Two phototransistors. One influences the other?

Forum 2005-2010 (read only) Hardware Interfacing
1

Hello forum.

I have connected two phototransistors to my arduino as seen in this diagram:
http://img3.imageshack.us/img3/9756/deviceq.png

I have discovered the following which surprises me:

  1. Cover sensor A: Value for sensor A drops as expected.
  2. Cover sensor B with A still covered: Value for BOTH sensors drop. B a lot and A a "little".

Why are the readings from sensor A being influenced by sensor B? Could I rewire to avoid this? How?

Data sheet for phototransistor:

Thanks

2

This is because the pull down resistors are too high.
Two possible solutions:-

  1. Read each sensor twice in a row and ignore the first reading.
  2. Change the resistor to about 47K - but the sensitivity to light might not be so great.
3

Hello....

Thanks a lot! I ignored the 1st reading as you suggest. The readings in the covered sensor still drop but much less than before.

I'm having problems in low light situations. That is, I register no light in low light scenarios. Is there any way to make a setup with phototransistors which is more sensible to light? Can I somehow "move" the sensitive area? Maybe another sensor is better for low light scenarios?

As always, you help is much appreciated.

4

The readings in the covered sensor still drop but much less than before.

Then put a delay between the first and second readings. The interaction occurs because the input capacitor on the A/D has not got time to charge up to the correct value due to the high impedance of those resistors.

which is more sensible to light?

Assuming you mean sensitive, there are several solutions to this. The simplest concept is just to amplify the signal with and external operational amplifier. You can run into dynamic range issues with this.
A more involved method is to replace those 470K resistors with a capacitor, say 0.1uF. Then switch the pin to be a digital output and write a logic zero to it. Then when you want to make a measurement, make the pin an input, delay and then take an analogue reading. What happens is that the current trickling through the photo transistor charges the capacitor. The longer the delay the more time this current is gathered (or integrated) in the capacitor and so it is more sensitive. Push this too far though and you just end up with noise.

5

Thank you for your advice.

The simplest concept is just to amplify the signal with and external operational amplifier.

What would such a circuit (The amplifier) look like?

Thank you

6

It would look like lines on paper a bit like this:-

For more information see:-

7

I would insert this "just" before the analogue pin on the Arduino and 0 to ground? My supplier has a bunch of different ones. Would anyone do?

8

Looking through my salvaged circuit boards, I can see that I have some TI RC4558P op-amps. Would they be a good/reasonable choice? It looks to me as if I can use just one of these for my two phototransistors as it has two sets of ins and outs? Is that so?

9

it has two sets of ins and outs

Yes there are two amplifiers in the package.

Would they be a good/reasonable choice

well the data sheet says the minimum supply voltage is +/- 5V so you would need a source of -5v as well as the arduino's voltage.

There are amplifiers that can operate on only 5V nowadays.

My supplier has a bunch of different ones. Would anyone do

No sorry.
You need a 5v one. And you need to connect the -ve input of the amp (show as going to 0v) to a potential divider with two 1K resistors going between 0V and 5V. Also if possible one that will output the full voltage rail range, sometimes called a rail to rail amplifier.

10

Thank you. I think adding the amplifier is beyond my abilities at this time.

I really do appreciate all you help!