Line Frequency Furnace

What is a Line Frequency Furnace

A line frequency furnace is an induction furnace whose inductor is energised directly from the power grid at its native utility frequency, without a solid-state converter to raise that frequency. In practice this means 50 Hz across most of Europe, Asia, Africa and Australia, and 60 Hz in North America and parts of South America. The terms mains frequency furnace, utility frequency furnace and power frequency furnace all describe the same machine. The defining contrast is with the medium frequency furnace, which inserts a rectifier and inverter to lift the inductor frequency to roughly 150 Hz up to 10 kHz, and with high frequency systems that reach hundreds of kilohertz for small precision melts.

Induction heating itself is simple in principle: an alternating current in a coil produces a rapidly reversing magnetic field, that field induces eddy currents in the conductive charge, and those eddy currents dissipate as heat through Joule heating, with an additional hysteresis contribution while the charge remains ferromagnetic. The eddy currents also interact with the field to drive vigorous stirring of the melt, which assures good mixing of alloy additions and temperature. What changes with frequency is not the principle but the balance of penetration depth, power density and stirring force, and it is that balance which makes line frequency a deliberate engineering choice rather than merely the cheapest option.

Historically, mains frequency was the original industrial induction melting frequency because it needed no power electronics: the coil connected to the grid through a transformer and a power-factor capacitor bank. Early coreless mains frequency melters dominated iron foundries through much of the twentieth century. The channel induction furnace, patented in principle as the Ajax-Wyatt vertical core type, ran at mains frequency from the outset because its transformer action is most efficient there. From the 1980s onward, robust solid-state medium frequency converters displaced mains frequency coreless melters for cold-charge melting, because medium frequency furnaces start from cold without a molten heel and pack more power into a smaller crucible. Mains frequency survives where its strengths are decisive: very large baths and continuous holding.

Scale spans a wide range. Induction power supplies are built from about 10 kW up to 42 MW, with corresponding melt sizes from roughly 20 kg to 65 tonnes of metal. Coreless furnace orders of 35 to 85 tonnes are now routine for large-scale foundry operations. Channel holding furnaces likewise reach tens of tonnes in casthouse pressure-pour and duplexing duty. At these capacities the installed power is so large that the frequency must be kept low to keep stirring controllable, which pulls the optimum naturally toward, and sometimes onto, the grid frequency.

Four engineering metrics decide whether a line frequency furnace fits a given duty: power density (how fast it can add energy), stirring intensity (whether mixing helps or harms the process), electrical efficiency (how much grid energy reaches the metal), and refractory lining life (the dominant recurring cost). These four interact. A furnace optimised for gentle, efficient holding of a single alloy at high power factor is a different machine from one optimised for fast, flexible melting of mixed scrap, even though both rely on the same induction principle.

Coreless vs Channel Types

Two furnace architectures share the mains frequency principle, and confusing them is the most expensive mistake in this category. The coreless furnace is a crucible wrapped in a coil that heats the whole charge directly. The channel furnace is a transformer in which the molten metal itself forms the heated secondary winding. They differ in startup, flexibility, efficiency and the role frequency plays. The table below sets out the core differences before each type is discussed in turn.

The coreless furnace is a non-conductive refractory crucible encircled by a precisely wound, water-cooled hollow copper coil. Alternating current in the coil induces eddy currents directly in every part of the charge, so the entire mass heats and there is no concentrated hot loop. This direct heating gives the coreless type its flexibility: it accepts cold scrap, ingot or returns, melts virtually any conductive metal, and can be drained completely and recharged with a different alloy. The price of that flexibility is efficiency. Normal coreless electrical efficiency is approximately 75 percent, because at any frequency some field energy is lost as coil resistance heating, cooling-water heat removal and stray losses.

The channel furnace behaves as a transformer with a laminated iron core. A primary coil is wound on the core, and a narrow channel of molten metal threading through the core forms a single-turn secondary. Current induced in that metal loop heats it, and natural convection plus the loop geometry circulates the heated metal up into the main bath. Because the magnetic circuit is closed by an iron core rather than air, coupling is extremely tight and efficiency reaches 95 to 98 percent. The catch is absolute: the secondary only exists while the channel stays full of liquid metal, so a channel furnace must hold a permanent molten heel, cannot start from cold scrap, and cannot be poured empty without freezing the inductor.

This difference dictates application. Channel furnaces excel at holding and superheating large volumes of a single alloy on continuous and multi-shift schedules, and at pressure-pour casthouse duty for aluminium, copper, brass and ductile iron. Their high efficiency makes them economical when metal must simply be kept hot and ready. Coreless furnaces, by contrast, are the melt shop workhorse where charge composition changes, where the furnace must be emptied for alloy changeovers, or where cold-start capability matters. Many large operations pair the two: a coreless melter feeds a channel or coreless holder, a configuration called duplexing.

A practical hybrid worth noting is the coreless holding furnace fitted with a mini-heel coil. Standard coils lose efficiency at low metal levels, but a mini-heel coil keeps coupling effective down to a roughly 10 percent heel, allowing continuous duplexing while freeing up to 90 percent of the bath as working metal. This narrows the historical efficiency gap with channel holders while keeping the coreless advantages of cold-start capability and easy lining replacement.

Mains-Frequency Stirring Physics

The single most important consequence of choosing line frequency is electromagnetic stirring. Stirring intensity rises as frequency falls and as coil power rises, so a low fixed frequency combined with large installed power produces powerful bath motion. This is the property that makes mains frequency attractive for some duties and unacceptable for others, and understanding it is the difference between a furnace that mixes alloys cleanly and one that drags slag and gas into the melt. The table below maps the stirring and power-density trade-off across the frequency band.

The reason frequency and stirring are coupled lies in penetration depth, the skin depth at which the induced current concentrates. Lower frequency means deeper penetration and a thicker driven layer, which produces stronger body forces and more vigorous circulation. Higher frequency confines current to a thin surface skin, heating efficiently but stirring little. At mains frequency the penetration is deep enough that a large bath rolls and mixes thoroughly, which is ideal for dissolving carburiser, homogenising carbon and temperature in iron, and blending alloy additions, but is excessive in a small crucible where it would slop metal and entrain slag.

Because of this, there is an optimum frequency band that scales with furnace size. Manufacturer engineering guidance illustrates it clearly: a coreless furnace melting about 2.25 tonnes of steel has an ideal supply frequency near 600 Hz, while a furnace melting about 16 tonnes of iron has an optimal frequency near 150 Hz. The trend is monotonic, larger capacity wants lower frequency, and at the largest capacities the optimum approaches the grid frequency itself, which is one reason mains frequency coreless melters were historically built large. Power density tells the complementary story. Mains frequency coreless furnaces are rated no higher than about 300 kW per tonne because higher power at that frequency would stir too violently, whereas medium frequency furnaces are rated at 600 to 1,000 kW per tonne and melt cold scrap far faster.

Two failure modes bracket the useful window. Frequency too low for the furnace size, or power too high, causes violent surface motion, slag inclusion and gas pickup that degrade metal cleanliness. Frequency too high for the size leaves the bath poorly mixed with cold spots and uneven temperature. The art of furnace specification is choosing a frequency and power rating that land the stirring inside the useful band for the intended metal and bath size. For holding duty, where the metal is already molten and uniform, gentle is preferred, so channel furnaces run at mains frequency but at modest power, exploiting the efficiency of the iron core without overdriving the bath.

Stirring also interacts with refractory wear. Vigorous circulation scours the lining at the bath line and in the lower crucible, so high-stir furnaces demand tougher linings and disciplined operating practice. This is why stirring is not only a metallurgical parameter but a cost driver, feeding directly into the lining-life economics covered next.

Refractory Linings and Melt Media

The refractory lining is the consumable heart of an induction furnace and one of its largest recurring costs, so lining chemistry must be matched to the metal and slag. Coreless furnaces use a monolithic, dry-vibratable lining that is rammed in place around a steel former and then sintered by the first heats, producing a sintered working face backed by an unsintered buffer layer that protects the coil. Channel furnaces use a castable or dry-vibratable lining around the inductor and throat, the narrow zone where the metal loop passes the core and where oxide buildup tends to accumulate. The table below summarises the three main lining families and their fit to common melts.

Acidic silica ramming mass, typically more than 98 percent SiO2, is the default lining for grey and ductile iron and many carbon steels. It resists acidic slag, tolerates thermal cycling well, and is the lowest-cost option, which matters because the lining is replaced on a cycle measured in heats. Its limitation is chemistry: it is attacked by basic slags and is unsuitable for high-manganese melts, where the slag chemistry would dissolve silica rapidly. In channel furnaces the throat region around the inductor sees both high temperature and metal flow, so a robust aluminosilicate is specified there to control buildup.

Neutral alumina and spinel-bonded linings serve higher temperatures and alloy steels, and are common for copper and copper-alloy melting because alumina resists the relevant slags and tolerates the higher superheat. Spinel-forming dry vibratory mixes are used around channel inductors where the combination of temperature and flow is severe. Basic magnesia linings are chosen for high-manganese and stainless grades whose basic slags would destroy a silica lining, accepting higher cost and more careful thermal-shock management in exchange for chemical compatibility.

Melt media follow the same logic. Line frequency furnaces handle the full range of foundry metals: grey and ductile cast iron, carbon and low-alloy steels, copper and copper alloys such as brass and bronze, and aluminium, with the channel design particularly favoured for the lower-melting non-ferrous metals where its holding efficiency pays off. The table below gives a first-pass media-to-lining map for selection; before engineering implementation, always obtain the manufacturer corrosion and slag-compatibility data and confirm against the specific alloy, slag practice and operating temperature.

Key Specification Parameters

A furnace datasheet lists many numbers, but only a handful drive the selection and the lifetime cost. The parameters below are the ones a purchasing engineer must read, verify and compare across bidders, because together they fix throughput, energy bills, metal quality and serviceability. Each is explained in turn.

Capacity and installed power. Capacity is the nominal molten metal mass the crucible or bath holds, from tens of kilograms in laboratory units to 65 tonnes and beyond in large melters; coreless orders of 35 to 85 tonnes are now routine for big foundries. Installed power ranges from about 10 kW to 42 MW. The ratio of the two, power density in kW per tonne, sets melt rate and is where the frequency choice becomes visible: mains frequency coreless furnaces are rated up to about 300 kW per tonne, while medium frequency units reach 600 to 1,000 kW per tonne.

Frequency. For a true line frequency furnace this is fixed at the grid value, 50 Hz or 60 Hz. It is not a free parameter but a consequence of the furnace size and duty, since stirring must stay within the useful band. Where a datasheet quotes a higher fixed frequency, the unit is a medium frequency furnace and should be compared on that basis. One electrical benefit of frequency choice worth noting is that operating at a higher frequency halves the capacitance needed for power-factor correction, so mains frequency furnaces carry comparatively large capacitor banks.

Electrical efficiency and energy consumption. Coreless efficiency is approximately 75 percent; channel efficiency is 95 to 98 percent. In energy terms, the thermodynamic minimum to take a tonne of iron to about 1,500 degrees Celsius is roughly 396 kWh, while practical coreless melting of a tonne of iron to about 1,450 degrees Celsius runs under 600 kWh, with typical specific consumption of 500 to 800 kWh per tonne depending on charge, grade and operating discipline. The shortfall versus theory is lost to bath-surface radiation, refractory conduction, coil cooling water and supply losses.

Power factor and electrical supply. Induction furnaces present a highly inductive load and require a capacitor bank for power-factor correction; sizing it correctly avoids grid penalties and harmonic problems. The supply chain comprises a rectifier transformer, in medium frequency units a frequency converter, the capacitor rack and the water-cooled power cables to the coil. Specification should confirm the transformer rating, the correction package and harmonic mitigation, since large furnaces are significant grid loads.

Cooling. The coil is hollow copper tubing through which cooling water flows continuously, and a reliable cooling system is critical: lose the water and the coil can overheat and fail. Datasheets specify cooling water flow, pressure, inlet temperature and circuit redundancy. Closed-loop chillers or cooling towers are sized to the installed power, and the cooling water also carries away a share of the energy that shows up as the efficiency shortfall.

Lining package and life. The lining family, thickness, push-out or knock-out replacement method, and expected life in heats are core economic parameters because relining is a major recurring cost and a planned downtime event. Specify the lining chemistry to match slag practice, and confirm whether the design supports rapid push-out relining to minimise lost production.

Safety and standards. Electroheat installations are covered by the IEC 60519 series on safety in installations for electroheating and electromagnetic processing, with IEC 60519-3 addressing induction and conduction heating equipment specifically. Water cooling integrity, ground-fault and earth-leakage detection on the coil, and interlocks against running dry are standard safety expectations that should appear in the specification.

Selection Decision Factors

To turn the preceding chapters into a purchase decision, follow the ordered sequence below. Most selection errors come not from one wrong number but from deciding the architecture too late, after a frequency or power figure has already been fixed. These steps double as an RFQ template.

1. Duty first: melt, hold, or duplex. Decide whether the furnace must melt cold charge, hold and superheat already-molten metal, or do both in a duplex line. Melting flexibility points to coreless; efficient single-alloy holding points to channel; a combined line points to a coreless melter feeding a holder.
2. Architecture: coreless or channel. If the duty needs cold start, alloy changes or pour-empty operation, choose coreless. If it is continuous single-alloy holding at maximum efficiency, choose channel and accept the permanent molten heel.
3. Capacity and power density. Set bath capacity from the casting plan, then derive installed power from the target melt rate. Mains frequency coreless ratings cap near 300 kW per tonne; if you need faster melting, you are specifying a medium frequency unit instead.
4. Frequency and stirring. Confirm the frequency is appropriate to the capacity so stirring lands in the useful band: large baths near mains frequency, smaller coreless melters at medium frequency. Excessive stirring causes slag and gas pickup; insufficient stirring causes cold spots.
5. Metal and lining. Match lining chemistry to the alloy and slag: acidic silica for iron and many steels, neutral alumina or spinel for higher temperatures and copper, basic magnesia for high-manganese and stainless grades. Confirm expected lining life in heats and the relining method.
6. Electrical supply and cooling. Specify the rectifier transformer rating, power-factor capacitor package and harmonic mitigation, and the closed-loop cooling water flow, pressure and redundancy. Confirm the cooling system is sized to installed power with fail-safe interlocks.
7. Safety and standards. Require compliance with the IEC 60519 series for electroheat safety, ground-fault detection on the coil, run-dry interlocks and water-failure protection. Confirm local electrical code and grid-connection approval for the load.
8. Total cost of ownership. Sum purchase, installation, the recurring relining cost and downtime, energy at 500 to 800 kWh per tonne, cooling and maintenance over the expected service life. A cheaper furnace with shorter lining life or worse efficiency frequently costs more across a two-decade life than a better-specified unit bought once.

One dimension that is easy to overlook at purchase is manufacturer serviceability: availability of lining materials and relining support, spare coils and converter parts, field commissioning and calibration of temperature and power instrumentation, and long-term firmware and controls support. These determine repair response time after a decade of production. Inductotherm Group, Otto Junker and ABP Induction Systems all maintain global furnace, lining and service support networks, which makes them defensible choices for large, long-lived installations where a frozen channel or a failed coil halts an entire casting line.