Updated: April 22, 2026
Measuring level inside a blast furnace shaft is one of the hardest jobs in process instrumentation. The environment combines red-hot gas, dust clouds, pressure pulses, and electromagnetic noise. Guided-wave probes and mechanical plumb-bobs wear out in months. An 80 GHz or 120 GHz non-contact radar, installed with the right purge and alignment, is the only sensor that survives and delivers continuous burden-profile data. This guide explains what the radar must do, why frequency matters, and how the purge system keeps it alive.
Contents
- Can You Measure Blast Furnace Level With Radar?
- What Makes Blast Furnace Level Measurement So Hard?
- Which Radar Frequency Works on a Blast Furnace?
- Design Features a Blast Furnace Radar Must Have
- Purge & Air-Cooling System Design
- Installation Geometry & Signal Path
- Radar Level Transmitters for High-Temperature Service
- FAQ
Can You Measure Blast Furnace Level With Radar?
Yes — high-frequency FMCW radar (80 GHz or 120 GHz) is the only level sensor proven to run continuously inside a blast furnace shaft. It works because radar is immune to dust, steam, and thermal noise that blind optical, mechanical, or capacitive systems. The radar installed with an air or nitrogen purge and ceramic-window isolation reports burden profile within ±30 mm at a measuring range of up to 30 m.
Mechanical plumb-bob gauges and radioisotope systems are still used on older furnaces, but they are single-point or require licensed sources. A 120 GHz radar sweeps the full burden cross-section when scanning-aimed, and is maintenance-free between scheduled refractory shutdowns.
What Makes Blast Furnace Level Measurement So Hard?
Five conditions combine to kill most level instruments.
- Process temperature 150-250 °C at the sensor flange — much higher inside the vessel, with thermal radiation load on any exposed surface.
- Dust loading of 10-30 g/m³ in the top-gas space, fouling windows and antennas within days.
- Internal pressure 2-3 barg with pressure pulses during charging.
- Coke oven gas (CO, H₂) at reducing atmosphere — oxygen-sensitive sealants fail.
- Charge surface irregularity — the burden is sloped, asymmetric, and moves as material is charged and descends.
A radar needs to see through the dust, not collect it on a lens. It needs a flange seal rated for reducing atmosphere. And it needs a narrow antenna beam to scan the sloped burden profile, not just report one point. Miss any of these and the instrument either fails early or reports misleading data to the charging system.
Which Radar Frequency Works on a Blast Furnace?
80 GHz or 120 GHz FMCW radar is the only frequency band that produces a narrow-enough beam for blast furnace burden scanning. A 120 GHz radar with a 100 mm lens antenna gives a 2° beam angle — tight enough to aim at a specific burden zone and see through the 20 m shaft without false echoes from the walls.
Lower frequencies (6 or 26 GHz) were the first generation of radar used on blast furnaces and are still common for cheaper bin-level jobs. But at 26 GHz the beam opens to roughly 8-10°, which washes out the sloped burden into a single average and generates mirror echoes from the shaft wall. On a 10 m diameter furnace that is good enough for an interlock, not for burden-profile control.
| Frequency | Typical Beam Angle | Fit for Blast Furnace |
|---|---|---|
| 6 GHz (pulse) | 20-30° | Not suitable — hit by walls |
| 26 GHz (FMCW) | 8-10° | Single-point only, bin use |
| 80 GHz (FMCW) | 3-4° | Good — single or dual beam |
| 120 GHz (FMCW) | 1.5-2° | Best — burden-profile scanning |
Another 120 GHz advantage: the wavelength (2.5 mm) is close to the dust particle size, so strong dust clouds scatter less of the signal compared to what 26 GHz users expect. Field data from Chinese and Indian steelmakers published between 2022 and 2025 consistently show 120 GHz outperforming 80 GHz by 20-40% in heavy-dust campaigns. For background on the core technology, see our guided wave vs non-contact radar comparison.
Design Features a Blast Furnace Radar Must Have
A catalogue 80 GHz radar for silos will die within weeks on a blast furnace. The specific features below separate a bin-radar product from a blast-furnace-rated one.
| Required Feature | Typical Spec | Why It Matters |
|---|---|---|
| Antenna material | High-purity alumina ceramic | Survives 400 °C radiated heat, not damaged by alkali vapor |
| Process window | Alumina or borosilicate disk | Isolates the waveguide from process gas |
| Flange rating | PN40 / 300# or higher | Handles pressure pulses during charging |
| Sensor ambient rating | -40 to +80 °C at electronics | Electronics sit above the cooling flange |
| Air/N₂ purge port | G½” with flow meter | Keeps antenna surface clean |
| Scanning beam (optional) | Motor-aimed ±30° | Measures burden profile, not just one point |
| Signal processing | Multi-echo tracking | Rejects false echoes from shaft wall, ring deposits |
Motor-aimed (scanning) radars are relatively new. They sweep across the shaft cross-section every few seconds and build a 2D burden map. This is how modern automated charging systems stack coke and ore in shaped layers — the radar tells the chute where material has actually piled up.
Purge & Air-Cooling System Design
The purge system is not optional. Without continuous gas purge and flange cooling, the antenna window fouls in days and the electronics overheat. Build the purge loop with three jobs in mind.
- Window cleaning. Feed instrument air (or N₂ where a reducing atmosphere is guaranteed) at 40-80 Nm³/h through a tangential port below the window. The jet pattern sweeps dust outward.
- Thermal barrier. A secondary flow of 20-40 Nm³/h cools the flange, limiting the electronics ambient to below 80 °C.
- Back-pressure control. Include a pressure regulator with a local gauge and a low-flow alarm wired to the DCS. Losing purge means losing the instrument.
Instrument air must be dry and oil-free to ISO 8573-1 Class 2 or better. Oily air deposits carbon on the window at process temperature, and within a week the radar reports a “distance = window” false echo. For furnaces running hydrogen injection, nitrogen is mandatory to prevent explosive atmosphere inside the purge line.
Installation Geometry & Signal Path
Installation geometry decides whether the radar can see the burden or fights it. Work through these points before cutting the top-cone nozzle.
- Stand-off distance. Mount the antenna face at least 1.5 m above the top-cone to keep it out of the direct charging stream.
- Aim angle. Tilt 5-10° off-vertical so specular reflections from the top-cone walls do not re-enter the antenna.
- Clearance from wall. Keep the beam footprint at least 500 mm from any refractory wall at the measurement depth.
- Nozzle length. Use a short nozzle (≤300 mm). Long nozzles create multiple reflections that confuse echo tracking.
- Valve isolation. Include a DN80 or DN100 ball valve below the flange so the radar can be swapped during short shutdowns without losing furnace pressure.
The most common mistake is mounting a single radar right on the top centerline. The center position receives descending charge material and ring deposits build up fastest there. Off-center mounting with a small aim angle clears most of these issues and also aligns better with the sloped burden when using burden-profile scanning. For a broader view of radar mounting, compare the stilling-well approach for liquids — the logic is different because the blast furnace needs open beam scanning, not a contained waveguide.
Radar Level Transmitters for High-Temperature Service
64/80 GHz FMCW Level Radar
Narrow-beam FMCW, up to 120 m range, ceramic lens antenna. Suited for dusty silos and furnaces.
80 GHz Radar Level Transmitter
Compact 80 GHz unit, IP67, flange mount, 4-20 mA / HART output for high-temp solids.
Guided Wave Radar Transmitter
TDR probe alternative for hopper bottoms or secondary bins where non-contact is impractical.
FAQ
What temperature can a blast furnace radar tolerate?
The radar’s ceramic antenna tolerates up to 400 °C continuous at the process side, while the electronics stay below 80 °C thanks to flange cooling and the purge flow. The sensor is not mounted directly in the flame zone — it sees the burden surface from 1.5 m above the top-cone where gas temperatures are 150-250 °C.
Why is 120 GHz better than 26 GHz for blast furnaces?
A 120 GHz radar with a lens antenna produces a beam under 2° wide. A 26 GHz radar with the same physical antenna diameter produces an 8-10° beam, which hits the shaft wall and averages the burden profile into a meaningless single number. Narrow beam equals better burden mapping.
Do I really need air purge on the radar?
Yes. Without continuous purge, the antenna window fouls with dust and alkali condensate within days, and the electronics overheat from radiant load. Specify 40-80 Nm³/h instrument air (dry, oil-free) plus a flow alarm wired to the DCS.
Can one radar give me the full burden profile?
A single fixed-aim radar gives one point. For burden-profile data you need either a scanning (motor-aimed) radar or multiple fixed radars at 3-6 positions across the top-cone. Scanning radars are more common on new furnaces; multi-point fixed arrays are typical on retrofits.
How accurate is blast furnace radar level?
A correctly installed 120 GHz radar with a narrow beam resolves burden height to ±30 mm over a 25 m range. Accuracy degrades if the purge fails or if refractory ring deposits create ghost echoes that the signal processing cannot filter.
Quotation for a Blast-Furnace-Rated Radar
Send us your furnace top-cone drawing, nozzle size, burden diameter, and whether you need single-point or scanning burden profile. We’ll come back with frequency recommendations (80 vs 120 GHz), antenna size, purge specification, and delivery timeline.
Request a Quote
Wu Peng, born in 1980, is a highly respected and accomplished male engineer with extensive experience in the field of automation. With over 20 years of industry experience, Wu has made significant contributions to both academia and engineering projects.
Throughout his career, Wu Peng has participated in numerous national and international engineering projects. Some of his most notable projects include the development of an intelligent control system for oil refineries, the design of a cutting-edge distributed control system for petrochemical plants, and the optimization of control algorithms for natural gas pipelines.