Construction Classics

Design and Construction of Synchro-Cyclotron

Why this document can be hosted

NYO-780 is a joint ONR-AEC contract report — the title page carries “United States Atomic Energy Commission / Office of Naval Research” and “Work performed under Contract No. N7onr-303” — issued and sold (at 30 cents) by the AEC’s Technical Information Service. AEC-issued contractor reports are treated as US Government works, which carry no copyright, and the pre-1978 distribution without copyright notice stands as an independent public-domain basis. This copy is the report itself, not a journal reprint. Verified against the title page and the OSTI record; determination made August 2026.

Reading guide

This is the record of one institution building one synchrocyclotron from scratch — magnet models to shielding doors — compressed into fifteen sections. Creutz opens with a candor rare in final reports: the machine was not finished, so “this report might better be termed a status report” (p. 1). What was finished, and is documented here, is nearly all of the construction. Page references are the report’s printed numbers; the hosted PDF runs five pages ahead.

The magnet chapter (pp. 2–5) is the collection’s best account of model-magnet methodology: a rough 6-inch model, then a 2-inch model “changed quickly and cheaply” to sweep parameters, then two 9-inch models, then a final model machined from the very forgings used for the full-scale poles. The sweeps found the coil height that minimizes combined power-plus-steel cost, proved shimming workable at 20 kG (2.0 T), and produced a pole contour holding 96.7% of central field at 96.5% of pole radius. Engineering design (pp. 6–14) then scales the result to 1500 tons of steel and 130 tons of copper: pole machining held to ±0.0025 inch, and an oil-cooled coil wound at the Brooklyn Naval Shipyard from sixteen parallel copper bars per turn, insulated with fishpaper and pressboard. Vacuum (pp. 15–19) reports the payoff of helium-mass-spectrometer leak testing every assembly after manufacture: two leaks total at first pumpdown, both in gaskets, and 3 × 10⁻⁵ mm Hg reached in 2.5 hours.

The oscillator section (pp. 23–31) is an honest development narrative — parasitic modes, a band gap near 25 Mc traced to the condenser teeth acting as a lossy resonant line, and the admission that the three-quarter-scale model “could serve as little more than a guide” to full-scale behavior. The rotary condenser gets its own section (pp. 32–38), including why half the ground capacitance sits in end-ring condensers to kill an axial transmission-line mode. Shielding (pp. 38–42) describes magnetite concrete mixed to 200 lb/ft³, one wall loaded with 320 tons of surplus Navy projectiles, blocks stacked in staggered double layers against straight-through cracks, and 80-ton hydraulic doors interlocked to the oscillator and field. Cooling (pp. 43–50) and the homopolar generator with bus bars built from coil-winding scrap (pp. 51–53) close the construction story. Section XV (pp. 67–75) condenses the whole machine into parameter tables, from the 15.615-inch pole gap to the pump-down schedule.

For a small-machine builder the transferable content is the method: models before metal, cost curves before forgings, leak-test before assembly, and a written trail — the report cites its own progress reports throughout. The hosted copy is an OSTI scan with an ABBYY text layer that is searchable but rough, so treat text extraction as an index, not a transcript.

The engineering behind this report is also here. TID-454 collects seven Carnegie technical reports written under the same contract — N7onr-303 — on magnets, vacuum, frequency modulation, RF and extraction timing. Where Creutz records that the model sweeps found the coil height minimizing combined power-plus-steel cost, TID-454’s first report is that cost optimisation worked out in full, unit costs and nomograph included. NYO-780 is what was built; TID-454 is why it was built that way.

Design rules extracted from this document

23 Design Guide rules cite NYO-780 — the largest yield of the wave-2 recovered documents, most of it magnet and vacuum practice. Representative examples:

• dg-635 — survey magnet parameters on cheap small-scale models before freezing the design
• dg-637 — shim the pole contour to 96.7% of central field at 96.5% of pole radius (n ≈ 0.2)
• dg-648 — helium leak-test every vacuum subassembly before installation
• dg-651 — suppress parasitic oscillator modes with light-bulb loads and a series-resonant trap
• dg-654 — interlock access doors to the oscillator and the magnetic field

Used by the magnet design deep dive.