[Solved] Arduino Timers – no pain and all gain!

@will I know conceptually what a NAND is, but I am talking about push buttons and relay contacts, the circuit proposed is totally different.

@zander 

They can also be used in different circumstances as well. The circuit shown using them isn't a good candidate for what you want.

How long a delay between the actual button push and the actual trigger event can your project stand ? That is, how fast does the debounce have to happen ?

There's an interesting article about debouncing at digikey which illustrates the use of NAND gates for this purpose (amongst other alternatives).

https://www.digikey.ca/en/articles/how-to-implement-hardware-debounce-for-switches-and-relays

@will The problem with that NAND circuit is it is a SPDT, not SPST. I just realized the kind of push button I am using is part of the problem, it's spring loaded or something. I will either find a switch that doesn't bounce, or look into capacitive touch sensors. If I just touch a wire to the pin there is no bouncing, there must be a switch somewhere that does the same thing.

Sorry for taking up so much bandwidth for a dumb switch.

Posted by: @zander

Sorry for taking up so much bandwidth for a dumb switch.

Well, I personally am finding it to be interesting reading the replies.

@zander, I found several articles on debouncing switches, and just thought I'd share them with you;;

The first one is a link to actual switches that claim to be bounce free...;

https://www.digikey.com/en/products/filter/pushbutton-switches/199?s=N4IgjCBcoLQdIDGUBmBDANgZwKYBoQB7KAbRACYBOAdgA4QBdAXwJgBYpRlJ1t8jSIWtQBs1CAWFjyISaOoBmWSBEBWatRnMWIVZxAAHKGAIGjkcqqZMgA

thos one is an article on methods to de-bounce switches;

https://www.digikey.com/en/articles/how-to-implement-hardware-debounce-for-switches-and-relays

Good luck with your project Ron,

kind regards,

LouisR

Posted by: @will

Posted by: @zander

Sorry for taking up so much bandwidth for a dumb switch.

Well, I personally am finding it to be interesting reading the replies.

Agreed.

Posted by: @zander

@ronbentley1 I think we are talking about two different things. I do not plan on any kind of polling, there may not even be a main loop if I can avoid it. My system is interrupt driven, one of the things generating the interrupts is a mechanical switch. Anytime there is a hardware solution that is the best solution in most cases. 

Ron 

Your solution therefore lies in a hardware solution for debounce.

How long the debounce period is is an interesting question, but it should not matter so long as the interrupt is triggered post debounce when switch has reach a settled state.

Good luck

@zander

I read a mercury doesn't bounce?

Another example I see on the internet is using two inputs per switch.

The internal pull resistors could also be used.

To enlarge image right click image and select open link in new window.

  If (PIN_2 == LOW)
    switch_ON = true;
  else
    switch_ON = false;

@robotbuilder 

Hi,

I'm afraid a mercury switch will also exhibit bounce.

I also dont feel the circuit above will take out any bounce, and 2 pins? I would be interested to understand others' point of view on this.

There are many sources for a hardware debounce switch circuit on the net, so you 'pays your money and takes your pick' - a hardware solution or a software solution.

Hi again,

This is a very helpful article and covers hardware  debounce  in several ways.

And here is a link to the ultimate debounce circuit, or so he says 😎 

@ronbentley1 I am beginning to think the software approach is the way to go if I can figure out an ideal number of tasks and task priorities and on which of the two cpu's to run them. It's been a while since I did something like that but I do recall it is NOT intuitive, and in fact many/most get it exactly backwards. I have not come across the scheduling parameters and what if any API calls I have available other than the priority setting and CPU affinity, but it's still early.

The reason for the re-think is two-fold, if the hardware fix proves to be not correct, it's a pain to change it whereas software is just that, soft and now with OTA quick. The second reason is that so many of you have encouraged me that way. 

I will still experiment with a touch sensor but two of the stations are in the bathroom and the humidity could be a problem.

What about an IC that handles debouncing for you?

@yurkshirelad Ok, I am curious, tell me more.

Posted by: @zander

@yurkshirelad Ok, I am curious, tell me more.

I can't remember where I read about them, so try a Google search. I remember reading about them somewhere so I don't know if they're any good or even available any more (!). I haven't tried using them myself so I can't really comment on them. Sorry. 😀 

A very random search gave me this, though I don't know if they're still available: https://www.onsemi.com/pdf/datasheet/mc14490-d.pdf

Edit: if they are available, they might be damn expensive!