The tuning reserve: why founders cast bells too thick on purpose

01

Why founders leave metal on purpose

A bell's tone comes from a set of partial frequencies, led by the hum, prime, tierce, quint, and nominal, whose relationships to one another are controlled largely by the bell's thickness, especially at the soundbow. Getting that relationship exactly right from the mould alone has never been reliable, even for skilled founders. A conservation account of English bellfounding records that most early founders knew a bell's shape and thickness affected its sound, and were careful to keep the mould uniform, yet remained mystified as to why many technically good castings still sounded discordant ². Acoustic research now explains why: the low partials do not fall into musical relationships automatically, and getting them there depends on precise thickness control that trial-and-error moulding alone could rarely guarantee ¹.

The practical answer to that uncertainty is the tuning reserve: casting a bell's wall a little thicker, and its weight a little heavier, than the acoustic target requires, so a tuner has real material to work with afterward. The reserve is not a hedge against a founder's incompetence, it is a rational response to a genuine limit, that no mould, however carefully built, guarantees a perfect result on the first and only pour a bell will ever have.

02

One-way metal, one-way decisions

Tuning removes metal, and only metal removal, from a cast bell. As bell acoustics researcher Bill Hibbert puts it plainly, a tuner cannot stick metal back on, so if a bell's underlying shape is wrong, too tall, too short, or thin in the wrong places, there is little tuning alone can do to fix it ¹. That constraint is why every cut a tuner makes is effectively permanent, and why the reserve exists in the first place: it converts an irreversible operation into a correctable one, but only up to the amount of extra metal the founder chose to leave.

The tools available to make that cut changed dramatically over the trade's history. Before the close of the 19th century, tuning was done with a hammer, chisel, and file, and generally reached only the soundbow and lip, where a founder could chip away at the rim to raise a bell's strike note in a practice sometimes called skirting. Large vertical boring machines introduced late in that century let a tuner mount a bell on a rotating lathe and cut metal precisely from almost any point on its interior, and sometimes its exterior, transforming tuning from a blunt correction at the rim into a controlled shaping of the whole interior profile ².

03

What tuning actually buys

Modern tuning on a vertical lathe pursues two distinct goals. The first is internal: bringing an individual bell's own partials, its hum, prime, tierce, quint, and nominal, into a consistent, ideally octave-related, relationship with each other, a standard known as true-harmonic or Simpson tuning after the researcher who formalized it in the 1890s. The second is external: adjusting the nominal, or strike pitch, of each bell in a ring so the whole peal sits on a musical scale together. Because tuning removes metal and lowers pitch, a peal is tuned down to match its flattest bell rather than up to match its sharpest, since sharpening a bell beyond modest amounts risks the same over-thinning that spoils tone ¹.

The Hemony brothers solved true-harmonic bell shape and individual partial tuning in the Low Countries in the 17th century, using a tuning machine and chime bars of their own devising, but the knowledge died with them and was not recovered elsewhere for over two hundred years. It took until the 1890s, with Canon Arthur Simpson's published analysis and Taylor's bellfoundry's near-simultaneous experiments with tuning forks and a purpose-built machine, for true-harmonic tuning to become a repeatable, teachable practice rather than a lost secret ¹.

04

Radical tuning and the ceiling it cannot break

Bell founders describe removing 10 to 15 percent of a bell's total weight as radical tuning, and even that scale of correction will not necessarily make an old-style bell true-harmonic if its underlying shape is wrong to begin with, because tuning can only rebalance partials within the range a given shape allows ¹. A documented 2002 case at Lyminge, Kent, illustrates the ceiling concretely: the bellfoundry Whitechapel removed about 10 percent of the weight of the fifth bell in a ring of eight, a 1785 casting by Mears, and its hum moved from roughly two semitones sharp of true-harmonic to about three-quarters of a semitone sharp, a genuine and audible improvement. Its prime, however, stayed at roughly three-quarters of a semitone flat throughout, unmoved by the same tuning pass ¹.

That asymmetry is the point: tuning corrected what the bell's shape allowed it to correct, and left untouched what the shape did not allow. A bell cast with the wrong proportions cannot be argued into a different shape family by removing more metal, it can only be pushed closer to the best result that shape family permits, and pushed too far, it stops sounding like a bell at all.

• Radical tuning: removing roughly 10 to 15 percent of a bell's weight
• Radical tuning rebalances partials within a shape's limits, it does not change the shape
• A 2002 Lyminge retuning cut hum error from about two semitones to three-quarters of a semitone
• The same pass left the bell's prime unchanged, roughly three-quarters of a semitone flat

05

Over-tuning: when the reserve runs out

Every reserve has a bottom, and cutting past it does not fail gracefully. Hibbert's research is direct on the risk: taking out too much metal makes a bell too thin, and an overly thin bell can spoil its tone and also shift its strike note in ways a tuner did not intend, since thickness governs how tightly or loosely a bell's rim partials space themselves, and that spacing is what the ear assembles into a strike note in the first place ¹. A documented example is the 4th bell at Bagshot in Surrey, an 1883 casting by Warner that was heavily tuned by Gillett and Johnston in 1914; compared with an untuned Warner bell of similar date and nominal, the Bagshot bell sounds distinctly poorer, and its strike note sits noticeably sharper than expected precisely because the tuned bell is so thin ¹.

Over-tuning is therefore not a hypothetical caution, it is a documented, named failure mode with real bells as its evidence. It is also the clearest argument for the reserve's existence: a founder who casts too close to the acoustic target leaves a tuner no room to correct an imperfect casting without crossing straight into this failure, while a founder who casts generously gives the tuner room to aim for true-harmonic without gambling the bell's tone to get there.

06

Both halves of the rule are true at once

"Cast it right the first time" and "leave room to correct" sound like competing instructions, but bell founding treats them as two parts of the same discipline. The strickle and mould set the shape family a bell belongs to, and no amount of tuning changes that family once the metal has cooled; a founder who ignores this and simply casts bells thick and heavy without getting the underlying profile right does not get a better bell, only a heavier one. Taylor's bellfoundry's 19th-century work with the clock designer Baron Grimthorpe is a documented cautionary case: Grimthorpe favored bells cast very thick and heavy for their note, on the theory that mass alone improved tone, and the resulting peals were widely regarded as sounding poor, with many later recast ¹.

The reserve only pays off when it sits on top of a sound shape, not instead of one. Cast the profile as close to the acoustic target as the founder's skill allows, using the shape family the strickle encodes, then leave enough extra thickness that a tuner working on a vertical lathe can correct the residual error without crossing into over-thinning. Both instructions are true because they answer different failures: bad shape defeats tuning entirely, and too little reserve turns a good shape into an unfixable one the moment the tuner's cut goes slightly too deep. Cleaning and finishing operations around a tuning machine, including swarf, chips, and a heavy rotating mass, fall under the same industrial safety practice that governs the rest of a metal casting operation ³.