Heat treatment: fixed recipes, hot furnaces, late hardness checks

Fixed recipes regardless of load, furnaces hot while empty

What happens. Each furnace runs the recipe laid down for the part, with a soak time that is long enough for the heaviest load. A light load is held just as long. Between loads, the furnace is often kept hot.

What it costs. Gas and power are burnt to keep parts at temperature after they are through, and to heat a furnace that is waiting for the next order. Furnace time is also capacity the next job could use.

Why the usual fix fails. Cutting soak times by hand is a risk to the quality that no shift leader wants to carry. A thermocouple in a test piece shows one point in a load of many parts.

How a fixed-price remote diagnosis checks it. From furnace logs, load descriptions and thermocouple data, I compute how the temperature in the core of the parts develops. The fixed-price remote diagnosis tells you whether recipes and the furnace schedule can be shortened with a computed part temperature, and what test data would back it up.

Strip temperature in annealing lines is not measured everywhere

What happens. In a continuous annealing line the strip passes through heating, soaking and cooling zones. Pyrometers measure the surface at a few points, but the temperature that gives the material its properties is the one inside, along the whole path.

What it costs. If the strip is too cool, properties are off specification. If it is too hot, gas is wasted and the strip may suffer. Operators keep a margin.

Why the usual fix fails. More pyrometers have limits: the furnace atmosphere and the strip’s surface change what they read. Setpoints from a table do not follow changing strip thickness and width.

How a fixed-price remote diagnosis checks it. From line data (zone temperatures, speed, strip dimensions, pyrometer readings), I compute the strip temperature along the furnace. The fixed-price remote diagnosis tells you whether a computed temperature could lead the zone setpoints and reduce the margin.

Hardness is checked afterwards

What happens. After quenching, the parts are tested for hardness, usually on samples. If a batch fails, it is too late: parts are reworked or scrapped. The quench itself, how fast each part really cooled, was not measured.

What it costs. Rework means a second pass through the furnace, with the gas, time and distortion that come with it. Hard-to-reach shapes in a batch cool differently from the sample.

Why the usual fix fails. Tighter quench settings reduce the spread but do not show which batch is at risk. Testing every part destroys it or takes too long.

How a fixed-price remote diagnosis checks it. From quench records (bath temperature, agitation, batch layout, part geometry), I compute how each part cooled. The fixed-price remote diagnosis tells you whether a computed cooling per batch can warn you before the hardness test, and what it would take.