In penetrant testing and magnetic particle testing, the human eye decides between a finding and a missed indication. Whether an indication is seen depends directly on the UV-A irradiance and on the amount of white light at the inspection station.

ISO 3059: measurement, instruments and calibration

ISO 3059 defines which values have to be met, how they are measured, and which requirements the measuring instruments have to fulfil. This page deals with measurement and calibration. The method itself, the UV sources and the process parameters are covered on the page Fluorescent penetrant testing & industrial inspection.

ISO 3059 (current European adoption: EN ISO 3059:2012; German edition DIN EN ISO 3059:2013-03) covers the control of viewing conditions for magnetic particle testing and penetrant testing. It sets out the minimum requirements for illuminance and UV-A irradiance and for their measurement, and it applies wherever the human eye is the primary means of detection. It does not apply to purely camera-based, automated evaluation systems, nor does it cover actinic blue light sources.

ISO 3059 is a cross-cutting document: both ISO 3452-1 (penetrant testing) and ISO 9934-1 (magnetic particle testing) refer to it. Anyone inspecting to either of those methods is therefore also inspecting to ISO 3059.

The standard distinguishes two evaluation methods:

• Colour contrast method (non-fluorescent): sufficient white light is needed for reliable contrast.
• Fluorescent methods: sufficient UV-A irradiance is needed, combined with as little white light as possible.

In penetrant testing, a combination of very high UV-A irradiance and long exposure time should be avoided. As a general guideline value the standard states an irradiance of no more than approximately 5,000 µW/cm² (50 W/m²).

Two points are regularly overlooked in practice. The 20 lx figure is a total value: it covers the residual light in the room and the visible light component emitted by the UV-A lamp itself. An inadequately filtered lamp can use up this budget on its own. And all values apply at the test surface – not at some arbitrary distance beneath the lamp.

For the white light source used with the colour contrast method, the standard additionally specifies a colour temperature of at least 2,500 K, preferably above 3,300 K.

Depending on the inspection method, two different quantities are relevant.

Illuminance E_v in lux (lx) describes the effect of visible radiation weighted according to the spectral sensitivity of the human eye. It is required in particular for colour contrast penetrant testing and non-fluorescent magnetic particle testing.

UV-A irradiance E_e in W/m² or µW/cm² describes the optical radiant power incident on a surface. In fluorescent penetrant and magnetic particle testing, the UV-A irradiance in the region around 365 nm is the decisive excitation quantity.

Both quantities must be determined directly at the relevant test surface and under actual working conditions.

ISO 3059 specifies a UV-A source with its intensity maximum at 365 ± 5 nm. The spectral bandwidth at half maximum should not exceed 30 nm. A UV LED with its maximum at 385 nm or 395 nm therefore does not meet the standard – regardless of how high the measured irradiance is. Such sources excite fluorescent test media less efficiently while at the same time increasing the visible light component.

Accordingly, not every UV meter is suitable for an assessment that conforms to the standard. The UV-A radiometer must have a defined spectral responsivity. A suitable sensor has to meet the following requirements, among others:

• maximum of the relative spectral responsivity between 355 and 375 nm
• relative responsivity at 313 nm below 10 %
• relative responsivity at 405 nm below 2 %
• defined limitation of the relative spectral responsivity across the entire measuring range

A measuring system suitable for ISO 3059 should have the following characteristics in particular:

• defined spectral UV-A responsivity
• a measuring range up to at least 5,000 µW/cm²
• good linearity across the relevant measuring range – check points are typically 1,000, 3,000 and 5,000 µW/cm²
• traceable calibration
• illuminance measurement with appropriate photometric weighting
• cosine-corrected input optics with a stated f₂ value – without diffuse input optics, obliquely incident radiation is systematically under-recorded
• a lux meter with a stated f₁′ value for the match to V(λ): the smaller the value, the smaller the error under LED light
• adequate resolution at the low end of the lux range – the 20 lx limit lies far below the range standard lux meters are designed for; an instrument with 1 lx resolution cannot provide a meaningful assessment here
• a stated measurement uncertainty – only this makes it possible to judge whether a reading just above the limit actually demonstrates compliance

How a sensor's spectral responsivity affects the reading is covered in detail under spectral mismatch of UV sensors; for assessing borderline readings see measurement uncertainty in UV metrology.

The Opsytec RMD Pro is a digital radiometer for irradiance and illuminance measurements. Different calibrated sensors can be connected to the same display unit, and two sensors can be read out simultaneously. Sensor identification and calibration data are stored in the sensor itself.

For measurement tasks to ISO 3059, the RMD Pro is combined with a suitably specified UV-A sensor and a lux sensor. Both inspection quantities are then covered by one system: UV-A irradiance for monitoring fluorescent inspection, and illuminance for monitoring visible ambient light as well as for colour contrast penetrant testing and non-fluorescent magnetic particle testing.

Because both sensors are read out at the same time, the readings are taken at the same point and in the same operating state. This avoids a common error: measuring UV-A and white light one after the other under slightly different conditions.

What matters for a measurement conforming to the standard is not the display unit alone, but the spectral characteristics, the measuring range and the valid calibration status of the sensor in use.

ISO 3059 requires measurement under actual working conditions. The sensor position must therefore correspond to the position of the surface that will later be evaluated – at the actual inspection location and at the intended working distance. The ASTM methods additionally specify a measuring distance of 380 mm.

1. Let the lamp stabilise. Light sources do not reach a stable state immediately. With mercury vapour lamps, wait at least 10 minutes after switch-on before measuring.
2. Align the sensor perpendicular to the direction of radiation. Tilting noticeably reduces the reading; a sensor without cosine correction distorts it further.
3. Locate the maximum and define the measuring point. UV-A lamps have a pronounced spatial distribution. Only a documented measuring point makes the measurement reproducible.
4. Determine the usable field. Record the area over which the required 1,000 µW/cm² is maintained – this is the inspection area you can actually use.
5. Assess white light separately. The lux meter reading must not be affected by the UV-A radiation; only then can the permissible residual white light be reliably demonstrated.
6. Check battery-powered hand lamps towards the end of their operating period, not only when fully charged.
7. Document. Reading, date, instrument including calibration status and inspection station belong in the record.

ISO 3059 governs the measurement itself; how frequently measurements are taken follows from the method standard, the customer specification and the in-house written procedure. In the aerospace sector (Nadcap, EN 4179 / NAS 410, SAE AMS 2647, OEM specifications), requirements are frequently stricter than the minimum requirements of the standard – in particular additional upper limits for UV-A irradiance and shorter intervals.

Viewing conditions are not only a matter of equipment. Dark adaptation of the eye must be allowed for before inspection begins – ISO 3059 states a minimum of one minute. Photochromic lenses must not be worn, as they darken under UV-A and impair the visibility of indications. UV-B and UV-C components must be effectively filtered out at the radiation source.

With the colour contrast method, illuminance is determined using a lux meter whose spectral weighting matches the photopic sensitivity of the eye. Illumination of the test surface should be uniform; glare and reflections are to be avoided.

Regular calibration is a central element of ISO 3059. The working range of UV-A irradiance meters and illuminance meters must be calibrated at the intervals recommended by the manufacturer; that period must not exceed twelve months. Compared with the previous edition of the standard, this verification interval was explicitly shortened to twelve months.

For UV-A measurement, the standard imposes additional requirements:

• calibration using narrow-band radiation at 365 nm
• metrological traceability to national, European or international standards
• calibration of the relevant working range, at several points rather than at a single point – for NDT applications typically from about 100 to 10,000 µW/cm²
• documentation by means of a calibration certificate
• a stated measurement uncertainty in the calibration certificate – without it the evidence is incomplete: at a reading of 1,020 µW/cm² with an uncertainty of ± 8 %, compliance with 1,000 µW/cm² is not demonstrated
• recalibration after damage or relevant maintenance work, as well as in the event of implausible readings, a noticeable deviation from a comparison instrument, or prolonged storage outside the specified conditions

Where a measuring system consists of a detachable sensor and a separate display unit, the calibration must cover the complete measuring chain of sensor and display unit. On setting the interval in your own operation, see calibration interval for UV meters.

ISO 3059 requires a traceable calibration and its documentation, but does not as a matter of principle require a laboratory accredited to ISO/IEC 17025. An accredited calibration does, however, provide more far-reaching evidence of competence and a measurement uncertainty assessed within the accredited scope; for testing laboratories that are themselves accredited, it is the recognised route.

Opsytec Dr. Gröbel operates its own calibration laboratory for UV meters and UV sensors. In addition to manufacturer calibrations, accredited calibrations to ISO/IEC 17025 are available for suitable measurands and instruments – both for Opsytec instruments and for numerous suitable UV meters from other manufacturers.

A UV radiometer always evaluates a spectrum according to its own spectral responsivity. Two sensors can therefore display different readings for the same radiation source even though both are working correctly.

For ISO 3059, UV-A measurement is deliberately aligned with the spectrum used in fluorescent inspection. This is why the standard requires narrow-band radiation at 365 nm for calibrating the UV-A radiometer – traceability is thereby established directly at the wavelength that governs the inspection.

This is particularly relevant when switching from classic mercury vapour lamps to UV LED inspection lamps: differing source spectra lead to spectral mismatch in broadband radiometers. An application-specific calibration reduces this influence.

A single acceptance measurement is not sufficient to ensure reproducible inspection conditions over time. UV-A irradiance can change through ageing of the UV lamp or UV LED, contamination of filters and protective screens, changes to reflectors, failure of individual LEDs, a change in working distance, mechanical changes to the lamp position, and temperature changes.

The measuring instrument itself is affected in the same way: contamination or scratching of the sensor diffuser lowers the reading unnoticed and makes an intact inspection station appear defective.

In addition to the annual calibration of the instrument, regular in-house monitoring of the actual inspection conditions is therefore advisable. Measurements should always be taken and documented at defined reference points. On larger inspection areas, several positions have to be recorded in order to assess the spatial uniformity of the irradiance as well.

A revision of ISO 3059 is currently in progress at ISO; publication is expected in 2026. The driver is the widespread adoption of UV LED inspection lamps. The announced focal points are more precise requirements for the peak wavelength around 365 nm, narrowly defined limits for the spectral bandwidth at half maximum, requirements for radiation stability over the operating period, requirements for monitoring battery-powered UV LED lamps, defined minimum working distances, and rules excluding pulsed UV LEDs and those operated with significant ripple.

For measurement technology this means that, alongside irradiance alone, spectral characterisation of the radiation source is gaining importance. Anyone investing today should check whether the peak wavelength and spectral bandwidth of the lamps in use can be reliably demonstrated.

What UV-A irradiance does ISO 3059 require?

For the actual fluorescent inspection, the UV-A irradiance at the test surface must be at least 10 W/m², i.e. 1,000 µW/cm². At the same time, the visible illuminance must not exceed 20 lx.

Which wavelength is used for fluorescent inspection?

ISO 3059 specifies a UV-A source with its intensity maximum at 365 ± 5 nm.

Does a 395 nm UV LED meet the standard if the irradiance is high enough?

No. The standard requires an intensity maximum at 365 ± 5 nm. A higher irradiance does not compensate for the wrong wavelength.

How often must a UV-A meter be calibrated?

The calibration interval follows the manufacturer's specification in the first instance, but must not exceed twelve months. Recalibration is required after damage or relevant maintenance work.

Does the calibration have to be performed at 365 nm?

Yes. Under ISO 3059, the UV-A irradiance meter is to be calibrated using narrow-band radiation at 365 nm.

Is an ordinary UV-A meter sufficient?

Not necessarily. ISO 3059 sets specific requirements for the spectral responsivity of the UV-A radiometer. A general-purpose UV-A sensor may well be suitable for other applications without meeting the spectral weighting required for fluorescent inspection.

Why does illuminance have to be measured as well?

In fluorescent methods, visible ambient light reduces the contrast between the fluorescent indication and the background. The standard therefore limits illuminance during inspection to a maximum of 20 lx. With colour contrast methods, by contrast, adequate white light illumination is required.

Does ISO 3059 also apply to automated camera systems?

No. The standard applies where the human eye is the primary means of detection.

Is an accredited ISO 17025 calibration mandatory under ISO 3059?

No. What is required is a traceable calibration; an accredited ISO 17025 calibration is the recognised proof of this and is customary in accredited testing laboratories.

When will the new edition of the standard be published?

The revision is in progress at ISO; publication is expected in 2026. The focus is on requirements for UV LED lamps.

• Fluorescent penetrant testing & industrial inspection – method, UV sources and process parameters
• Spectral mismatch of UV sensors
• Measurement uncertainty in UV metrology
• Calibration interval for UV meters
• Calibration laboratory of Opsytec Dr. Gröbel