I'm trying to make an IR beam break detector, TSOP1738 is pretty neat, but it has gain attenuation and it makes things complicated for me.

I decided to use a simple IR phototransistor, I amplified the output by two Darlington pairs but the range is too short! It only works if the IR LED is within a few centimeters shining straight through. But TSOP1738 works very fine from meters away. Is it normal? What is inside receivers like TSOP1738? Is it a simple IR phototransitor with special amplification circuit?

  • \$\begingroup\$ The data sheet has a block diagram \$\endgroup\$ – Icy Jan 28 '16 at 16:12
  • 1
    \$\begingroup\$ micropik.com/PDF/tsop17xx.pdf this explains it. \$\endgroup\$ – user1582568 Jan 28 '16 at 16:17
  • \$\begingroup\$ More generally, this link provides lots of useful information google.com \$\endgroup\$ – user1582568 Jan 28 '16 at 16:20
  • \$\begingroup\$ Two Darlington pairs should provide more than enough gain, if used in a sensible configuration. The important property of the TSOP is its filter. It provides gain around 38 kHz, but not at lower frequencies. There is DC background and a lot of noise at 50/60Hz and harmonics up to 1kHz which will overdrive the amplifier if gain for those components is as high as gain at 38 kHz. \$\endgroup\$ – Michael Karcher Jan 28 '16 at 16:35
  • \$\begingroup\$ Not an answer (exactly) to your question, but you can use lenses both to focus the irLED output towards your receiver in a narrower beam & to focus more of the incoming light on your phototransistor. This approach should significantly increase your range, though using a pre-built/focused ir laser would give best range with less work/complexity than getting appropriate lenses sourced, installed and focused. \$\endgroup\$ – Robherc KV5ROB Jan 28 '16 at 17:47

The reason it works so much better is because of its ability to ignore background light that is not modulated at 38kHz.

If a plain photodiode is to reliably detect the output of an unmodulated IR LED, then the brightness of the IR LED must be significantly above the background illumination level, e.g. from sunlight etc. That means that they are close together, the transmitting LED is incredibly bright and/or the background light level is basically dark.

If the IR LED is modulated (38kHz being typical but there's nothing much special about that frequency) then the detector can implement a narrow bandpass filter to exclude all the interfering infrared sources which are mostly around DC and 50-60Hz. That way, as long as the detector is not completely saturated (i.e. in direct sunlight), it can detect the desired modulated source with excellent sensitivity no matter what the background light level is.

  • \$\begingroup\$ Thank you for your answer, I think I'll have to abandon working on a bare IR phototransistor. To solve the gain attenuation problem I decided to PWM the power input of TSOP1738 at a frequency of 1kHz and duty cycle of 90%, do you think it is a good idea? \$\endgroup\$ – ahmadx87 Jan 29 '16 at 8:58
  • \$\begingroup\$ The TSOP1738 has only an optical input, and it expects you to modulate the IR transmitter (LED) at 38kHz not 1kHz. \$\endgroup\$ – William Brodie-Tyrrell Feb 1 '16 at 3:19
  • \$\begingroup\$ I think you didn't get me quite right, I modulate the IR transmitter at 38kHz but on the receiver side to avoid entering gain attenuation mode I modulate the power of TSOP1738 at say 1kHz. \$\endgroup\$ – ahmadx87 Feb 4 '16 at 6:28
  • \$\begingroup\$ Don't modulate the power input; it will malfunction. Just use the device as designed (including power decoupling capacitor) and it will work properly. \$\endgroup\$ – William Brodie-Tyrrell Feb 6 '16 at 21:56

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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