Displacer Level Transmitter

What a Displacer Level Transmitter Is and What Is Being Decided

A displacer level transmitter reports where the level is along its span, as a continuous 4 to 20 mA signal proportional to the submerged fraction of a calibrated cylinder. The boundary jobs are easy to state: a continuous band of level belongs to this class of instrument, while a single alarm line belongs to a level switch, and mature practice splits alarm and control functions onto different measurement principles so one fault cannot silence both. On a boiler drum, the displacer or its cage sibling feeds the three element control loop, while independent switches and gauge glasses carry the alarms that boiler code demands. The displacer is bought for the control quality of its signal, not for the paperwork of its alarms.

Its nearest architectural cousins share the same external hardware. A magnetostrictive level transmitter follows a float along a bypass chamber, and a magnetic level gauge hangs a float in a parallel chamber for local indication, as the lead photograph shows. The head based hydrostatic level transmitter reads the same vessel from pressure taps. Top down microwave instruments measure without touching the liquid: a radar level meter in free space, a guided wave radar level probe along a conductor. An ultrasonic level meter serves open atmospheric vessels, and a capacitance level transmitter tracks level along a probe. The full comparison between these families belongs in the selection chapter, and only there; this chapter settles what a displacer is and what its purchase decides.

What is being decided at RFQ stage is a configuration, not a model number. First the service mode: single liquid level, liquid liquid interface, or density and concentration output, because each mode calibrates the same hardware differently. Second the mounting: external cage with its valve set, or cageless internal displacer hung from a single nozzle. Third the sizing: displacer diameter and length against span, liquid density, and, for interface duty, the density difference between the two liquids. Fourth the pressure and temperature envelope, which fixes the cage class and the torque tube architecture. Fifth the compensation: what the transmitter does when density moves with temperature, stated as a requirement rather than left as a hidden error. A quotation that answers all five in writing is a specification instrument; one that answers three is a commodity with a sticker.

Five references organize the datasheet. ASME Boiler and Pressure Vessel Code Section I, paragraph PG-60, governs water level indication and alarms on power boilers and names what hardware must sit on the drum. API 2350 frames overfill management and alarm layering for petroleum storage, the context in which any level transmitter on a tank participates in a safety narrative. Performance figures are evaluated and reported against IEC 60770 practice. Safety function electronics carry IEC 61508 assessment, and wetted parts in sour service follow NACE MR0175, now ISO 15156, material selection. A datasheet quoting none of these is a price tag, not a specification.

How Buoyancy Becomes a Measurement Signal

The sensing physics is Archimedes' principle: a body immersed in fluid loses weight equal to the weight of fluid it displaces. A displacer is that body, manufactured so its volume and weight are known to the calibration shop. As level rises from dry to fully wetted, the buoyant force grows in proportion to the submerged volume, and the transmitter reads that force as level. In numbers: a representative displacer 76 mm in diameter and 610 mm long has a volume of about 2.8 liters, so full submersion in water changes the buoyant force by about 27 newtons, 2.8 kilograms force. The force span across the calibrated range equals volume multiplied by the density difference between the liquid and the vapor above it, and every sizing, interface, and error question on this page reduces to that one sentence.

The measuring element in the classic design is the torque tube: a thin walled alloy tube closed at the process end by a cross piece, with the displacer hung on a lever arm. Buoyancy change twists the tube a few degrees, typically 2 to 6, and the tube serves three roles at once: torsion spring, rotary bearing, and pressure boundary feedthrough. On the atmospheric side of the tube, the motion meets either an electronic angle sensor producing 4 to 20 mA, or, in legacy pneumatic builds, a flapper nozzle producing 3 to 15 psi. Range depends on the displacer weight and cross section: a fatter or heavier cylinder produces less angular travel for the same level change, and catalogs pair displacer geometry with tube stiffness to set each span. Order displacers as matched, documented sets, because a lookalike spare cylinder is not a calibration, it is a guess with hardware attached.

Interface duty measures the boundary between two liquid layers. With the displacer spanning both layers, the output follows the interface position only because the two densities differ: the force span across a full interface sweep equals displacer volume multiplied by the density difference between the liquids. Catalogs typically certify interface service down to a density difference of about 0.05 g/cm3, and 0.1 g/cm3 is the practical floor for a loop intended to be trusted, because every force error divides by that difference to become level error. As the two liquids converge in density, the instrument first loses accuracy, then resolution, then meaning, and the honest specification states the expected density difference from samples taken at operating temperature, not from a design basis table.

Density is also the error path, and it moves with temperature. Light hydrocarbons change density by roughly 0.6 to 0.8 kg/m3 per deg C, about 0.1 percent of reading per degree, so an uncompensated transmitter on an outdoor vessel swings 1 to 2 percent of span between night and day. Specify compensation: an integrated sensor, or a temperature from the thermowell by the cage fed to the transmitter or the control system as a density input, with the density versus temperature curve of the actual fluid in the datasheet. Steam drums are the extreme case, because saturated water density falls from about 958 kg/m3 near atmospheric saturation to roughly 735 kg/m3 at 80 bar: a drum transmitter calibrated on cold water compresses the true span by more than 20 percent at pressure, so drum loops carry saturation density compensation keyed to the drum pressure transmitter that every drum already has.

Force measurement earns drift by existing. Torque tube springs creep over thermal cycles, buildup adds weight to the displacer and shifts zero low, and temperature moves both the fluid and the instrument. The consequence is that a displacer installation is bought together with its verification method: equalizing and drain hardware for zero checks, a fill and drain procedure for span, and a scheduled density audit against samples. Chapters 4 and 6 turn those into datasheet rows and maintenance lines. What is being decided on this page is not whether buoyancy works, it does, but whether the bought configuration keeps its numbers between turnarounds.

Types, Construction, and Duty Boundaries

The external cage is the classic architecture: a vertical pipe chamber, DN50 or DN80, connected at the top to the vapor space and at the bottom to the liquid, with the displacer hanging inside on its torque tube assembly. Cage isolation valves, a drain, and a vent or equalizing arrangement, classically a three valve manifold, allow isolating, draining, and equalizing the cage for calibration checks without opening the vessel, which is the whole operational argument for the cage. Cages are built as pressure parts, hydrotested and stamped, with ratings from CL150 through CL2500 depending on build; the Fisher 249 family documents the architecture, including cageless variants that suspend the displacer directly in the vessel through a single nozzle, trading calibration access for fewer penetrations.

The cageless internal mount hangs the displacer on a flanged head through one top nozzle. It is chosen when nozzle real estate is scarce, when the service is fouling enough that a cage would slowly fill with solids and stop following the vessel, or when the vessel code makes a second penetration expensive. Its price is literal: any calibration or inspection work beyond an electrical zero check means a vessel entry or an isolation and blinding permit. Interface service and drum level service almost always pay for the cage, because those loops earn their keep through verification, and a measurement that cannot be checked between turnarounds quietly stops being trusted.

Materials follow the chemistry. 316L displacers and torque tubes are the catalog default and cover hydrocarbons, steam condensate, and most chemicals. Alloy 400 buys resistance in brine and seawater duty. Alloy C-276 serves sour and chloride acid service and is the standard answer when ISO 15156 governs material selection. Carbon steel cages with stainless internals keep oil service economical. Steam duty brings insulated cages, extended or air cooled torque tube housings, and elastomer free architecture at the hot end, all of which exist to keep the electronics within their rating while the process sits at saturation temperature.

The envelope is wide, which is the point of the technology. Process temperatures of minus 30 to 200 deg C cover standard builds, and steam duty variants with extended torque tube architecture work toward 340 deg C. Pressure ratings follow the cage hardware: CL2500 cages hold over 400 bar at ambient temperature, derating with heat exactly as the flange standards table. No other continuous level principle offers moving metal parts proven at that combination of pressure, temperature, and interface capability, and the limits that remain are mechanical, not electronic: span length, density difference, and fouling.

The boundaries deserve blunt statements. A displacer measures density weighted level, so any density change reads as a level change unless compensated. A cage in solids laden service plugs at the lower connection and then reports the cage, not the vessel. An interface span with 0.05 g/cm3 difference is a certificate, not a control loop. A 3,600 mm displacer is a pendulum: install it plumb or it touches the cage wall and turns every level reading into a friction reading. The table below maps common duties to the construction that survives them.

Key Specification Parameters

The rows below decide whether a quoted instrument matches the service; the paragraphs after the tables say what a complete datasheet states and what a weak one hides.

Several rows deserve prose. Span: state it in buoyancy terms, volume times density difference, not only in millimeters, because the force budget is what error divides into, and a 0.05 g/cm3 interface span has a force budget one tenth of a water level span on the same cylinder. Materials: demand the material certificates and, in sour service, the ISO 15156 compliance statement for every wetted part including the torque tube. Valves: the equalize and drain set is not an accessory, it is the calibration interface, and full port geometry is what keeps the cage following the vessel. Accuracy: the quotation must state reference conditions, fluid, density, and temperature, or the percentage is marketing.

The service mode table compresses the calibration decision for specification meetings.

Temperature compensation deserves its own datasheet line, because it is the difference between a reading and a fact. The compensation needs a temperature representative of the liquid in the cage, so the thermowell belongs adjacent to the lower cage connection, not on a distant line where steam tracing or sunlight lies to the calculation. Where the transmitter accepts an external temperature input, state the source; where it computes internally, state the sensor location and its calibration. A weak quotation on this line is a single phrase, temperature compensated, with no curve, no source, and no stated residual error, and it should be sent back.

Selection Decision Factors and RFQ Language

The decision sequence is short enough to run as a checklist in a specification meeting, and each step eliminates a class of quotations before price comparisons begin.

1. Fix the service mode: level, interface, or density output, because everything downstream calibrates against this choice.
2. Write the fluids as they are, with densities at operating temperature from samples, not from the design basis datasheet inherited from another project.
3. For interface duty, compute the density difference honestly and reject the application if it sits near 0.05 g/cm3 and the loop matters.
4. Choose cage or cageless on verification needs: cages buy in service checks at the price of two nozzles and freeze protection.
5. Size the displacer so the span of interest covers 60 to 90 percent of its length, keeping signal budget away from both mechanical ends.
6. Set the pressure and temperature envelope, then pick cage class, torque tube architecture, and materials with certificates.
7. Specify density compensation explicitly: curve, source, and residual error, with the thermowell location named.
8. Fix the verification regime: equalize check interval, fill and drain span check at turnaround, and proof test path for any safety function.

Measured against the modern families, honestly and only here in the selection chapter: a guided wave radar level probe has no moving parts, needs no density assumption, and installs on a single top nozzle; its weaknesses are heavy coating on the probe, emulsion layers that scatter the pulse, and the mechanical realities of long probes in tall cages. A radar level meter is non-contact and is the default new build choice for clean vapor spaces; foam, tall narrow nozzles, and heavy vapor can still defeat the reflection. The hydrostatic level transmitter is the historical rival at the same vessel: built on a differential pressure transmitter, simple and proven, but density dependent by the same physics as the displacer, and in hot service it drags impulse lines or a diaphragm seal whose fill fluid limits temperature and shifts zero with ambient; fitted with a diaphragm seal on both taps it climbs in cost until the comparison resets. An ultrasonic level meter works only on open, atmospheric vessels and shares vapor sensitivity. A capacitance level transmitter fights buildup on its probe. A magnetostrictive level transmitter in a bypass cage delivers millimeter class accuracy in clean liquids but needs a clean float path. The displacer keeps four defensible grounds: interface duty with a real density difference, density and concentration as a native output, pressure and temperature ratings proven at CL2500 separator and drum service, and retrofit into the enormous installed base of cages, where a modern electronic head upgrades the loop without a single new nozzle.

Two purchase patterns follow from that comparison. Retrofit: many drums and separators already carry cages and torque tube hardware, and replacing the head while keeping the documented displacer is the cheapest defensible modernization on site, provided the original displacer volume and weight records still exist; order the records search before the RFQ. Interface: where the density difference is real, 0.1 g/cm3 and up, the displacer measures the boundary directly and predictably, while a probe reading the upper layer needs an inference and a tank gauging approach needs stillness. Where the vessel already carries a differential pressure transmitter on the lower taps, compare quotes on total cost of verification and drift between turnarounds, not on tag price.

The application map turns the reasoning into a table the project file can carry.

The RFQ block should be written once and reused. A complete block reads: quantity and service mode; span with the band of interest marked; fluids with densities at operating temperature, from samples, and density difference for interface duty; mounting, cage or cageless, with connection size, class, and facing; displacer diameter, length, and material; cage material and rating with hydrotest certificate; torque tube architecture and rating; output with fail behavior and HART where used; accuracy statement against the actual fluids with reference conditions; density compensation curve, temperature source, and residual error; valve and manifold arrangement; material certificates, ISO 15156 statement where sour; and acceptance as a wet calibration witnessed at FAT with the documentation of the displacer set, volume, weight, and matched calibration constants. A quotation returned without the span in buoyancy terms, the reference conditions, or the compensation curve is incomplete at any price.

Installation, Standards, Failure Modes, and Maintenance

Installation quality decides whether the cage reads the vessel or reads itself. Set the cage plumb within about 2 degrees, because a 3,600 mm displacer swinging even 10 mm sideways touches the wall and adds friction to the measurement. Align both cage connections to true vessel elevations, upper in the vapor space, lower in the liquid band, with full bore piping and no pockets where gas can collect on the liquid leg; a vapor locked lower connection freezes the reading mid range while the vessel drains underneath it. Thermal growth between the two nozzles must go somewhere: support the cage on one nozzle and let the piping flex, or the vessel teaches the cage its stress. Heat trace and insulate water and condensate service, ground and bond the loop properly, and orient the torque tube housing so its electrical head stays out of the weather and the drip path.

Commissioning proves the loop with liquid. Fill the cage through the vent in measured steps and record output versus fill height in percent of span, the two point check that catches a transposed range constant or a damaged displacer before the plant hands over. Verify the zero with the equalizing arrangement, confirm the fail behavior of the electronics by breaking the loop, and cross check the installed reading against the gauge glass or magnetic level gauge on the same vessel. Document the reference densities used, because every later argument about drift resolves against that piece of paper.

The standards set the verification frame. ASME Section I paragraph PG-60 governs drum level indication and alarms on power boilers, including what hardware the code requires beside the transmitter. API 2350 frames overfill management for storage, the layer of protection context in which any tank level instrument participates. IEC 60770 practice underlies comparable performance data, IEC 61508 assessment covers any output feeding a safety function, ISO 15156, formerly NACE MR0175, governs materials in sour service, and the pressure boundary carries PED conformity where the plant sits under it. Citing these by name in the RFQ moves the burden of proof to the supplier, which is where it belongs.

Failure modes follow a short and litigious list. Torque tube fatigue is the classic: years of thermal and pressure cycling end in a broken tube, a pinned or wildly unstable signal, and an unplanned unit trip, and the warning sign is span drift between annual checks, which is why span trend records exist. Displacer buildup adds weight and shifts zero low, the signature in produced water and heavy oil, and it is caught by weighing the displacer against its record card at turnaround. A plugged lower connection freezes the cage, the separator complaint, and the countermeasures are flushing, heat tracing, and roddable connections. A pinhole in a welded displacer slowly fills with process liquid, the buoyant force collapses, and the reading walks to permanent low; the fix is inspection and a matched spare. Vapor lock, freezing of unprotected legs, and mechanical overrange from slugs complete the list. Every one of these modes announces itself early in the zero and span record, which is the entire argument for keeping one.

Maintenance is verification on a schedule. Annually: equalize zero check with the record signed, span trend reviewed against previous years, and the compensation temperature source audited. Each turnaround: pull the displacer, weigh it against its record, inspect the torque tube and bearings, clean the cage, and recalibrate with a witnessed fill test; sample the process liquid for a density audit read together with the cage thermowell. Keep the matched spare displacer and its calibration constants in stores, because the matched set is the calibration. Where the transmitter feeds a safety narrative, proof test the alarm path through an independent level switch and record the cross check. A displacer installation that keeps its numbers between turnarounds was bought with its verification method attached, and one that does not was bought on price.