DT_INTS-18 - problem with INT0

[HenrikOlsson]

Hi,
What do you trig INT0 with? If it's a pushbutton you may have contact bounce that makes the interrupt fire multiple times. The 10mS pause in the ISR provides some debounce but perhaps it's not enough.

Also, I don't see

@   INT_ENABLE   INT0_INT     ; enable external (INT) interrupts

anywhere. I see that you "manually" enable the INT0 interrupt but try using the provided macro, to begin with at least.

/Henrik.

[Charles Linquis]

To solve debounce problems, you can test for PORTB = 0 (or 1 if you are triggering on a rising edge). Don't leave the ISR unless the button is in the un-pressed state.

[AndrewC]

I am triggering off a pushbutton but it is not debounce - it triggers once when I press the button, and again when I release.

I thought of adding an "exit delay" to miss the release, but that fixes the symptoms and I'd like to understand the cause

Macro for enabling the interrupt: hmmm - I'll try that tomorrow but is there a macro anywhere that defines which edge to interrupt off or should I stick with INTCON2.6 line ? I'm also going to try moving that statement after the interrupt create ASM section.

Thanks, Andrew

[Gusse]

Remove this one. And reason can be found from here .

INTCON.4 = 1 'enable INT0

BR,
-Gusse-

[AndrewC]

Okeydokey - so I'll take out the INTCON lines but then where do I tell the D_INTS whether to trigger on a rising or falling edge ?

Andrew

ps: I'll add this question to the main D_INTS thread as it might be of interest to others.

[Acetronics2]

Hi, Andrew

you sure ?

INTCON2.6 = 0 ' Int 0 on falling edge of C3

a look to INTCON2 register could help ...

Darrel's goodie only enables/Disables the interrupts ... it's you to set interrupt conditions.

Bouncing looks here a more than a valid reason for "false triggering" ...

Alain

[Gusse]

Hi Andrew,

This is OK, interrupt on falling edge.

INTCON2.6 = 0

Also add that INT_ENABLE part as Henrik suggested. Then it should work.

Q: Isn't it so that INT_Handler execution cannot be interrupted (e.g interrupt in interrupt sequence)?
If yes, then debounce filtering is not needed, because first falling edge trigger INT_Handler. Ringing is ignored during interrupt execution time.

BR,
-Gusse-

[AndrewC]

Hi, Andrew

you sure ?




a look to INTCON2 register could help ...

Darrel's goodie only enables/Disables the interrupts ... it's you to set interrupt conditions.

Bouncing looks here a more than a valid reason for "false triggering" ...

Alain

At first I thought it couldn't be a bounce, but then I was just thinking of a "bounce" as I pressed the button. Obviously, a push button can bounce as you release it as well. Simplest thing is probably to just build a delay into the interrupt routine - timing isn't critical in this application and there is no problem holding in the interrupt routine for a few hundred msecs.

Andrew