Voltage spikes due to switching disturbing transistor switch circuit

Question

^{simulate this circuit – Schematic created using CircuitLab}

I want to interface a 5v logic signal to an input of a microcontroller (STM32G030) that operates at 3.3V. I made the attached circuit to detect the rising edge of the input 5v signal. However, whenever a nearby AC mains socket switch is flipped, I get an interrupt (I configured the input pin of the microcontroller to generate an interrupt on detection of a falling edge). Can anyone help me improve this circuit and avoid getting such false interrupt detections due to the voltage spikes?

Answer 1

The simplest circuit to get the falling edge for your circuit is the one below:

^{simulate this circuit – Schematic created using CircuitLab}

It can be used to either detect a relatively slow-falling edge, or a low level input when the 5V supply is off.
The only drawback is a small current flowing from 3.3V to the 5V rail, but in most cases it should not affect anything.
R1 value of 100k should be good enough for almost anything, but you can experiment with resistor values from 10k to 1M, keep that line very short to reduce the effect from nearby voltage/current spikes.
I don't know how much is your +5V rail loaded, you can experiment and see how much voltage will there be when you disconnect the 5V supply line.

P.S.: I have re-read your question and realized you want a falling edge when the 5V comes UP.
Here is a schematic for that:

^{simulate this circuit}

C1 can be decreased to 100nF if it will be sufficient for detection by the MCU, or it can be increased if insufficient for detection.