Line Follower Robot Unable To Sense Lines

Thread Starter

Utkarsh Verma

Joined Oct 30, 2016
102
I made a Line Follower Robot as directed here. The only changes I have made are that instead of TCRT5000, I have used a pair of 3mm IR LED and Receiver. My PIC is also programmed well and all the modes work as stated on the "instructable". However, the main mode, that is the Line Following Mode doesn't seem to work. I checked all the wiring and I assure you, it's all clean and well done. Regarding the Line Following mode, when I power up my robot, it starts to rotate at one place due to forward motion of Left motor and backward motion of Right motor. I tried placing the Line Follower over a line as well, however it didn't yield any results. Please help me in debugging this project.
 

AlbertHall

Joined Jun 4, 2014
12,345
I made a Line Follower Robot as directed here. The only changes I have made are that instead of TCRT5000, I have used a pair of 3mm IR LED and Receiver. My PIC is also programmed well and all the modes work as stated on the "instructable". However, the main mode, that is the Line Following Mode doesn't seem to work. I checked all the wiring and I assure you, it's all clean and well done. Regarding the Line Following mode, when I power up my robot, it starts to rotate at one place due to forward motion of Left motor and backward motion of Right motor. I tried placing the Line Follower over a line as well, however it didn't yield any results. Please help me in debugging this project.
Do you have a shield between the LED and the phototransistor?
If you move it over the line do you see the voltage on the phototransistor changing? How much does it change?
 

be80be

Joined Jul 5, 2008
2,072
Wrap the photo transistor with a piece of tape black tape leave the bottom open
shielding.jpg
This is even better its heat shrink tube
homeShield.jpg
 
Last edited:

Thread Starter

Utkarsh Verma

Joined Oct 30, 2016
102
Do you have a shield between the LED and the phototransistor?
If you move it over the line do you see the voltage on the phototransistor changing? How much does it change?
Yes i have used a piece of cardboard in between the IR LED and the phototransistor. If you want to know the voltage the phototransistor is allowing to pass, would it be okay if i measure it between the cathode(- terminal) of the phototransistor and GND of the entire circuit?
 

be80be

Joined Jul 5, 2008
2,072
The thing you have to be careful with is that when using IR led and a phototransistor is that the phototransistor doesn't get any light from the side and the end that you wire up.
Oh and that a lot of so called phototransistor that you find cheap and some not so cheap are photodiodes and don't work the same as a real phototransistor.
http://educypedia.karadimov.info/library/Sharp photodevices.pdf
radio shack use to sale a set some would be photodiodes and some where phototransistor.
 
Last edited:

Thread Starter

Utkarsh Verma

Joined Oct 30, 2016
102
The thing you have to be careful with is that when using IR led and a phototransistor is that the phototransistor doesn't get any light from the side and the end that you wire up.
Oh and that a lot of so called phototransistor that you find cheap and some not so cheap are photodiodes and don't work the same as a real phototransistor.
http://educypedia.karadimov.info/library/Sharp photodevices.pdf
radio shack use to sale a set some would be photodiodes and some where phototransistor.
Thanks for your suggestion, I will keep this in mind while shielding the phototransistor.
 

Thread Starter

Utkarsh Verma

Joined Oct 30, 2016
102
shielding the phototransistors didn't change a thing! However, I would like to state that I checked whether the IR LEDs were lighting up using my mobile camera. The IR LEDs are lighting up. The conclusion to which I have come is that maybe the phototransistor isn't generating any pulses. I think so because when I used to power up my project before(when no IR pairs were soldered), the same thing happened. The robot just rotated at one place. Therefore I think that the IR receiver isn't generating any pulse to LM324 which in turn doesn't send any pulse to PIC16F84A. Please express your views on this. By the way, I happen to have a fully functioning multimeter as well as a continuity tester.
 
Can you change the pulse rate of the LEDs to verify the operation of each transmitting and recieving pair? Slow everything way down so you multimeter can easily measure the voltage changes in the circuit.
 

MaxHeadRoom

Joined Jul 18, 2013
28,619
I have worked on line finders in the past on trace cutting machines that use a drawing to follow in order for the cutting heads to produce a part as per drawing.
These work on the principle of the reflected line is oscillated over the photo cell in order to find the line and which side it wanders off.
I am curious as to how this system knows which side of the line it is on if it strays off and also how can it decide on T junction which way to travel?
Max.
 

AlbertHall

Joined Jun 4, 2014
12,345
From the robot site:
"This robot has 4 IR sensors that scan the path.
If the Right and Left sensors has same values and the values of them are different from the Middle or Front sensors the robot is on the line and motors running forward. Else the robot is out of line so the robot keeps moving till one of the side sensors value changes then it turns to the direction that its sensor value changed."

Also from that site:
"Remove the PIC and test your sensors one by one using a voltmeter. Black wire to GND, red wire to sensor pin on the PIC socket, hold sensor over a black area it must shows 0V and hold sensor on a white area it must shows 5V."

Have you tried this last method? What readings did you get?
 

Thread Starter

Utkarsh Verma

Joined Oct 30, 2016
102
From the robot site:
"This robot has 4 IR sensors that scan the path.
If the Right and Left sensors has same values and the values of them are different from the Middle or Front sensors the robot is on the line and motors running forward. Else the robot is out of line so the robot keeps moving till one of the side sensors value changes then it turns to the direction that its sensor value changed."

Also from that site
"Remove the PIC and test your sensors one by one using a voltmeter. Black wire to GND, red wire to sensor pin on the PIC socket, hold sensor over a black area it must shows 0V and hold sensor on a white area it must shows 5V."

Have you tried this last method? What readings did you get?
Yes, I tried it and found out that all my wirings are perfect except the "TCRT5000" part. I found out that the pins of IR receiver should be reversed. I gues the TCRT5000 might be manufactured differerntly. Anyways, now my robot is able to send pulses to PIC accurately. So this part is finished. However I have faced yet another issue which I will post in a detailed way tomorrow for I'm really exhausted currently.
Finally, thanks to all in easing the debugging process!
 
Last edited:
Top