pull_up_down=GPIO.PUD_UP issue

[Robodanjal]

I have been working on a script which will run a photobooth that I have built for my wedding.

After testing it for 30 minutes or so today it started to trigger the photo preview sequence even though the button had not been pressed. To test if this was being caused by the button I simply removed it; it still triggered on it's own.

I removed everything from the Pi and now have it connected to the monitor, keyboard, and mouse. I have run the script below to see if I can rule out any mistake made in my photobooth script.

Code: Select all

import RPi.GPIO as GPIO
import time

GPIO.setmode(GPIO.BCM)

GPIO.setup(23, GPIO.IN, pull_up_down=GPIO.PUD_UP)

while True:
    input_state = GPIO.input(23)
    if input_state == False:
        print('Button Pressed')
        time.sleep(0.2)

With this code I am getting "Button Pressed" over and over again without the button being pressed.

Does anyone perhaps know what could be causing this?

[joan]

What have you got connected to GPIO 18?

[Robodanjal]

What have you got connected to GPIO 18?

My momentary switch.

[joan]

A photo may help.

The internal pull-up is about 50k ohms. That's normally enough to prevent random changes at the GPIO. However an attached wire will be affected by electrical noise and radio waves. The longer the wire the more susceptible. You could try something like an external 5k pull-up.

[Robodanjal]

A photo may help.

The internal pull-up is about 50k ohms. That's normally enough to prevent random changes at the GPIO. However an attached wire will be affected by electrical noise and radio waves. The longer the wire the more susceptible. You could try something like an external 5k pull-up.

It' is quite simply GPIO 18 to switch and switch to ground.

My issue is that this has just started to occur, it was working correctly earlier.

Oddly enough if I do not connect anything to the GPIO pin I still get the output "button pressed".

Quick video to show what is going on. At no point is a button pressed.

https://www.youtube.com/watch?v=7rWMD4l78-0

[joan]

The video appears to show the switch is connected to 5V. If that's the case you may have destroyed the GPIO.

[Robodanjal]

The video appears to show the switch is connected to 5V. If that's the case you may have destroyed the GPIO.

You're mistaken, it is connected to ground.

[DougieLawson]

The code in the video is set-up for GPIO 23 pin#16, the code you've posted in this thread is GPIO 18 pin#12.

[Robodanjal]

The code in the video is set-up for GPIO 23 pin#16, the code you've posted in this thread is GPIO 18 pin#12.

Granted, but I am using GPIO 23 on the Pi and have changed the code to match that. This doesn't explain the issue I am having.

I shouldn't be getting the "button pressed" output when nothing is being pressed. The issue I am raising is reproduce-able regardless of which GPIO I am using, or if there is anything connected to the GPIO at all.

[DougieLawson]

If the pin is floating you can get anything (it will pick up all sorts of stray RF noise). Add an external pull-up or pull-down 10K resistor (depending whether you're detecting a low or high in your code) so that you know what value is expected when the switch is open and when it's closed.

[Robodanjal]

If the pin is floating you can get anything (it will pick up all sorts of stray RF noise). Add an external pull-up or pull-down 10K resistor (depending whether you're detecting a low or high in your code) so that you know what value is expected when the switch is open and when it's closed.

I still don't understand how this was a non-issue for the last 8 months and all of a sudden it is having this issue.

Also, should it be picking up interference when no cabling is attached to the GPIO at all?

[DougieLawson]

Because you've got some long unshielded wires attached to your GPIO. Those things are acting as an antenna.

What's the difficulty with adding a 10K resistor?

[Robodanjal]

Because you've got some long unshielded wires attached to your GPIO. Those things are acting as an antenna.

What's the difficulty with adding a 10K resistor?

No difficulty, I'm just not convinced that this is the cause of the issue I'm having. Also, should I still be having this issue when I have no cables connected at all? As in I run the script with nothing attached to the GPIO and I still get the button pressed output.

[DougieLawson]

Just a bare pin acts as an antenna if there's enough RF noise nearby.

[boyoh]

Some times under standing why you use pull up and pull down resisters can be hard to understand.
The GPIO in it's normal state as a very high impedance( resistance) some times called TriStat
It will neither sink or source a signal( Floating) To give it a working impedance level you use
Resisters to set the level, I think 10k is a good level,

If your in/put signal is a active 1( high) you connect the 10k to the 0v- Gnd Rail.

If your in/put signal is a Active 0 ( low ) you connect the 10k to the 3.3v+rail

[Robodanjal]

Some times under standing why you use pull up and pull down resisters can be hard to understand.
The GPIO in it's normal state as a very high impedance( resistance) some times called TriStat
It will neither sink or source a signal( Floating) To give it a working impedance level you use
Resisters to set the level, I think 10k is a good level,

If your in/put signal is a active 1( high) you connect the 10k to the 0v- Gnd Rail.

If your in/put signal is a Active 0 ( low ) you connect the 10k to the 3.3v+rail

Thanks for your reply Boyoh, I appreciate you taking the time to reply.

Unfortunately I have tested my Pi against a borrowed one and it seems that the GPIO on mine is faulty. Same code, same connected hardware, and the borrowed one works correctly.

[DirkS]

Unfortunately I have tested my Pi against a borrowed one and it seems that the GPIO on mine is faulty. Same code, same connected hardware, and the borrowed one works correctly.

Did you also swap the SD card?