Chocolate making, professionally

How Chocolate Drops and Chips Are Made Industrially

By Oliver BennettPublished October 15, 2026Updated October 15, 202615 min read

Chocolate drops and chips are made by dosing small, measured amounts of liquid chocolate or compound onto a moving belt, letting them set as the belt carries them through a cooling tunnel, then taking them off the belt and packing them. A piston depositor or a rotary drop former does the dosing. Chunks are different: they’re cut from a slab cast on the belt or broken from solid chocolate.

The route from cocoa bean to finished chocolate is in the bean-to-bar process. A drop line starts where that route ends, with chocolate that is already made.

Independent literature on drop lines is thin, so most of what follows comes from patents. They describe machines in the inventor’s own words but are filed by machine makers and food companies with a stake in them, so each one below names the kind of company behind it. Makers’ product pages are used only to describe how a machine works, never for capacity.

Drops, chips, chunks: what is the difference

No standard separates them. US standards of identity define chocolate by its ingredients, so under 21 CFR Part 163 a chip takes its legal name from its mass: semisweet chocolate, milk chocolate, or a chocolate and vegetable fat coating. The naming rules are in real vs compound chocolate. The USDA’s purchase description for candy and chocolate confections specifies some candies by count per pound, but it has no item for chips or drops.

In patents and makers’ descriptions, the words follow the method. A drop or chip is a single dose of liquid mass that sets on the belt with a flat base. A depositor maker’s European patent, EP 3 888 469 B1, calls depositing straight onto the belt the typical route for high-volume “chips” and “buttons.” A chunk is an irregular piece, cut from a slab or broken from a block. None of these documents separates “drop” from “chip”: both name the same deposited piece.

Size has no standard either. Suppliers state it in their own terms, usually pieces per kilogram, and the figures vary widely even within one product family. When you buy or specify, ask for the count per kilogram and the average piece weight rather than a size code like “M” or “1M.”

One trap in older sources: a 1969 US patent titled “Chocolate drop making machine,” US 3,605,644, is about a fondant center coated with chocolate. It describes a candy that has nothing to do with baking drops.

The recipe: chocolate, compound and bake-stable versions

Three kinds of mass run through drop lines: chocolate, compound, and bake-stable formulas of either. Makers offer the same depositors and formers for “chocolate or compound,” so the equipment after tempering is largely shared. What changes is whether the mass needs tempering at all, and what’s in it to keep the piece in shape in the oven.

Chocolate and compound

Compound chocolate replaces some or all of the cocoa butter with other fats. Sources disagree on which group to call a “substitute” and which a “replacer,” so the plain names are used here: lauric and non-lauric. Both set in a stable form straight from the melt, so neither needs tempering, as real vs compound chocolate explains. On a drop line, compound goes to the depositor after melting alone.

Lauric fats and cocoa butter don’t mix well. A 2020 study by Yoshikawa and colleagues notes that in blends they form eutectics, which lower the melting point and hardness and can lead to bloom, and that makers of lauric compound keep cocoa butter below about 5% of the fat phase to avoid it. A line that switches between chocolate and lauric compound needs a cleaning plan between runs as well as a recipe change.

What “bake-stable” means

Bake-stable is a supplier term with no standard behind it. A food company’s 1992 US patent, US 5,344,664, gives the clearest functional definition: chips in baked goods need the thermal stability to keep a distinct shape through baking, and the mechanical stability to stay whole while the dough or batter is mixed. A chip that survives the oven but shatters in the mixer fails too.

The nearest thing to a measured method belongs to one ingredient supplier. Its 2016 patent application, US 2019/0098912 A1, defines bake stability as keeping shape, color and texture during baking, and tests shape by “flow out”: filling sticks are baked on puff pastry at 190 °C for 15 minutes, and whatever has run out of the pastry is cut off and weighed. It’s a test for fillings, written by a company that sells the ingredient, and it isn’t an industry method for chips.

The nearest public test is for heat rather than baking. The USDA purchase description requires its heat-resistant milk chocolate bar to keep its shape at 49 °C (120 °F) for 16 hours. The research literature is looser still: a 2019 study by Dicolla and colleagues lists published definitions of heat-resistant chocolate that range from not sticking to the wrapper above 30 °C to staying stiff at 50 °C, and notes that such subjective criteria make products hard to compare.

A bake-stable chip doesn’t stop the fat from melting. The 1992 patent describes the cocoa butter, together with the sugar, giving the chip “a fragile but definite structure” during baking so it keeps its shape. The shape is carried by the solids, not by solid fat.

How chips hold their shape in the oven

The peer-reviewed work on shape retention is about heat-resistant chocolate for hot climates. Its mechanisms apply to a chip in the oven, but the studies below didn’t test baking. A 2021 review by Suri and Basu groups the approaches into fat modification, sugar modification with added polyols, and process changes.

Less fat, or harder fat. Raising the melting point of the fat is the easiest route to heat resistance, Dicolla and colleagues write, and the review cites work showing that chocolate with less fat takes longer to melt. Patent examples don’t agree on a typical fat level for chips: the 1992 patent puts cocoa butter at 36% of a chocolate chip formula, while a food company’s 2003 application, US 2007/0160722 A1, used control chips with 24%. The sources support no “normal” figure.

A second structure that isn’t fat. Water, sugars, polyols, fiber, starch or protein can build a non-fat network that holds the piece together, according to Dicolla and colleagues. With water, the review explains, the surfaces of the sugar particles dissolve and stick to each other, forming a sugar network that resists deformation and traps the liquid fat. In the 2019 sensory study, panelists rated low-fat samples made with gelatin or polyols as the most heat resistant at 38 °C. A 2024 study by Laughter and colleagues made white chocolate heat resistant by curing solid pieces at 83% relative humidity and 29 °C for a week. Both studies had an author from, or support from, a large chocolate company.

There’s a catch for a drop line. Small amounts of water or polyol stirred into molten chocolate make it seize into an unusable, solid-like mass, the review warns, and a depositor needs a mass that flows.

Gelling the fat. Oleogels and networks of high-melting emulsifiers raise melting points in research, but the review notes that the amount of one such emulsifier needed for heat resistance exceeds what food standards allow.

Each route has a cost. Some heat-resistant chocolates are reported to be coarser, especially those with added water, and the review says these techniques generally fall short on palatability and mouthfeel. They also run into the legal definitions: add water, polyols, protein or other fats and the product may no longer be allowed to be called chocolate.

After baking, the picture is less settled. A 2012 study by Frazier and Hartel asked why chips baked in cookies showed less bloom even though the oven had melted them and broken their temper. The authors said fat migration contributes, but the exact mechanism was still unknown.

Step by step on the line

Patents and makers’ descriptions follow the same order: melt and temper the mass (compound is only melted), dose it onto a cooled belt, cool it through a tunnel, release it from the belt, then pack it.

Melting and tempering

By the time chocolate reaches a drop line it has already been refined and conched; how that sets the texture is in chocolate refining explained. Here it’s melted and tempered, so the cocoa butter sets in its stable crystal form. The basics are in what is chocolate tempering.

A steel-belt maker’s patent application, EP 3 107 404 A1, describes this stage in one line: the tempered mass may be deposited into mold trays or directly onto another surface, which is typically cooled in a temperature- and humidity-controlled space. Drop lines take the second route.

Tempering settings aren’t universal. A 2025 review by Stobbs and colleagues notes that tempering temperatures and times depend on the origin and composition of the cocoa butter, so the same settings can give bloom with a different cocoa butter. A line running several chocolates needs settings per recipe. Compound skips this step.

Forming the drops

Depositing with pistons is the most widely used way to dose exact amounts of confectionery mass, either into molds or onto a moving plastic or steel belt, according to the depositor maker’s 2021 patent. In the machine it describes, rotary valves open a path from the hopper to a row of bores, the pistons pull back to draw in a set volume, then the valves turn and the pistons push the mass out through the nozzles. While it deposits, the distribution block travels with the belt at the same speed, then moves quickly back against the belt’s direction for the next stroke.

Makers’ pages describe two arrangements. One moves the depositor onto the moving belt; another uses a table that lifts the belt toward the depositor to break each tail cleanly, with light vibration to settle the drops into shape. Either way, the nozzle and the belt don’t move relative to each other during the deposit, and a vertical movement separates the drop from the nozzle.

The nozzle has limits. The same patent notes that very low-viscosity chocolate deposited for small-diameter chips often can’t hold a stable surface at the nozzle tip between deposits, so it tends to drip.

A rotary drop former works continuously. A 1992 patent written for drop forming in general, WO 93/19843 (in German), describes a fixed, heated inner cylinder with a perforated tube turning around it. As each row of holes passes the outlet, material drips onto a cooling belt, where the drops harden as the belt moves on before they’re taken off. The outlet zone is heated so the material leaves at the viscosity that suits drop formation best. A maker that sells this type for chocolate describes the same parts, a heated stator fed with liquid chocolate inside a perforated rotating shell, with the shell and belt speeds matched so the drops land without deforming. It changes drop size by swapping the shell and nozzle bar. A belt maker describes chip production the same way, as dropping hot chocolate onto a steel belt where it cools gradually.

Chunks are formed another way. One equipment maker casts a slab of chocolate on the belt and cuts it into chunks with a separator blade and guillotine. Chunks can also be broken from solid chocolate, but grinding a meltable confection heats it until the pieces melt into a sticky mass on the cutting teeth. An industrial gas supplier’s 1992 patent, US 5,333,802, solves that by spraying liquid nitrogen into the grinder. The applicant sells liquid nitrogen, and the patent was written with ice cream inclusions in mind.

Filled drops use one-shot depositing. The 2021 depositor patent describes a ring-shaped inlet for a second mass arranged around the nozzle, and a maker of steel-belt depositing lines lists filled drops among its products.

Three other methods turn up in patents. A food producer’s 2001 US patent, US 7,018,670, spreads liquid food as a film on a chilled roller or belt and breaks it into particles once hard, mainly for products like yogurt, and admits that particles cooled that fast lack some properties of conventionally made chocolate. It also describes an older drop tower, about 10 to 15 meters high with cooling gas flowing up against the falling drops, and names its drawbacks: low capacity and a complicated build. The 1992 chip patent extrudes a water-based, low-fat mass, cuts it into short cylinders with a rotary knife, and leaves them overnight to even out the moisture. There, the recipe chose the forming method.

Cooling

The deposited drops set as the belt carries them through a cooling tunnel. The food producer’s patent describes the known setup as a cooling belt led through a tunnel of roughly 10 to 30 meters, with cold air at about 10 to 15 °C flowing against the product. The steel-belt application says cooling typically takes between 10 and 40 minutes, depending on the product. Both figures come from the background sections of patents, so read them as descriptions rather than settings to copy.

A long, gentle tunnel protects the crystals. The food producer’s patent says the temperature gradient must stay low, or the chocolate sets with the wrong crystal structure, and that slow cooling is what gives the drops their gloss and resistance to fat bloom. Independent work by Lonchampt and Hartel agrees on the direction: cooling too fast can produce unstable crystals and fine surface cracks. The steel-belt application adds that sudden changes in temperature or humidity can destabilize the crystals formed during tempering.

Both steel and plastic belts are used, and makers describe cooling by air blown over the product, by cooling from beneath the belt, or both. Some coolers are built in passes: the steel-belt application sets the tunnel conditions and belt speed so the chocolate deliberately sticks to the belt and rides the underside for a second pass. There, sticking is by design; the problem is drops that don’t let go at the end.

Zones and dwell time in cooling tunnels are covered in the production line guide.

Release from the belt

At the end of the tunnel the drops come off the belt. The 2021 depositor patent has them scraped off and carried on to be “wrapped and shipped.” The steel-belt application describes two ways to release them: a knife or scraper against the belt, or a sudden change in the belt’s angle that pops the chocolate free.

Once packed, drops store like any chocolate. The storage conditions that slow bloom are in why chocolate turns white.

Forming methods compared

The table sets the sourced facts for the three main methods side by side, on the same criteria.

The piston depositor column draws on EP 3 888 469 B1 and makers’ product pages, the rotary drop former column on WO 93/19843 and makers’ product pages, and the slab and cut column on one maker’s product pages. Capacity is left out: the only kg/h figures available are single vendors’ claims for their own machines, and they can’t stand for a whole method. The film-and-break, drop tower and extrusion methods each appear in one patent, so they stay out of the table.

Problems on a drop line and how they’re solved

The problems here come from patents, which describe the earlier technology’s faults in order to sell their fix. They show what goes wrong, not how often.

Tails and strings. Two companies’ patents describe the same nuisance from different machines. In the depositor patent, long “tails” or “threads” of mass between the drops and the nozzles end up on nearly every surface of the line, and cleaning them off is a constant job. In the rotary patent, threads between the drop on the belt and the mass left in the holes wrap around the shell if they don’t break in time, partly because of the material’s viscosity and partly because the shell cools.

The fixes differ by machine. Older depositors used tail knives or closure plates; the depositor patent’s own answer is a nozzle that narrows at the waist, with the piston reversed slightly after each deposit so surface tension holds the mass back. The rotary patent heats the shell at the outlet and draws the leftover mass back in. One maker of belt-lift depositors relies on the lift and vibration to break the tail. Both patents tie tails to how the mass flows and its temperature, and the depositor patent links them to the mass’s yield value, but none of this is measured independently.

Drops sticking to the belt. On a multi-pass cooler the steel-belt application controls sticking through the cooling conditions, belt speed and the belt’s surface material. The trouble starts when drops don’t release at the scraper or belt turn. Research on chocolate releasing from molds shouldn’t be stretched to belts. Where chunks are broken rather than cut, sticking moves to the machine: the heat of grinding melts the broken pieces onto the cutting teeth.

Weight and size variation. One rotary maker says its former delivers the mass at even pressure across the full belt width, so drops at the edge cool the same way as those in the middle and meet shape, size and weight tolerances. The depositor patent calls its weight accuracy highly satisfactory to manufacturers. Neither gives a tolerance in grams or percent, so ask vendors for one measured on your mass.

Planning a drop line

Start with the product. Decide what goes through the line (chocolate, compound, bake-stable versions or all three), the size range, and whether you need filled drops or chunks. Those answers settle whether you need a tempering machine, which forming method fits, and how often you’ll clean between runs, especially if lauric compound shares the line with chocolate.

The machines before the depositor, from tempering back to bean processing, are covered in chocolate production line: machines you need, in order, along with an RFQ checklist and the EU rules on who is responsible for a line built from separate machines. For drops, a capacity figure means little until it states the recipe and its viscosity, the piece weight, how running time is counted, and which machine sets the pace. Piece weight matters twice: the same kg/h means very different piece counts at different sizes, so ask for both.

Every product interacts with equipment differently, as a Kansas State guide on buying food equipment (Rausch and Erickson, MF-2096) puts it, and on a drop line tails, release and weight all depend on how your mass flows. Run a trial with your own chocolate or compound before you sign, and use the RFQ checklist in the production line guide to put the weight-variation figure into the acceptance terms.

Sources and open questions

These points have no source yet:

• an independent test for bake stability, or a study of heat-resistance mechanisms in baking chips;
• whether commercial baking chips use fat gelling;
• sieving, grading, rework and packing formats on a drop line;
• an independent comparison of steel and plastic belts, or of cooling methods, for drops;
• why a drop’s base releases from the belt or holds;
• a published tolerance for drop weight, in grams or percent;
• the causes of flat or misshapen drops.

The USDA purchase description cited is the 2009 edition (A-A-20177E); we couldn’t open the 2019 and 2025 revisions. The 2012 Frazier and Hartel paper was read only through a trade press summary.