help!! sensors not working........

[forever99482]

Hi all,

I'm currently making a line robot that needs to follow a maroon track with white background. The sensors used is OPB704, optimum distance from ground is about 0.5cm.

But the problem is, I can only get about 0.5V of voltage difference from the collector when it is changed from the white background to the maroon track.
Furthermore, the track is expected not to be flat, meaning the distance between the sensor and the ground would be changing.

Any idea of solving this problem?

Thanks in advance

 

[Hero999]

The maroon track is probably infrared reflective. Have you tested the senser with an object that isn't reflective to infrared radiation?

 

[forever99482]

i test it on a black track, and it gave me about 0.5V, on a white track, its about 3.0V.

on a maroon track, its about 2.5V.
but the racing track is expected not to be flat, so the distance between the sensor would be changing, and 0.5V of margin is not enough to make sure the sensor "see" the right thing....

 

[Hero999]

I'm not sure if you understand, colour doesn't mean anything at wavelengths other than the visible spectrum, your maroon track will appear white to an infrared camera because it reflects infrared light. There are two possible solutions to this problem; you can either use a track that doesn't reflect infrared light or use a sensor that uses a different wavelength light so won't suffer from this problem.

 

[ikalogic]

Hi all,

I'm currently making a line robot that needs to follow a maroon track with white background. The sensors used is OPB704, optimum distance from ground is about 0.5cm.

But the problem is, I can only get about 0.5V of voltage difference from the collector when it is changed from the white background to the maroon track.
Furthermore, the track is expected not to be flat, meaning the distance between the sensor and the ground would be changing.

Any idea of solving this problem?

Thanks in advance

i am sure i can quite help, i am actually building very similar stuff for a robot competition. plz send some schematics

also, a 0.5v can be very easily detected using an op-amp....

using IR can solve the problem of range..

 

[forever99482]

I'm not sure if you understand, colour doesn't mean anything at wavelengths other than the visible spectrum, your maroon track will appear white to an infrared camera because it reflects infrared light. There are two possible solutions to this problem; you can either use a track that doesn't reflect infrared light or use a sensor that uses a different wavelength light so won't suffer from this problem.

dont think changing track is possible, and the demo for the race is on this coming friday, so changing sensors would probably not be an option either........ T_T the robot worked well on black track though, cuz in the rules it stated black track.......

For the circuit schemetics, dont think its anything special, just connecting the sensors(opb705) to the comparators(LM339N) then to the PIC16F648A, 7 inputs(sensor), 8 outputs(for motor driver).

from 8 outputs, separate into 4 each and feed it to the 2 motor driver(uln2803A) (to prevent excessive current in the motor driver). Then from motor driver just connect it to the stepper motor.

 

[ahmedragia21]

ikalogic,
use an ultrabright red led and it will solve ur proble, also make sure the Rc of the phototransistor is a variable one .
and inshallah it will work

 

[Pommie]

If it's reflects IR then it will probably reflect red. Try a green LED.

Mike.

 

[ikalogic]

ikalogic,
use an ultrabright red led and it will solve ur proble, also make sure the Rc of the phototransistor is a variable one .
and inshallah it will work

IR is known for its high reflectivity. .. read that in my physics books..

 

[Hero999]

The maroon track probably only reflects infrared and red light at the far red end of the visible spectrum. Using a red emmitter might work but green would be better. The only problem is the reciever might have an infrared filter on it do it only receives infrared. One solution could be to replace the receiver with a cadmium sulphide cell (providing it's fast enough) and using a green LED.

Having said that (as correctly stated above) 0.5V is a large enough change in voltage to be able to detect using a comparator so I don't see the problem.