UV radiometer selection is dominated by three engineering decisions: spectral band match, detector class, and cosine/directional error budget, with the governing field practice codified in ASTM G183-15 [S3].
Outdoor solar monitoring, accelerated weathering per ISO 9370, and disinfection-lamp process control each pull the spec in different directions, so a single broadband instrument almost never covers all three duties [S3].
Spectral Band and Detector Class
UV-A (315-400 nm), UV-B (280-315 nm), and UV-C (100-280 nm) require separate filtering and detector chemistries; a UV-A silicon-carbide (SiC) photodiode has near-zero responsivity below 280 nm, while a UV-C GaP or NiMo hybrid must suppress visible bleed [S3]. ISO 9060 classifies pyranometers by class (secondary standard, first class, second class) and the same instrument-class logic carries across to UV radiometers for accuracy tiering.
For solar UV index (UVI) reporting the World Meteorological Organization Guide (WMO-No. 8) pairs a broadband UV-A/UV-B detector with a tailored spectral response weighted to the CIE erythema action spectrum, not a flat radiometric response [S3].
Cosine Response, Linearity, and Calibration Tier
Cosine correction at 60° from normal should stay within ±5 % for general outdoor work and ±2 % for primary reference stations; departures above ~10° from normal dominate the error budget on clear days [S3].
Calibration traceability is the single spec that separates a field radiometer from a laboratory instrument. ASTM G183-15 specifies verification frequency, field intercomparison against a reference radiometer, and that laboratory radiometric work on weathering light sources fall under ISO 9370 [S3].
Duty Match: Outdoor Solar vs Accelerated Weathering vs Process UV
Outdoor solar monitoring under ASTM G183-15 needs WMO-traceable calibration, Class A cosine response, and a desiccant or heated head to suppress dew on the diffuser; accelerated weathering per ISO 9370 calls for narrowband radiometers tuned to the specific xenon-arc or fluorescent lamp spectrum being used [S3].
Process UV duties (germicidal 254 nm, curing 365 nm, or 222 nm Far-UV-C) are a separate category: the source is a known line, and a filtered narrowband detector sized to the line gives a far better signal-to-noise than a broadband device. This is the same sensor-class logic used in weather station selection, where matching the transducer to the measurand beats a generic instrument every time.
Signal Output, Power, and Environmental Specs
Field solar radiometers typically output 0-20 mV analog or 4-20 mA current loop; networked weather stations add Modbus RTU/TCP or SDI-12 to the same head, which lets one logger pull wind, temperature, and irradiance from a single bus. Process UV heads add 0-10 V or RS-485 Modbus for line integration. [S3]
Ingress protection needs to be specified explicitly: IP65 is the floor for permanent outdoor mounting, and a heater draw of 1-5 W is common for dew suppression in humid coastal sites. Operating temperature range for outdoor silicon-carbide heads usually runs -30 to +70 °C, with active cooling required above that ceiling.
Comparison: Four Common UV Radiometer Configurations
1) Broadband UV-A SiC head: spectral range 315-400 nm, cosine ±5 %, calibration to WMO traceable source, best for solar UVI and materials-weathering screens. 2) Erythema-weighted UV-B/UV-A: WMO-style weighted, cosine ±3 %, used at meteorological stations for public UVI broadcast. 3) Narrowband 254 nm UV-C process head: filtered photodiode, cosine not critical (collimated source), traceable to NIST/FIZ calibration, used in germicidal lamp control. 4) Electrically calibrated radiometer (ECR): pyroelectric, NIST-traceable absolute responsivity, intended for laboratory lamp characterisation under ISO 9370 [S3].
Specifications like spectral range, cosine error, and calibration standard are non-negotiable; interface protocol, mounting, and data logger are configurable around the same head. The same hierarchy applies to a compact PLC selection, where I/O count and fieldbus type are gated by the application, not by marketing preference.
Who Should NOT Buy a Generic Broadband UV Meter
Anyone with a regulatory or process target on a specific photobiological action spectrum should not buy a flat-response UV-A meter. A broadband 320-400 nm sensor cannot discriminate UVB erythemal dose from UVA, and a non-weighted 254 nm probe cannot be used to verify germicidal dose without a calibration factor tied to the specific lamp's spectral line shape [S3].
For solar PV plant monitoring, the same principle that drives pyranometer selection in ISO 9060 (secondary standard vs first class) applies: the radiometer class is set by the uncertainty budget demanded, not the budget. As with the spec-first map laid out in the temperature monitoring device selection guide, the right instrument is the one whose accuracy class matches the application, not the one with the best datasheet headline.
Standards to Anchor the Spec
ASTM G183-15 governs deployment, maintenance, verification, and calibration frequency for field radiometers and is the primary anchor for outdoor use [S3]. ISO 9060 sets the instrument-class framework borrowed from solar pyranometry, and ISO 9370 covers instrumental determination of radiant exposure in laboratory weathering tests. The WMO Guide to Meteorological Instruments and Methods of Observation (WMO-No. 8) governs the UVI reporting chain.
Recalibration interval is typically 12 months for field solar radiometers and 6 months for high-exposure process UV heads, with field intercomparison against a transfer-standard instrument between formal calibrations per ASTM G183-15 [S3].
Next nodes to chase before spec sign-off: request the manufacturer's cosine-response curve (not just the spec-sheet number), ask for the calibration certificate naming the reference standard (WMO, NIST, or PTB traceable), and confirm whether the supplied spectral response is measured at the diffuser input or computed from the bare detector. If the answers match your duty, you can shortlist two heads and run a side-by-side intercomparison before commit.
For component-level specifications, see uv radiometer, pressure transmitter, and flow meter.