TDR (time-domain reflectometry) level meters use a low-power nanosecond pulse launched along a metallic probe immersed in the process fluid; the reflected echo from the media interface is timed to derive distance, with stated accuracy commonly inside ±2 mm and repeatability inside ±1 mm for top-tier OEM units [S2].
For 2026 specifiers, the relevant decision is no longer "guided-wave radar versus everything" but "which probe architecture, which validation regime, and which Dk (relative dielectric constant) window the duty can guarantee." Buyers working on chemical, pharmaceutical, food and bulk-solid duties should treat TDR as the default for interface and low-DK liquid service, not as a general radar replacement; reference comparison work for adjacent technologies is collected in TDR Level Meter vs Level Transmitter: Spec-First Decision Guide for 2026 Specifiers.
Probe architectures and their real operating envelopes
Coaxial (still-well) probes confine the electromagnetic field inside a perforated outer tube and deliver the strongest signal return, allowing operation on media with Dk as low as ~1.4 and tolerating low-viscosity hydrocarbons, LPG and liquefied gases with stated process temperatures from -196 °C up to +450 °C and pressures to 160 bar on competing non-contact radar units in the same catalogue band [S2]. The trade-off is mechanical: coaxial probes trap process, complicate CIP (clean-in-place) on hygienic lines, and cannot be used in vessels with internal obstructions.
Single-rod and single-wire probes tolerate build-up, accept side-mounted fittings and survive longer insertion lengths, with VEGAFLEX 83 type units rated 0 m to 32 m level range, process pressure 0 bar to 16 bar, and process temperature -40 °C to +150 °C, and they remain the default pick for aggressive media and pharmaceutical interface service [S2]. Twin-rod probes sit between the two on signal strength and mechanical robustness; flexible cable probes (rope) extend the 32 m envelope to deep silos, but at the cost of reduced accuracy in the top blanking zone and susceptibility to side-load failure in agitated tanks.
The classification of probe type governs the maximum measuring range, the minimum Dk that returns a stable echo, and the response to foam, turbulence and condensation on the probe shaft, and these are the three variables the spec sheet must lock before a vendor shortlist is even opened [S2].
Process variables that decide TDR suitability
Dielectric constant of the medium is the single most important gate: a Dk of 1.4 to 1.6 is the practical lower limit for rod or cable probes, while coaxial probes can resolve down to Dk ~1.2 for hydrocarbon service, and engineered interface applications (oil over water, foam over liquid) require a Dk differential of at least 10 between the two layers to produce a clean second echo [S2].
Process pressure and temperature drive probe material selection and gasket class.
Foam, turbulence, agitator speed, vapour and condensation are the second-order variables: TDR rides through these better than ultrasonic and free-space radar because the field is guided, but a coating-prone medium (oil, latex, crystallising brine) will still shift the apparent interface unless the probe is specified with a coaxial shield or periodic auto-falsification diagnostics.
Who should pick TDR, and who should not
TDR is the right call for interface measurement in separators and desalters, for low-DK hydrocarbon tanks (LPG, butane, light solvents), for hygienic interface measurement in pharmaceutical reactors, and for high-pressure narrow-nozzle installations on chemical reactors where free-space radar loses too much energy [S2]. For 80 GHz non-contact radar versus TDR on liquids, the GWR-side comparison is laid out in GWR vs TDR Level Meter: 2026 Spec-First Selection, and a broader 25 GHz versus TDR versus portable radar buying view is in Radar Level Meter 2026 Buying Guide: 25 GHz vs TDR vs Portable.
Skip TDR for low-density bulk solids below ~20 kg/m³, where the echo is too weak on rod/cable probes and free-space radar is the only credible option; for very tall bulk silos above 32 m, where flexible cable probes stretch accuracy beyond useful limits; and for CIP-only hygienic lines that demand a flush, drainable, dead-leg-free installation, where a tuned coaxial probe only works if the line is straight and the gasket face is food-grade polished.
Criteria-based comparison of the main probe types
The four realistic TDR probe options line up against four buyer-side criteria as follows. Coaxial probes: minimum Dk ~1.2, maximum recommended length ~6 m, build-up tolerance low, hygienic suitability medium (cleanable but not drainable). Single-rod probes: minimum Dk ~1.6, maximum length ~32 m, build-up tolerance high, hygienic suitability high with polished finish. Twin-rod probes: minimum Dk ~1.4, maximum length ~32 m, build-up tolerance medium, hygienic suitability medium. Flexible cable (rope) probes: minimum Dk ~1.6, maximum length ~32 m (rod) and 75 m (cable, reduced accuracy), build-up tolerance low to medium, hygienic suitability low (rod variant only). [S2]
These envelopes are the same envelope family that published vendor data uses for adjacent radar families in the same specification band (VEGAPULS 6X 0 m to 120 m, -1 bar to 160 bar, -196 °C to +450 °C), so the comparison is grounded in real catalogue data rather than generic claims [S2].
Output, signal integrity and integration
[S2]
For hygienic interface service, the validator should also check 3-A, EHEDG and ASME BPE gasket and surface-finish documents, and the mechanical drawing should show the dead-leg length below the process connection, because a >2×DN dead leg in a pharmaceutical or food line will fail a CIP review even if the electronics are perfect. The signal side carries its own risk: foam thicker than ~50 mm at the probe entry can swallow the launch pulse on rod/cable variants, and the OEM software must support a top dead-zone blanking map matched to the actual probe length, not the nominal catalogue value.
Standards, validation and what to demand on the datasheet
For explosion-proof service, datasheets should reference IEC 60079 series for the zone classification (typically Zone 0/1 inside the tank, Zone 1/2 at the housing) and ATEX category 1G/2G marking, with IECEx for markets outside Europe and FM/UL for North American sites; hygienic lines should additionally cite 3-A Sanitary Standard 74- and EHEDG Doc. 2 or Doc. 8 for cleanability, and the wetted material certificates should follow EN 10204 3.1 for the probe shaft and 3.2 for the gasket compound lot. [S2]
For overfill prevention, the functional-safety certificate should state SIL 2 (or SIL 3 on dual-channel configurations) under IEC 61508, with a documented SFF (safe failure fraction) above 90% and a proven PFDavg below 1×10⁻³ for the safety loop, not a generic "SIL capable" claim. Buyers who lock the standard reference on the purchase order rather than on a marketing PDF avoid the most common acceptance-test failure mode, which is the gap between the certification the OEM holds and the certification the EPC contractor stamped on the IO list.
For complementary selection work on adjacent level technologies, see Capacitance level transmitter selection: a spec-first map and Radar Level Meter vs Level Transmitter: Spec-First Selection for Liquids and Bulk Solids.
A useful TDR selection question is also answered on the encyclopedia side, where TDR level meter gives the technology overview and radar level meter covers the non-contact sibling; for non-radar comparison work the ultrasonic level meter entry sets the contrast in physical principle and minimum-Dk window.
Shortlist logic for 2026: lock Dk window and probe length first, eliminate anything that cannot hit the Dk minimum on a single-rod or coaxial architecture, then filter by process pressure/temperature envelope, then by hygienic/ATEX/SIL certification, then by protocol and total cost of ownership over a 10-year life cycle; this order, not vendor familiarity, is the difference between a TDR that holds calibration and one that drifts within 12 months.