In the dosage measurement application, the irradiance is accumulated over time and the total energy density in [J/cm2] is measured. This is the true measurement of the number of photons reaching an area to perform the curing process. If the UV source is scanning over the work area, as is the case in UV 3D printers, or if the work area is a conveyor passing underneath a static UV source, as is often the case in large UV curing machines, the sensor should be able to sample at a high rate and track the changes in irradiation (see Fig. 3). Good cosine response is also important because the sensor will be illuminated from different directions due to the motion of the source or the sensor.
In printing, curing, and 3D printing of polymers, LEDs in the UVA range (mostly 365-395 nm) are used. Some applications also make use of violet LEDs at 405-425 nm. Unlike the fixed and narrow UV spectral lines of traditional mercury lamps at 254 nm, 365 nm, and 405 nm, LEDs have a spectral bandwidth of 10-25 nm. Their peak wavelength is specified with an uncertainty of several nanometers and tends to shift when they heat up.
The first graph in Fig 4 shows the spectra of three different LED sources with the same peak wavelength and different spectral bandwidths (a). The second graph shows three different LED sources with same spectral widths but different peak wavelengths (b). With the third graph, the spectra of three different LED sources are presented, each with a specific peak wavelength and different spectral bandwidth (c).
To see how this affects measurement accuracy, we need to consider three types of irradiance sensors.
Sensor without spectral response calibration Sensor with spectral response calibration Sensor with flat spectral response