In this chapter I am going to write about dust sensor. If you follow my project then you will know I plan to use Sharp GP2Y1010AU0F dust sensor.
Again I will provide the information here in case someone is not read my previous chapter.
Link to Element 14
http://sg.element14.com/sharp/gp21010au0f/dust-sensor/dp/9707956?COM=IntheAirDesignChallenge
Link for datasheet
http://www.farnell.com/datasheets/73805.pdf
https://www.sparkfun.com/datasheets/Sensors/gp2y1010au_e.pdf
Related connector
http://sg.element14.com/jst-japan-solderless-terminals/zhr-6/connector-housing-6way/dp/3357570
This is a simple dust sensor, it shine an LED light and measure light power or density to determine the dust or particle in the air.
Given picture below show the internal design of the sensor.
T
To anyone who want to use this sensor please noted that this sensor is not accurate calibrated and user need to calibrated them to get correct and good data (base on my experience).
They never mention to use different supply for LED and the amplifier circuit. In my case I using same supply 5V for LED and amplifier circuit.
For this testing I using workbench power supply unit but in the real application I going to use DC-DC converter or Power Bank to run the circuit.
I hope those dc-dc not affected the stability of the output data. My selection power bank is Xiaomi power bank due to price and battery given (Mi Power Bank - Mi Singapore).
Will DIY the power management part if possible but for now just keep with power bank.
Warning: Please remember to install the 150Ω resistor before power up the LED to prevent damage to the LED.
LED is driving using pulse wave, LED is activated when the input is low. Mean LED ON when the drive pulse at 0v or lower then 2.5V.
You need a digital output or PWM output for this LED drive pulse.
For data it look straight forward. Output of amplifier is analog signal form about 0.9v to 3.5V.
The important is the data taking need to be taken 0.28ms after the on stage for LED to get the accurate data.
Dust sensor detail:
This is front view of dust sensor(I think). They have 6 pin 1.5mm pitch header and some test point with opening hole from the metal housing.
The variable resistor to set the sensitivity of the sensor also can be access from this side.
Higher resolution picture:
https://www.dropbox.com/s/d4begl1qdhi6z5x/IMG_3672.JPG?dl=0
This is back view of dust sensor(also I think). The hole in the middle is to allow air pass and where the air dust will be measure.
Here you also can identifier the sensor brand and model. They are indicated on the housing (this angle of picture normally see for Element 14 order) .
Higher resolution picture:
https://www.dropbox.com/s/82l3oh7225bcc0s/IMG_3673.JPG?dl=0
This view is after metal cover been remove.
Here you can see the transistor, amplifier and PCB where all the component been put together.
The metal cover hold by two screw that screw to the plastic housing.
Higher resolution picture:
https://www.dropbox.com/s/a8wual3fe6yeppi/IMG_3677.JPG?dl=0
The last view is LED, photo detector and lenses in black plastic housing. They been arrange nicely to detect the light reflection
Look like this sensor is not design for servicing. Everything is keep hidden inside and lenses also unable to access from outside.
Higher resolution picture:
https://www.dropbox.com/s/z09ity0ommgrrhn/IMG_3682.jpg?dl=0
Dust sensor testing:
Higher resolution picture:
https://www.dropbox.com/s/yi6ktuqaeh24r7e/IMG_3675.JPG?dl=0
Higher resolution picture:
https://www.dropbox.com/s/wemnngbtam4045i/IMG_3685.JPG?dl=0
For testing I'm using function generator to drive LED and oscilloscope for output reading and measurement.
Look like a particle counter is needed to calibrate this sensor. The output is just difficult to see from oscilloscope and it come in random unless you put something in the hole 
Vibration will affected the output of this sensor, mean that this sensor is not suitable to use in a vibrating or very hash environment.
The output signal will keep jumping although the LED pulse is not given.
Lighting also will affected the output signal, this mean when there are strong light shine to sensor it will provide some reading.
This will affected data and provide an error data when in the event of data reading.
Some advice:
Try to make the connection as short as possible to get nice and more accurate data. This also will help to reduce the noise and prevent delay that will affect data taking.
Solder directly to the PCB on this sensor will be better way of connection. Having some issue with the connection using DIY clamp header.
Sharing:
Some example form other work on same sensor.
http://www.howmuchsnow.com/arduino/airquality/
http://sensorapp.net/sharp-dust-sensor-and-arduino/
Please check the code I not sure it working
Arduino code for this sensor http://www.howmuchsnow.com/arduino/airquality/dust.ino
Another similar sensor found just for sharing.
Buy Grove - Dust Sensor [SEN12291P] | Seeedstudio
Previous
Dust, Temperature and Humidity Monitor Chapter 2
Next
