0
\$\begingroup\$

In this schematic, we see a 10μF cap (C1). What is its purpose, wired like that?

enter image description here Source: https://www.electroschematics.com/9023/infrared-beam-break-detector/

I duplicated this schematic with two TSOP1236 devices, connected to a LM393.

The TSOP1236 datasheet tells me that, in its typical application, I need a cap between pin 1 GND and pin 2 VS. However, here the cap is inserted between pin 1 GND and pin 3 OUT.

enter image description here

So I thought it must be for protecting the LM311. And since I use a LM393, I now have one cap on each side. However, I would think one cap would be sufficient as connected between LM393 pin 4 GND and pin 8 VCC.

enter image description here

Can someone help me understand?

Would it be better to remove the two caps, one on each side of my LM393, and

  • put one 10μF between LM393 pins 4 and 8, and
  • put a 4.7μF between TSOP1236 pins 1 and 2?

EDIT: The TSOP1236 tries to detect a 36kHz signal and is connected to an Arduino, via the LM393 voltage comparator. The attachInterrupt() is triggering unexpectedly and I think using caps correctly across my receiver set-up could solve that.

\$\endgroup\$
  • 3
    \$\begingroup\$ SInce the cap is on input to the comparator, the purpose is to remove high frequency variations What is the circuit trying to detect? \$\endgroup\$ – analogsystemsrf Jun 13 at 13:17
  • 1
    \$\begingroup\$ The circuit will work both with and without that 10 uF capacitor, it will just respond slower with the capacitor. It might also be able to detect short IR pulses better with the capacitor in place. I suggest that you include the capacitor and try what works best for you. On a PCB you can always include the place for a component and later decide that you leave that spot empty. \$\endgroup\$ – Bimpelrekkie Jun 13 at 13:21
  • \$\begingroup\$ The output of the TSOP1236 is an NPN transistor (open collector) and a 30 kΩ resistor to V_s, so the output goes to low-level quite fast but to high-level very slow with the capacitor; to discharge the capacitor a very high current flows through the output transistor, which will be greater than the maximum output current of 5 mA; hence if you want to slow everything down, place a resistor (say 6 V ∕ 5 mA = 1.2 kΩ) at the output to limit that current... \$\endgroup\$ – aschipfl Jun 13 at 13:29
  • \$\begingroup\$ In your second circuit, capacitor C1 is connected between Vs and GND on the TSOPxxx - no connection to Out. \$\endgroup\$ – Peter Bennett Jun 13 at 15:53
  • 1
    \$\begingroup\$ @Nils Make sure that when you copy photos, you provide attribution (a link to the site where you got it from) to comply with copyright law \$\endgroup\$ – Voltage Spike Jun 13 at 16:08
0
\$\begingroup\$

The circuit from electroschematics is for a break detection circuit, and the 10uF cap is probably for a delay. The buzzer would sound when the infrared signal is lost (and a short delay from the cap). In the original schematic the detector is for a specific frequency (like 36kHz). When the 36kHz signal is detected, it brings its output high or low. The cap, with the RL resistor will form a low pass filter with a time constant of ~1 second.

The internals of the sensors (GP1UX51QS or TSOP1236 ) actually look like this:

enter image description here Source: https://media.digikey.com/pdf/Data%20Sheets/Sharp%20PDFs/GP1UX51QS.pdf

If your getting bad triggers, make sure your power supply is clean, you could still get bad triggers if the voltage came up and down quickly. Verify the output of the LM393 with an oscilloscope or logic analyzer (they are cheap, if you don't have one get one) It may be beneficial to add some hysteresis to the comparator (which will require some math)

\$\endgroup\$

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.