Reverse engineering an automatic air freshener spray machine

There are lots of commercial electronic products available in the market with simple to intermediate to complex and more complex circuit designs. Now to clarify how they works we have to study the circuit and try to apply our existing knowledge to figure it out, this is generally known as reverse engineering also known as hardware hacking. Practically reverse engineering is a way to progress further and to explore different short techniques which can be very fruitful in our future designs.

In this hack, I tried reverse engineering an automatic air freshener spray machine picked up from the junk. The machine works on two AA batteries i.e on 3 volts. It was not working at first, I unscrewed it and found the the battery connector leads got rusty, cleaned it up but still no response.

It got a motor gear attachment to increase the torque to press the nozzle of the air freshener can, there are two LEDs and two buttons one is for ON and OFF another is to set the timer which are 9 mins, 18 mins and 36 mins. I took out the circuit and cleaned it with thinner then found that the main problem is in the power input of the PCB, two pads got uprooted causing a discontinuous path to the main circuit so I fixed it and then it respond.

So, the main function of this instruments is to spray the air freshener in the selected time intervals. The green LED blinks in each second 540 times in case of 9 mins interval then the motor is ON for 2 seconds providing enough torque to push the nozzle down and stops then the backward force applied by the nozzle pushes the gear back to the initial position, when the motor is on the red LED turns on. So using my multimeter and oscilloscope I went for a deeper understanding of the behavior of the IC as the manufacturer scratched out the name and part number of the IC. After observing the circuit for half an hour, I made a schematic and as per my assumption it is a 8 pin microcontroller running on 3 volts.

Most of my assumptions matched when I probed the circuit to my oscilloscope. The pin configurations are –

  • Pin 1. Vcc i.e it is connected to +3 volts.
  • Pin 2. This pin is unused in this circuit.
  • Pin 3. Its connected to the base of a NPN transistor having a 100 Ω current limiting resistor in series.
  • Pin 4. It is connected to the delay selection switch and pulled down to ground with a 100 KΩ resistor which provides the delay of 36 mins.
  • Pin 5. It is connected to the 18 mins delay.
  • Pin 6. The green LED is connected in this pin.
  • Pin 7. It is connected to the 9 mins delay.
  • Pin 8. It is the ground pin of the microcontroller.

Observations made :

  1. I probed Pin.3 of the circuit and switched it on and got a pulse which is 2 seconds high that means at starting the motor turns on for 2 seconds.
  2. The delay selector switch pulls the corresponding pins high.
  3.  Pin.6 is programmed like a decade counter the connected green LED blinks in the interval of 1 second in that interval the pin goes low(ground), the current passes through the green LED to ground turning the LED on for one second. And after 1 second it shifts one bit like a decade counter. It covers 12 shifts 9 times for 1 minute. According to my assumption, a counter counts the 12 shifts x 9 and increases its count value correspondingly. Another counter is set to count the minutes, whenever the second counter hits 9 mins. Pin.3 again sends a 2 second high pulse to the motor.

  • The NPN transistor is working as a switch and it delivers the sufficient current needed to drive the motor. A diode is connected in reverse bias in parallel with the motor to protect the transistor from the current peaks formed due to electro magnetic energy stored in the motor coil.
  • The 10uF electrolytic capacitor is connected in parallel with the ON/OFF button to filter the noise due to button debouncing and 0.01uF ceramic disc capacitors filters the noise of the power supply.
Written by : Subhadeep Biswas

Precautions to maintain before using a PC oscilloscope

In every stage of designing a circuit you need an oscilloscope to see whether you are getting the desired output signal or not, and if you are reverse engineering a circuit then oscilloscope is very much essential tool for that. A wide variety of oscilloscopes are available in the market and they are very expensive so most of the hobbyist choose to buy a PC oscilloscope. After some research I bought a cheap one for my basic experiments it’s a Hantek 6022BE, it has 20MHz of bandwidth and 48MSa/s of real time sampling rate. But before using it we have to cope with some problems that can blow this device in seconds.

Here is a awesome video by Dave from EEVBlog on this purpose.

Many of us use a computer SMPS as power supply personally I also use, but when you are probing a circuit with your PC oscilloscope which is powered by the SMPS you need to do a small and simple hardware hack in the power supply. At first there are two sections present in the SMPS, high voltage(HV) side and low voltage(LV) side. HV side and LV side are isolated either by isolation transformer or by opto-couplers.

But in most cases the ground of the LV side is connected with the earthing of the HV side and this is quite dangerous so our hack will be to isolate ground form earthing in HV side.

Now, we have to test the continuity to be sure that ground and earthing of LV side and HV side are connected. Probe the multimeter in the continuity mode if there is no continuity then the HV and LV side is isolated otherwise we have to connect the earth wire of the LV side to the HV side or the circuit

The metallic casing of the SMPS is earthed, so the screw mounts of the LV side needs to be isolated form the LV side circuitry by isolation tape I’ve used thick double sided tape.

From these pictures we can see that the earthing is attached only with the HV side screw terminal which is isolating the the earth and ground and after the hack in working condition a 5V output is tested.

I added a buck converter circuit to +12V output so that I can use it as a variable voltage output between 0-12 volts also drilled the casing and attached banana connectors to make the supply more portable.

Hacked! power supply from old SMPS
Written by : Subhadeep Biswas