Most steam heats a process. Tracing does something humbler: it stops a pipe from cooling. A tracer is a small steam line — DN15 or DN20 — strapped along a product pipe under shared insulation, quietly replacing the few kilocalories the line loses so that whatever is inside stays hot enough to flow. India runs on traced lines more than most engineers notice: bitumen at road plants, palm and rice-bran oil in winter, molten sulphur, fifty-percent caustic that freezes at a spring-morning 15 °C, molasses, and the impulse lines of outdoor instruments. None of these needs heating. All of them need not cooling — and the difference decides the whole design.
A loss to replace, not a load to heat
The tracer's duty is the traced line's heat loss, nothing more. That loss belongs to the insulation, not the product: an insulated assembly holding 40 °C against a winter night sheds tens of kilocalories per hour per metre — the arithmetic of lesson C7, run at holding temperature instead of steam temperature. Grade the answer by how much loss you must replace: light tracing (one tracer) holds easy products warm; medium (two or three tracers) holds hotter products against bigger losses; heavy tracing gives up on tracers entirely and jackets the pipe. Most plant lines are light-tracing problems wearing dramatic faces.
The practice plant pumps refined oil 50 m from the tank farm, outdoors, DN100. It must stay pumpable at 40 °C or better; the winter night touches 10 °C. Tracing runs round the clock — the line must not set while the plant sleeps.
Read that last line twice. Tracing steam is nearly free; tracing neglect is not. The economics of a traced system live in its traps (lesson G2), not its steam meter.
The rules that decide success
Tracing fails by layout, almost never by thermodynamics. Four rules carry the whole craft. The tracer runs along the bottom of the product pipe, in firm contact. Heat conducts into the pipe wall where metal touches metal and rises through the product from below; a tracer that sags away is heating an air gap. Never spiral-wrap. A wrapped tracer looks thorough and fails twice: the tracer grows more than the cooler product pipe and every heat-up drags the spiral off its contact, and a helix has no low point that drains — it hammers instead (lesson C4's physics on a small scale). Keep each run short. A single tracer is a tiny two-phase pipe; past a few tens of metres — 30 to 50 m is the working ceiling — the tail runs on condensate and goes cold while the meter says everything is fine. Split long lines into short runs fed from a small manifold. Let it move. The tracer runs ~100 °C hotter than the product pipe; give it slack at supports and a loop at direction changes, or expansion will do the routing for you. Where contact matters most, a bead of heat-conducting cement along the touch line multiplies the metal-to-metal path — an honest upgrade, not a luxury.
One tracer, one trap
The temptation is obvious: six tracers, one trap, five traps saved. The physics says no. Runs of different length and exposure condense at different rates; teed into one trap, the lightest-loaded run blows its steam through first and holds the shared trap shut — the same short-circuit that condemns group-trapped drip legs (lesson G1). Every tracer run drains through its own trap. The natural choice is the balanced-pressure thermostatic trap (lesson D5): it deliberately holds condensate back until it subcools — a flaw on process equipment, a feature on a tracer, where banked condensate sits harmlessly in the line and its sensible heat is still tracing. Group the trap stations, not the traps: an accessible manifold at grade, or an elevated station the condensate rises to — workable at tracer loads, but designed with the lift counted against the differential (lesson G3's rule, in miniature), never improvised. Two habits finish the job: drain tracers at shutdown where frost is possible, and go gently on instrument impulse lines — light tracing only, because an overheated impulse leg boils and the transmitter starts telling stories (lesson D8).
- Walk the traced lines before winter: insulation closed over pipe and tracer, contact firm, no tracer sagging into the annulus air.
- Tracer traps are traps — survey them (lesson G2). One failed-open tracer trap wastes six to ten times its tracing duty; a failed-shut one announces itself as a line that will not flow at 6 a.m.
- Hunt the helpful tee: anywhere two tracer runs share one trap, expect one of them cold. Un-tee it.
- Long runs: feel the tail. A tracer hot at the manifold and tepid at 45 m is telling you to split the run.
- Price a solidified line honestly once — product loss, downtime, and the crew with heating torches — and the tracing budget approves itself.
- Tracing replaces a loss. The duty is tens of kcal/h per metre — the steam is nearly free.
- Bottom of the pipe, firm contact, never spiral-wrapped, room to expand.
- Runs of 30–50 m maximum; split longer lines.
- One tracer, one trap — balanced-pressure, which banks condensate on purpose.
- The traced system's economics live in its traps, not its steam meter.
Molten sulphur is the one product a tracer can ruin by working too well: above about 159 °C its viscosity climbs a thousandfold and the "liquid" stops pouring. Sulphur plants solved it with the saturation curve itself — trace with low-pressure steam at around 2.5 kg/cm²g and the tracer physically cannot exceed ~139 °C, because pressure fixes temperature (lesson A3). No controller, no sensor, no alarm: the thermostat is thermodynamics. It is the neatest illustration in the plant of why saturated steam, alone among heat carriers, carries its own temperature limit in its pressure gauge.
FAQ
Why not just put steam inside the product line and heat it directly?
Because the product does not need heating — it needs its losses replaced. Steam inside the line means contaminating the product, pressure-rating the product pipe for steam, and delivering a thousand times the duty required. The tracer sits outside, replaces the tens of kcal/h per metre that actually leak away, and touches nothing.
Why must every tracer run have its own trap?
Runs of different length and exposure condense at different rates. Teed into one trap, the run with the least condensate blows steam through first and holds the trap shut against the others — the classic group-trapping short circuit. The saved traps are paid for many times over by the cold lines they cause.
Which trap type suits tracing, and why is holding back condensate acceptable here?
The balanced-pressure thermostatic trap — precisely because it discharges condensate only after a few degrees of subcooling. On a heat exchanger that banking floods the surface; on a tracer the banked condensate lies harmlessly in the line and its sensible heat is still doing the tracing. The trap's one vice becomes the application's virtue.
How long can a single tracer run be?
Keep it within a few tens of metres — 30 to 50 m is the working ceiling. A tracer is a tiny two-phase line: too long, and the far end runs on condensate instead of steam and goes quietly cold while the supply end looks perfect. Split long lines into short runs from a small steam manifold.
Is electric tracing better than steam tracing?
Each wins somewhere. Electric tracing holds precise low temperatures, reaches lines far from any steam main, and returns no condensate. Steam tracing wins where steam already exists: rugged, cheap per metre of heat, safe in flammable areas without certification, and — uniquely — self-limiting in temperature at the saturation point of its pressure. A steam plant almost always traces with steam.
Check yourself
Five quick questions on this lesson. No marks, no records — the score is for you.