1
\$\begingroup\$

I wanted to implement a one way wireless optical communication using arduino and LEDs but soon I had to choose sensors. LDRs are very slow and IR does not respond to visible light. My aim is to achieve 3-4 m of transmission so a sensitive sensor will be better for implementation.

\$\endgroup\$
8
  • 4
    \$\begingroup\$ Photodiode + transimpedance amplifier is the usual solution. \$\endgroup\$
    – pjc50
    Nov 30, 2016 at 17:03
  • 2
    \$\begingroup\$ 'slap on forehead' Yes \$\endgroup\$
    – Voltage Spike
    Nov 30, 2016 at 17:12
  • 1
    \$\begingroup\$ Okay my bad. The problem is local vendors in my area don't have them. Thanks for the help. Any suggestions for good sites to buy them. \$\endgroup\$ Nov 30, 2016 at 17:14
  • 2
    \$\begingroup\$ Two commonly used vendors for small orders (at least in the US) are Mouser and Digikey. You should be able to get photodiodes for visible light from either. You'll want to try to order other things at the same time--if you order a single photodiode, you'll probably pay considerably more for shipping than for the diode itself. \$\endgroup\$ Nov 30, 2016 at 17:35
  • 1
    \$\begingroup\$ Okay in my case 100 kHz is what I consider high speed. I want it to work in ambient light with the added 50 Hz (in India) noise. No typical frequency specifications until it is in the visible part of spectrum. As far as AWG noise is considered I will suppress it by increasing the power of transmitted signal. \$\endgroup\$ Nov 30, 2016 at 17:55

2 Answers 2

3
\$\begingroup\$

Any LED also works as photodiode, for exactly the wavelengths it emits.

In your 50 – 100 kHz case, you can simply use any LED. They should be fast enough. This is totally low-tech! Reverse-bias your LED, add a AC-coupling capacitor, and filter appropriately to extract the signals you're interested in.

India is fairly well-equipped with distributors. Most international distributors also deliver there.

\$\endgroup\$
2
\$\begingroup\$

Your design may benefit from Sunlight/lightbulb nulling.

schematic

simulate this circuit – Schematic created using CircuitLab

Get that DC/50Hz photon energy out of the signal. Wire an integrator to VDD/2 bias, attach Rin to output of photodiode, pick feedback cap for 10Hz response so your 100KHz data does not get attenuated but DC is fully nulled. Connect OpAmp output to a Darlington in series with the photodiode; if the photodiode produces lots of DC, the Darlington will conduct the DC current to the opposite rail. You'll need a resistor to VDD/2 to convert the desired 100KHz current into a signal voltage. And place back-back diodes across the resistor.

\$\endgroup\$
1
  • \$\begingroup\$ A circuit diagram would be much easier to follow. \$\endgroup\$ Feb 24, 2017 at 18:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.