Trim your cost. Opto-sensors in wearables and IoT applications


When you think of the expression ‘wearable technology’ the first thing that comes to you is for instance temperature measures, body fluids etc. But what you perhaps do not know is that light in fact plays a major role when you want to measure body functions like e.g. oxygen saturation. Today light can measure your physically health as well, by use of so-called chemical optical sensors. These sensors are placed inside your body.

The diode as a versatile sensor

The electronic component – also called a diode – is one of the key sensor elements when it comes to detect body measures. A diode is a component that, in theory, lets a current pass through in one direction but not the other. In the reverse direction, a small current – also called the leakage current – passes the wrong way through the diode. This leakage current is typically very small and often negligible.

A diode can measure temperatures (2mV/C) but it can also detect photons in light applications. Photon detection is the basis of all optical applications and measurements. The biggest and perhaps also the best known diode is found in solar cells. Solar cells generate electrical current when the diode absorbs photons from the sunlight.

Using diodes as electronic detection you need a silicon diode. In typical CMOS technology these diodes are as small as 5 x 5 µm and they change the leakage current when exposed to light. As the photons hit the surface of the diode, the leakage current and subsequently the voltage over the diode will change.

The sensing effect is created by the current change that follows the exposure to photons. The leakage current increases with the amount of photons that hits the diode.

When used for measurements, the diode detects photons over time (an integration process) and the sum of the number of photons is an expression on how much light you have had over time. You actually integrate the number of photons over time. This means that the leakage current in the diode keeps going up relative to the number of photons.

After a measurement, you need to reset the diode in order to prepare it for the next measurement. You reset the diode by adding a reference voltage to it and then it is ready to start a new measurement.

Design an artificial eye from a photodiode sensor

You can generate a photodiode from a CMOS silicon diode and an optical filter packaged in a transparent package. Adding the filter, you make the photodiode sensitive to specific wavelengths depending on the filter characteristics.

The optical filter will enable you to tune which wavelengths you want your diode to receive. For fluorescence measurements, the photodiode needs to be sensitive to wavelengths around 450nm. Laser diodes are sensitive around 850nm wavelengths. Thus with different light sources and wavelength sensitive diodes, you will build sensors for many different applications.

Designing an artificial eye takes thus three diodes and three different filters that let wavelengths for the colours red, blue and green pass. This way you determine the 3–axis colour coordinate (the relationship between the colours red, blue and green). Now you have created a pixel which is a one-pixel artificial eye.

Today, modern compact cameras have around 20M pixels in a picture. With 60 million diodes, they are able to make a very detailed image. A detailed image calls for a very powerful on-camera chip to capture the value of the 60 million diodes. And, clearly, there is a very high demand for storage of these images.

Detect motion with an artificial eye

You will meet two limitations if you wish to detect motion with a standard compact camera: The price and the high resolution.

A chip with a 20 million pixel resolution is expensive. Such camera chip can be up to 25x25mm and the silicon cost is above USD 30.

The high resolution will demand an exorbitantly large memory to store all data. For instance an IoT-application where you typically capture data from a large amount of sensor nodes – such as surveillance cameras – you can imagine the storage and data transmission requirements.

As it is very difficult to transfer such huge amount of data to a database, it is necessary to reduce the data size. Local intelligence and algorithms will reduce the amount of data transferred. The processor on the camera chip runs an algorithm to reduce captured data and transfer them simultaneously. This, however, will not reduce the cost of the camera chip. The chip cost is dependent on the chip size. Therefore you have to be willing to reduce the number of pixels.

Very few pixels are needed for image recognition

Take a look at the image below. At a close-up it is very difficult to see what it is. But from a distance, you will recognize a human head. If you calculate the number of pixels from this experiment, you will see that it is only 15×15 pixels i.e. a total of 225 pixels.

Fig.1: Low pixel image of a head

Fig.1: Low pixel image of a head

The 225 pixels do not give a very clear image but it can be implemented at a price less than USD 0.25. An optical sensor element at a price of USD 0.25 is quite competitive even compared to many mechanical sensors. Such solution is adequate for many image sensing applications.

Diode sensor elements and your product

The great advantage of this particular photo diode is that a CMOS chip is easily integrated and then you have a System On a Chip (SOC).

For some applications you will have to add filters directly on the diode to become wavelength sensitive. Other applications do not require such additional filters if you will measure certain optical signals in the dark.

The optical chip can be integrated with many kinds of other functions. These could be reset, memory, wireless connection or a microprocessor. You can get an integrated solution from around USD 1 which makes the sensor method very competitive with many mechanical solutions.

Over the last three years, DELTA has worked with optical sensors for a wide range of applications. Extract of applications is listed here:

Industrial & safety                                         Motion control

Light curtains                                                         Optical touch detection

Production control                                                Backlight control

Contrast control                                                     Rotation and velocity

Position detection                                                  Distance measurement (TOF)


Sensor system                                                 Eye wear

Intelligent windows                                              Ski goggles

Barcode readers                                                     Motorbike helmets

Optical spectroscopy                                             Welding helmets

Colour measurement                                            Night vision

Lately, we have seen companies use high speed diodes for communication such as LiFi and fiber-repeaters. DELTA believes that the use of chips which both combines light and electronics will continue to grow. Many mechanical sensors will be replaced by optical systems that will result in reduced bill of material as consequence.

The DELTA way

Do you want to introduce light applications or make a proof-of-concept? DELTA has implemented our photodiode-IP in a chip. The chip is mounted in a transparent package on a module. It has a Bluetooth interface that connects it to a mobile phone. You can control the photodiode via an app on an Android mobile phone.

The size of the diode depends on the application. A large diode captures many photons, a small captures a limited number of photons. A strong light source emits many photons, while a weak source will emit a low number of photons.

In the app, you enable or disable diodes to find the right diode-size for your application and specific light source. DELTA has access to optical filters and the expertise to determine right filter characteristics for a specific application. We offer standard filters to be used for trial. For the actual proof-of-concept we can provide the exact filters you need.

Fig. 3: The Android app allows you to turn gain on and off

Fig. 3: The Android app allows you to turn gain on and off

Optical chips are one of the focus areas at DELTA. As mentioned earlier we believe that many mechanical sensors will be replaced by optical measurement systems. The bill of material for the sensor element is much smaller and of higher reliability. Typically the sensor element is on-chip and there are less soldering connections.

How can we help you?

DELTA has experts and skilled teams both with light and microelectronics. At present we are involved in several development project regarding optical chips.

Therefore, if you have design ideas to your product, we shall be happy to discuss any implementation methods with you.

You can also visit our website. Here you find information, data and downloads on e.g. optical touch system, movement sensors, luminance detection systems etc.


This is a unique website which will require a more modern browser to work! Please upgrade today!