The sensing element inside an RTD is a delicate thing, a fine platinum winding or a thin film that would not survive five minutes bare in a process. What survives is the sheath: the metal tube that houses the element, keeps the process fluid out, transfers heat in, and gives the assembly the mechanical strength to be stuffed into a thermowell or bolted to a pipe. The sheath's material, diameter, and construction quietly decide how fast the sensor responds, how long it lasts, and whether it drifts. This page covers what the sheath is made of, the diameter-versus-response trade-off, how it fits a thermowell, and how sheath integrity governs insulation resistance and drift.
RTD Sensor Sheath in one line: An RTD sensor sheath is the metal tube, commonly mineral-insulated stainless steel such as 316 or a high-temperature alloy like Inconel, that encloses and protects the RTD element from the process while conducting heat to it. The sheath material is chosen for corrosion resistance and temperature rating, and its diameter trades response time and immersion against vibration strength. Sheath integrity is critical because a breach lets moisture into the mineral insulation, which drops the insulation resistance and shows up as sensor drift or an unstable reading.
Most industrial RTDs use a mineral-insulated, or MI, construction, where the platinum element and its lead wires are packed inside a metal sheath filled with a tightly compacted ceramic powder, typically magnesium oxide. That compacted powder electrically insulates the element and leads from the sheath while conducting heat well, so the element is protected from the process yet stays in good thermal contact with it. The MI design is flexible and rugged: the sheathed cable can be bent to route around obstructions, cut to length, and sealed, which is why it dominates field RTD construction. The whole assembly's ability to keep the element dry and insulated depends on the sheath staying sealed against the environment.
The sheath material is selected to survive the specific process. General-purpose sheaths are 316 or 316L stainless steel, which handles a wide range of ordinary process chemistries and temperatures at reasonable cost. For higher temperatures or more aggressive atmospheres, high-nickel alloys such as Inconel are used because they resist oxidation and hold their strength where stainless would scale or weaken. In particularly corrosive service, more exotic sheath alloys are specified to resist the specific attacker, because a sheath that corrodes through is not just a mechanical failure, it is a path for process fluid and moisture to reach the element and destroy the measurement.
Matching the sheath to the service is a durability decision more than an accuracy one, but the two connect. A sheath that is under-specified for the temperature will oxidize and thin over time, and a sheath that is wrong for the chemistry will pit and eventually breach. Either way the failure ends the same: the protective barrier is compromised, moisture and contaminants reach the mineral insulation, and the sensor starts to misbehave. Getting the sheath material right for the temperature and the fluid is what lets the element inside do its precise, quiet job for years rather than months.
Sheath diameter is a genuine trade-off rather than a free choice. A thinner sheath has less metal mass between the process and the element, so heat reaches the element faster and the sensor responds more quickly to temperature changes, which matters where the control loop needs to catch fast transients. But a thinner sheath is mechanically weaker and more vulnerable to bending and to vibration-induced fatigue, especially in high-velocity flow or where the assembly hangs unsupported inside a well. A thicker sheath is sturdier and better able to survive vibration and rough handling, at the cost of a slower response because there is more mass to heat and cool. Choosing the diameter means weighing how fast the measurement needs to react against how much mechanical abuse the sensor will face.
Immersion length is the companion decision. For the element to read the true process temperature, it has to sit far enough into the process that it is dominated by the process temperature rather than by heat conducting away along the sheath and stem to the cooler mounting. Too shallow an immersion leaves the reading pulled toward ambient by that conduction path, a stem-loss error, and the sheath and sensor geometry set how deep is deep enough. This is why RTD assemblies are specified with an insertion length suited to the pipe or vessel and the required accuracy, so the sensing tip lands well within the process, not lingering in the thermal boundary near the wall.
Most process RTDs do not touch the fluid directly but sit inside a thermowell, and the fit between sheath and well bore matters as much as the sheath itself. An air gap between the sheath and the well bore is a thermal insulator that slows response and can add error, so RTD assemblies are often spring-loaded: a spring behind the sensor pushes the sheath tip firmly against the closed bottom of the thermowell to make good thermal contact and to keep it there despite vibration and thermal expansion. A properly sized sheath that seats against the well bottom under spring pressure gets the fast, accurate reading the geometry allows; a loose or bottomed-out fit gives a sluggish, unreliable one.
The single most important thing the sheath does for accuracy over time is keep the mineral insulation dry. The magnesium oxide powder inside is an excellent insulator when dry, but it is hygroscopic, meaning it readily absorbs moisture, and wet insulation conducts. If a crack, a corrosion pit, a failed end seal, or mechanical damage lets moisture into the sheath, the insulation resistance between the element, the leads, and the sheath falls. That leakage path effectively adds unintended resistance in parallel with the element, which biases the reading and makes it unstable, so a compromised sheath shows up as drift, noise, or a value that wanders with humidity and temperature rather than a clean failure.
Because of this, insulation resistance is a diagnostic worth measuring on an RTD assembly, and a falling insulation resistance is an early warning that the sheath's protection is failing before the sensor gives out entirely. A healthy MI RTD reads a very high insulation resistance; a degraded one reads progressively lower as moisture works in, often long before the reading is obviously wrong. This is why a periodic insulation-resistance check catches a marginal sensor while it can still be planned for replacement, rather than waiting for it to fail during operation. The sheath is the barrier, and insulation resistance is the health check on that barrier.
In the field these degradations are gradual, which is exactly what makes trending them valuable. An RTD whose sheath is slowly admitting moisture drifts a little more each month and grows noisier, a pattern that a single reading rarely reveals but that a monitoring history exposes clearly. When these sensors report into a SCADA or cloud monitoring layer, a platform such as Merobix shows the affected point creeping away from its neighbors and getting jumpier, which flags a likely sheath-integrity problem and lets the team schedule a replacement before the measurement fails a control loop or a safety function. The sheath's slow decline, invisible on any one day, becomes an actionable trend when the readings are kept and compared over time.
General-purpose RTD sheaths are usually 316 or 316L stainless steel, which suits a broad range of ordinary process temperatures and chemistries. For higher temperatures or oxidizing atmospheres, high-nickel alloys such as Inconel are used because they resist scaling and keep their strength, and for aggressive corrosive service more specialized alloys are specified to resist the particular attacker. The choice is driven by the process temperature and fluid, because a sheath that corrodes or oxidizes through exposes the element to moisture and fails the measurement.
Yes, a thinner sheath has less metal mass between the process and the element, so heat reaches the element faster and the sensor responds more quickly. The trade-off is mechanical strength: a thin sheath is more prone to bending and to vibration fatigue, especially in high-velocity flow or where it hangs unsupported. A thicker sheath survives vibration and handling better but responds more slowly, so the diameter is chosen by weighing required response speed against the mechanical environment.
The mineral insulation inside the sheath is dry and highly insulating until moisture gets in, and it absorbs moisture readily. If the sheath cracks, corrodes, or loses its end seal, moisture enters, the insulation resistance falls, and that leakage path adds unintended resistance in parallel with the element, biasing and destabilizing the reading. The result is drift, noise, or a value that wanders with humidity, which is why a falling insulation resistance is an early warning that the sheath is failing.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.