IR barrier debugging

HI!
Need help about debugging!

I have IR barrier setup of Arduino Pro mini, Vishay TSSP4038 sensor and Vishay TSAL6100 infrared diode. TSSP4038 is connected to IRQ pin 2 (also use resistor and capacitor as in datasheet) and diode is driven on 38 KHz @ 40mW 50% duty cycle. Distance from sensor to diode is 3m. This setup is used for precise timing system based on DS3231 clock to detect object crossing IR line. Setup has stable power from AMS1117 3.3V
Everything is working fine, but sometimes sensor false detect IR barrier brake. False detect interval is from couple hours till couple days. As it is used for timing system I definitely don’t need false brakes.
Possible troublemakers:
-diode should be driven at higher power level (I could use setup of 2N2222 transistor, diode can keep 100mW) to get more radiating power (this is outdoor setup – sun/rain)
-arduino generating 38 KHz for diode in long term could have some oscillating error, which cause sensor to detect brake (maybe I should put diode in simple 555 timer)

Any thoughts how to debug this situation? Anyone knows about industrial IR barriers precision - do they have false brakes?

Is the wiring from the receiver short and shielded, so it doesn't pick up RF from cellphones etc..
Sunlight, indoors? Post a picture of the setup.
Leo..

The following quote is from the first page of the TSSP4038 datasheet :

FEATURES
• Up to 2 m for presence and proximity sensing

Maybe you are expecting too much from it, if you are using it at 3m.

JohnLincoln:
The following quote is from the first page of the TSSP4038 datasheet :Maybe you are expecting too much from it, if you are using it at 3m.

from the same spec - Transmission distance
Ev = 0, test signal see fig. 1,
IR diode TSAL6200,
IF = 200 mA
d - 25 - m

Wawa:
Is the wiring from the receiver short and shielded, so it doesn't pick up RF from cellphones etc..
Sunlight, indoors? Post a picture of the setup.
Leo..

Wires for sensor and IR diode are long, but I use high quality video signal shielded cables so it can't be culprit (hope so).
Sunlight shouldn't be the issue - 38KMz modulated signal of IR and sensor is in IR resistand plastic tube.
It happens indoors too.
Sorry, but pictures will be no good - setup is in plastic box and all wiring is hidden.

Basic faultfinding can be done by dividing the setup in half.
Test the wiring and receiving software e.g. by shorting sensor output to sensor ground right on the sensor pins.
(not a problem, since the sensor has an open collector output)
That will mimic the sensor receiving a continuous signal.
Test the setup for a couple of days, or until the fault shows.
If the same thing happens, you can rule out transmitter and sensor.
Leo..

Hi,

Can you please post a copy of your circuit, in CAD or a picture of a hand drawn circuit in jpg, png?

Thanks.. Tom... :slight_smile: