Controlling UVC power with the L14U1 Sensor

UVC light germicidal properties are used in a variety of systems, ranging from air or water purification to hospital sanitization. The UVC light source can be either a gas lamp (mercury, Xenon Flash or Deuterium) or a solid state LED.

In all cases, it is paramount to control the UVC light intensity at all times to make sure that the disinfection process is optimum.

The main cause of UVC light output drop is ageing of the light source. If that drop is not compensated, the UVC light intensity would in time become too low to provide full disinfection.

This can be achieved by controlling the light intensity with a L14U1 sensor. This sensor features a high gain and a wide optical aperture. Those properties are ideal for the purpose of monitoring the UVC light radiated in a relatively wide space, which is generally the case in disinfection systems : the wide optical aperture provides an integration of the ambient UVC light level, and the high gain results in an electrical signal that is immune to noise and makes the design of the control circuit straightforward.

The most straightforward design consists in building a servo loop, where the L14U1 provides the negative feedback.

One example of such a design is shown in Fig.1 : in this example, the UVC light source is a solid state LED.

We assume that for this specific LED unit, and the specific configuration of the system (volume where the disinfection takes place, distance to the LED, attenuation factors, etc…) a driving current of 20 mA provides more than the minimum requirement of light intensity to achieve a full disinfection.

With that light intensity, and with this specific configuration, the L14U1 output current is 100 uA.

Note: this reading varies from unit to unit, depending on the gain of the unit.

Using resistors values R1 = 30 kOhm and R2 = 100 Ohm, Vin must be adjusted to 3 Volts to set the output current around 20 mA (the voltage drop in the LED is about 6 Volts).

Fine tuning of Vin around the 3 Volts value is necessary to adjust for the voltage drop actual reading.

Led Control Schematic
Fig. 1

A simulation of the system and of the decay in the light output has been run:

The L14U1 unit has been placed at a distance of 4 cm from the LED in a head to head configuration, with air in between.

An ampmeter and a voltmeter have been added to the circuit to read the current in the LED branch and the voltage on the negative input of the operational amplifier LM258.

Vin has been adjusted to result in a driving current of 20 mA.

Then the L14U1 unit has been moved away by steps from the LED (increasing the 4 cm distance) to simulate the decay of the LED output with time (as the distance increases the flux received on the sensitive area of the L14U1 decreases with an inverted square law).

At each step, the driving current in the LED increases and the voltage on the negative input does not change, which is the expected outcome. The driving current increase results in more total light being emitted by the LED, such that the flux on the sensitive area remains constant, despite the increase in distance.

Ultimately, the feedback effect of the servo loop ceased to be effective for a driving current of 28.6 mA, which is the maximum current that the LM258 can supply.

Similar circuits can be designed to work with other configurations (Lamps instead of LEDs, arrays of LEDs, different targets of light intensity). For each configuration, the choice of component values will have to be made taking into consideration the characteristics of the light source and the typical response of the sensor.

Then the initial setting of Vin could be made by the system controller with a programmed sequence taking as an input the measurement of the light output from a photometer, or the measurement of the driving current. This second option requires that the relationship between light output and current is well known and can be a simplified option when the margin between the effective light output and the required light output is wide.

Then the system would run until the servo loop runs into its limits. A periodic rerun of the initial setting routine could be chosen to detect the end of life or an additional current monitoring circuit could feed the controller and be compared with a preset limit.

Download PDF