Our Team Chemical Device
The chemical device our group was tasked to create is the self-sterilising door handle. What it does is that sterilises itself and eliminate most of the bacteria and germs left on a door handle after use, which minimises the spread of bacteria and hence infections from user to user. The finalised hand sketch is shown below:
By default, the UV bulb will be turned on to sterilise the surface of the door handle. When a person approaches the door handle, the proximity sensor (in this sketch, it shows a pressure sensor being used, however we switched it for a proximity sensor instead as it just works better overall for our product.
Team Planning, Allocation and Execution
Our team consists of:
Gideon: CEO
Gwyn: COO
Adam: CSO
Me: CFO
Finalised BOM (Bill of Materials) Table:
1. Initial Design Ideas (Sketching) Done by Me
We devised a new more practical design which involved an aluminium mirror to reflect the UV rays to both sides of the handle.
We did not follow through with the design above but combined the concept of all the designs we came up with. Which results in our actual design:
We decided to utilize a shorter transparent handle than the previous designs to cater to UV light bulbs instead of a tube. This way, our handle is much more compact, demanding less cost to produce.
At first, we had planned to buy an acrylic cyllinder so all I had to 3D print was a holder of sorts to attach it to the door which is this design I created with a circle and a square and extruded. I later on added a space at the bottom for the components. I used a lto of parametrics for this design as we did not have the materials yet and this was a rough idea.:
For our door handle, we wanted the UV light to be off, on default. When a person approached the handle, we wanted the handle to recognise the presence of a person, and turn on to sterilize the handle for the next use. Due to safety concerns with the exposure to UV light, it would wait for 1 minute for the person to have left the vicinity of the handle, and then turn on for 2-3 minutes since for a UV lamp held within 1 inch above a petri dish grown with E. coli, it will only take 1-2 min to show a complete sterilization. After which the UV light would turn off again. Due to safety, we could not use a UV lamp, so we substituted it with a LED bulb.
First, I had to figure out how the ultrasonic sensor worked.
Gwyn started the code by researching what the 4 pins in the proximity sensor were. Then the connections.
The trigpin is the output because it produces the sound waves while the echopin is the input since it receives the soundwaves. The a “pulseIn” command is used to repeated detect the soundwaves.
It all starts, when a pulse of at least 10 µS (10 microseconds) in duration is applied to the Trigger pin. In response to that the sensor transmits a sonic burst of eight pulses at 40 KHz. This 8-pulse pattern makes the “ultrasonic signature” from the device unique, allowing the receiver to differentiate the transmitted pattern from the ambient ultrasonic noise.
The eight ultrasonic pulses travel through the air away from the transmitter. Meanwhile the Echo pin goes HIGH to start forming the beginning of the echo-back signal. In case, If those pulses are not reflected back then the Echo signal will timeout after 38 mS (38 milliseconds) and return low. Thus a 38 mS pulse indicates no obstruction within the range of the sensor.
If those pulses are reflected back the Echo pin goes low as soon as the signal is received. This produces a pulse whose width varies between 150 µS to 25 mS, depending upon the time it took for the signal to be received.
To calculate the distance of the object detected, first divide the pulse by 2 since the time taken for the pulse to be received would be when it travelled from the sensor to the object and was reflected back, so it did twice the distance. After which to convert the distance to cm, divide the result by 29.1 since the speed of sound is 343 metres per second, or 29.1 microseconds per centimeter. That is how we get the distance.
This is the code:
#define trigPin 6 //Define the HC-SE04 triger on pin 6 on the arduino
#define echoPin 5 //Define the HC-SE04 echo on pin 5 on the arduino
#define bulb 13 //Define the relay signal on pin 9 on the arduino
void setup()
{
Serial.begin (9600); //Start the serial monitor
pinMode(trigPin, OUTPUT); //set the trigpin to output
pinMode(echoPin, INPUT); //set the echopin to input
pinMode (bulb, OUTPUT); //set the bulb on pin 13 to output
}
void loop()
{
int duration, distance; //Define two intregers duration and distance to be used to save data
digitalWrite(trigPin, HIGH); //write a digital high to the trigpin to send out the pulse
delayMicroseconds(500); //wait half a millisecond
digitalWrite(trigPin, LOW); //turn off the trigpin
duration = pulseIn(echoPin, HIGH); //measure the time using pulsein when the echo receives a signal set it to high
distance = (duration/2) / 29.1; //distance is the duration devided by 2 becasue the signal traveled from the trigpin then back to the echo pin, then divide by 29.1 to convert to centimeters
if (distance < 13) //if the distance is less than 13 CM
{
Light(); //execute the Light subroutine below
}
Serial.print(distance); //Dispaly the distance on the serial monitor
Serial.println(" CM"); //in centimeters
delay(500); //delay half a second
}
void Light() //Start the Light subroutine
{ digitalWrite(bulb, LOW); //turn off the light
delay (10000); //wait 15 seconds
digitalWrite(bulb, HIGH); //turn on the light
delay (20000); //wait 20 seconds
digitalWrite(bulb, LOW); //turn off the light
}
For the sensor distance, the reason why we chose 13cm was because through our research and testing, if we put a distance like for example 5cm, there was a chance that the sensor would not be able to detect the presence, and the LED bulb would not turn on. So 13cm was above the inaccuracy range, and through testing, it also was very accurate and did not fail to sense our presence using a hand in the stated distance of 13cm.
Here is a video during the testing phase, I had the LED be default on to ensure that the sensor was detecting my palm before switching off
First I created the design in Fusion 360
Then I had to separate the parts to print. To make the print faster and more stable, I elected to print the handle in 2 parts, top and bottom by splitting the design.
I then saved it as an stl file and exported it to cura and then saved it as a gcode file, with these settings:
We had 2 main problems:
We has trouble getting some of the materials such as getting an acrylic cylinder which resulted in having to change the design again. Initially, the plan was for me to create a holder of sorts to just attach the cylinder to the wall but then we had to adapt. I changed it to 2 possible designs where the handle was to be round or a square. We ended up going with the second design since we could not 3D print a transparent version of the cylindrical handle.
Another problem was the last of time when using the machines. A lot of teams want to use the same machines, 3D printers or laser cutters. I think that our group was lucky in that we were able to get slots for the 3D printer early. This gave us some lee-way in case something were to go wrong, we could still come back to print again. For the laser cutter, we did have to come back at a later date. Initially, our teammates booked in the Fab lab to bend the acrylic but when that did not work well, we switched to the rectangular model and booked another slot. However, the machine was down and so we had to return again. I think our team was able to overcome this challenge of limited slots by starting early and adapting if things go wrong.
Comments
Post a Comment