Candy / Sugar Thermometer
What it is
A high-range thermometer — reading to roughly 200 °C / 400 °F — built to live in a pot of boiling sugar syrup. Traditionally a glass spirit (or, historically, mercury) thermometer mounted in a metal cage with a pan clip; now frequently digital. Its job is to tell you, with the precision that confectionery demands, exactly how concentrated a boiling sugar solution has become — because in candy-making, temperature is a direct proxy for water content, and water content determines whether your finished sweet is a soft fondant or a hard lollipop.
The science & materials
This is one of the most elegant temperature/chemistry relationships in cooking. As a sugar syrup boils, water evaporates and the sugar concentration rises. By boiling-point elevation, a more concentrated solution boils at a higher temperature: pure water boils at 100 °C, but a syrup at the soft-ball stage boils around 112–116 °C, and one at hard-crack around 150 °C. So the temperature of the boiling syrup reads out its sugar concentration — and concentration, once the candy cools, dictates texture. The progression of named sugar stages is really a progression of decreasing residual water:
| Stage | Temperature | ~ Sugar conc. | Cold-water test | Used for |
|---|---|---|---|---|
| Thread | 110–112 °C / 230–235 °F | ~80% | forms a soft 2-inch thread | syrups, candied fruit, some glazes |
| Soft ball | 112–116 °C / 235–240 °F | ~85% | ball that flattens when removed | fudge, fondant, pralines, Italian meringue, Swiss/Italian buttercream |
| Firm ball | 118–120 °C / 245–250 °F | ~87% | firm but pliable ball | soft caramels, some marshmallow |
| Hard ball | 121–130 °C / 250–266 °F | ~92% | hard yet still pliable ball | nougat, marshmallow, divinity, gummies |
| Soft crack | 132–143 °C / 270–290 °F | ~95% | bendable threads that then crack | taffy, butterscotch |
| Hard crack | 149–154 °C / 300–310 °F | ~99% | brittle threads that snap | brittle, toffee, hard candy, lollipops, spun sugar |
| Caramel | 160–182 °C / 320–360 °F | ~100% (water gone) | (no water test — sugar is now decomposing) | caramel sauce, praline, decorative caramel |
Past hard crack, essentially all water is gone and a different chemistry takes over: caramelization — the thermal decomposition of sucrose. Sugar melts (~160 °C), then breaks down and re-polymerizes into hundreds of new compounds — diacetyl (butterscotch note), furans, maltol, and the brown caramelan/caramelen polymers that give color. Light caramel forms around 170 °C, dark amber by 180 °C, and bitterness/burning past ~190 °C. Note this is distinct from the Maillard reaction, which requires amino acids and reducing sugars; caramelization is sugar alone.
The cold-water test is the pre-thermometer method and still a fine backup: spoon a little boiling syrup into a glass of very cold water and judge the texture it forms — a soft ball that flattens, a firm pliable ball, threads that bend then crack, threads that snap. It works because dropping the concentrated syrup into cold water instantly cools it and reveals the physical state its sugar concentration will set to. It is reliable in skilled hands but slower and more subjective than a calibrated thermometer.
How it's used
Clip the thermometer so the bulb is fully submerged in the syrup but not touching the bottom of the pan (the metal base is hotter than the liquid and reads false-high). Read at eye level to avoid parallax. Do not stir a cooking sugar syrup once it's clear — agitation and stray crystals seed crystallization, turning your candy grainy; wash down the pan sides with a wet brush and add a little corn syrup or acid (cream of tartar, lemon) as an "interfering agent" to inhibit crystal formation. Crucially, calibrate for altitude: boil plain water, note the reading (it'll be below 212 °F / 100 °C above sea level — roughly 1 °F lower per ~500 ft), and subtract that same offset from every target stage. Skipping this overcooks your candy by exactly the altitude error. Because sugar holds enormous heat and keeps cooking off the flame, pull a few degrees early and account for carryover.
Regional & cultural traditions
French confiserie has its own beautiful stage vocabulary — nappe, filet (thread), petit boulé and grand boulé (soft/hard ball), petit cassé and grand cassé (soft/hard crack), caramel — codified over centuries of sugar craft. The cold-water test predates the thermometer everywhere sugar was boiled. In Indian sweet-making, syrup consistency is judged by chashni "thread count": ek taar (one-thread) for gulab jamun soak, do taar / teen taar (two/three-thread) for denser sweets like jalebi and the syrup that sets in barfi — a tactile thread-pulling test between finger and thumb, a parallel system to the European one. Middle Eastern and Mexican confectionery carry their own syrup-stage traditions.
Cultural & historical context
Boiling sugar to controlled stages was, for centuries, a near-alchemical skill held by confectioners and judged entirely by hand and cold-water test. The candy thermometer's arrival in the 18th–19th centuries made stages reproducible and was a quiet enabler of the industrialization of candy — you cannot run a factory on "boil until it forms a firm ball between your fingers," but you can run one on "118 °C." The thermometer turned a craft mystery into a specification.
Reference notes
Cross-link: Caramelization, Maillard Reaction (and why they're different), Italian Meringue, Tempering Chocolate, Deep-Frying (shared high-range thermometry), Infrared Thermometer (surface caramel), Bain-Marie (gentle sugar/chocolate work). Pairs with the heavy copper sugar-pan and saucepan vessel entries — copper's even conduction is prized in confiserie precisely to avoid scorching.
---
When to use
Use it any time texture is non-negotiable and stage-dependent — fudge, caramels, nougat, marshmallow, Italian meringue, hard candy, brittle, spun sugar. Choose the thermometer over the cold-water test for reproducibility and speed; keep the cold-water test in your repertoire as a sanity check, a backup when a thermometer fails mid-batch, and the authentic method when working a traditional recipe by feel.
What goes wrong
Bulb against the pan bottom — the most common error — reads high and undercooks the batch. No altitude calibration overcooks it. Stirring or stray crystals seed graininess. Carryover overshoot — not pulling early — takes a soft caramel into firm-ball territory. Parallax and steam misread glass thermometers; thermal shock can crack a cold glass thermometer dropped into hot syrup. And the ever-present hazard: molten sugar at 150 °C+ causes severe burns and sticks to skin — splatter discipline and long sleeves are not optional.