near a given pole diameter, sweep the directory’s size column — the fixed-frequency section spans Melbourne’s 39-cm variable-energy machine (p. 21) to the largest classical cyclotrons, so a small-machine builder can find the cluster of sub-half-meter designs and read exactly what fields, dee voltages, and beam currents they achieved. Second, sanity bands: collecting dee-to-dee voltage against energy across the fixed-frequency sheets gives an empirical envelope for judging whether a proposed RF budget is ambitious or routine. Third, survey statistics: the summary table (p. 1) counts the 1958 fleet — 76 fixed-frequency cyclotrons, 18 FM machines built, 50 machines then under construction or design — the baseline snapshot of the field at the AVF transition. Howard’s foreword states the data discipline: questionnaire returns were edited only for terminology, and unverified entries are flagged as such. The hosted copy is an OSTI scan with a text layer that is searchable but noisy on the tabular pages — numbers OCR badly — so use the text layer to find a machine and read the values from the page image. ## Design rules extracted from this document [13 Design Guide rules](https://cyclotroninfo.com/design-guide/?q=ORNL-2644) cite ORNL-2644 — census-population envelopes rather than single-machine numbers: what field, dee voltage, RF power, geometry, and budget the 1958 fleet of small machines actually converged on. A pass feeding this site's census records with historical entries remains queued. Representative examples: Used by the [beam dynamics](https://cyclotroninfo.com/learn/beam-dynamics/) deep dive and the [industry](https://cyclotroninfo.com/learn/cyclotron-industry/) reference. - [dg-697](https://cyclotroninfo.com/design-guide/#dg-697) — every working small classical cyclotron ran 12.5–18.5 kG center field - [dg-700](https://cyclotroninfo.com/design-guide/#dg-700) — magnet iron grows roughly with the cube of pole diameter - [dg-701](https://cyclotroninfo.com/design-guide/#dg-701) — census geometry template: gap, dee aperture, and dee diameter proportions - [dg-702](https://cyclotroninfo.com/design-guide/#dg-702) — shim for only 2–4% total field drop-off, center to maximum beam radius - [dg-709](https://cyclotroninfo.com/design-guide/#dg-709) — whole working cyclotrons were built for $5k–$110k in about two years > **Page provenance.** Last reviewed: 12 August 2026 · Hosted copy verified against the OSTI record; rights review August 2026 · [Methodology](https://cyclotroninfo.com/about/methodology/) · [Report a correction](mailto:info@cyclotroninfo.com?subject=Correction%3A%20ORNL-2644%20%E2%80%94%20Cyclotrons%20and%20High-Energy%20Accelerators%2C%201958) · [Changelog](https://cyclotroninfo.com/changelog/)