Gimzo 2 custom On/Off button?

Check out this post where I added an external power and reset button: http://www.element14.com/community/community/designcenter/single-board-computers/gizmo2/blog/2015/01/28/gizmo2-adding-external-pwr-and-rst-buttons

Oh wow, so much great tutorials I'm having hard time finding

So you didn't use a resistor?

Is there any requirement for what button can be used. All I know is buttons have a maximum current rating, this shows how little my electronics knowledge is. But it's going to be fun programming the Gizmo after I get this figured out.

PS. Is that a 3d printed enclosure? That's another hobby of mine.

Yes,

mine is here: http://www.element14.com/community/community/designcenter/single-board-computers/gizmo2/blog/2015/01/26/gizmo2-3d-printable-enclosure

and Brian made an even better one: http://www.element14.com/community/community/designcenter/single-board-computers/gizmo2/blog/2015/01/31/gizmo-2-another-3d-printed-enclosure

Nice.

About the connection to the pins though, you didnt need a resistor?

Hi Mark,

It is your choice. It will function either way. Personally I would use it, because no components are perfect and it will look after your switch (provides a current limit - those small board switches might not handle much current), provides a bit of debounce (not that it should matter for a power-on circuit but anyway..)  (EDIT - also provides some protection if you accidentally connect the wire to an incorrect supply wire accidentally) - plus the original designers used it, so even if we cannot see their purpose it is harmless to add it.

But equally a body of engineers would just omit the resistor - there is not a lot of energy stored in that bit of circuit for example. So if it were a courtroom it would be a hung jury ; )

You can use any switch you like for the purposes of this button function, as long as it is a normally-open switch.

No, I have not used the resistor with a reason (not sure it is correct). Seeing in the scheme just as has suggested Shabaz in his post, my though is the following: If I should manage a push button with a GPIO port it is correct to use a resistor and eventually also a capacitor, just as the circuit in the scheme of these two buttons. But in this case the push button sends to GND exactly the signal pin "reset" and "system power" that - in the same Gizmo2 scheme - it is shown with a resistor capacitor etc. So I have excluded to add one more.

Pay attention as it is not sure that I am right

Enrico

If I should manage a push button with a GPIO port it is correct to use a resistor and eventually also a capacitor, just as the circuit in the scheme of these two buttons. But in this case the push button sends to GND exactly the signal pin "reset" and "system power" that - in the same Gizmo2 scheme - it is shown with a resistor capacitor etc. So I have excluded to add one more.

 

Not sure I understand.

For the record I don't mind using a resistor, it's a <$1 difference, I just wanted to test out right now if it works with my switch.

If I should manage a push button with a GPIO port it is correct to use a resistor and eventually also a capacitor, just as the circuit in the scheme of these two buttons. But in this case the push button sends to GND exactly the signal pin "reset" and "system power" that - in the same Gizmo2 scheme - it is shown with a resistor capacitor etc. So I have excluded to add one more.

 

Not sure I understand.

For the record I don't mind using a resistor, it's a <$1 difference, I just wanted to test out right now if it works with my switch.

I mean that as the pins for power and reset are specifically done for this scope, I suppose that as they don't mention extra-components the two buttons are sufficient.

Anyway: I have done experiments and an extra-component is needed as the bare button it works but is not debounced. So I need a small 555 to debounce the buttons. Those shown in the images was just a stupid test to see it working and evaluate the idea for a prototype. Also and external power is needed to manage extra SSD and extra fan and other stuff.

Enrico

I see , however, is debouncing a problem with the on/off button (not the reset)? I just need the on/off button myself.

And did I get this right?: resistor will help with debouncing? Sorry, my electronics is lacking.

It is a problem of both.

In the rest it is not so influencing as a repeating reset pressing does not generate more than multiple rest and the effect is anyway the expected (but it is not the best solution without debouncing).

While powering-off debouncing instead is meaningless: you should keep the power pressed for few seconds. Instead I have noted it has influence when you power-on.

Enrico

While powering-off debouncing instead is meaningless: you should keep the power pressed for few seconds. Instead I have noted it has influence when you power-on.

I'm sorry, can you p[lease reword this part?

And will the resistor be a solution for debouncing?

Debouncing is the phenomenon that happens when you press a button but as a matter of fact the responsiveness of the system if so fast that with some other factors the effect is like you press the button twice or more.

The resistor is for current limitation (or other usages, depending on the kind of connection respect the button) when the button signal - high - is put to ground to generate the I/O event. This has nothing to do with the resistor and - unfortunately - it is a bit more complex. few more but it is. Take a look to this link: http://www.labbookpages.co.uk/electronics/debounce.html Here is clearly explained what happens when you press a button and how you can manage the physical event avoiding the debounce with an OpAMP, a diode and a capacitor. The principle is that the capacitor generate a minimum delay in the button press (due it is charging) then when the OpAMP AND the button signal and the capacitor discharge the output event is effective. Also keeping some microseconds more the button pushed (that is you finger pressing reaction time) nothing happens because before a second event occurs it is needed not only that the button is pressed but also that the capacitor is charged.

I suggest to try with a breadboard the case of the power off, that needs the button pressed for some seconds. The interesting aspect of the example circuit in the link above is that it shows a schematic with acceptable values for the discrete components.

Enrico

Sorry the OpAMP is a schmidt trigger to be precise (it is impossible to edit the comments ?)

I understand. But will that be an issue for the on button? To turn off you have to press and hold the same button for few seconds. Not like debouncing would changing anything here, right?

Frankly I suppose this can represent a problem but to be sure in this particular case (it is the first time I do this on a Gizmo2) I should test it. My suggestion is to add the resistor and ignore the debouncing circuit. This can be done in minutes with a breadboard. It is probable that this works as I suppose.

The debouncing effect is almost common with push buttons and can be solved - as shown in the page I got you before - via software or hardware. In this case it is not possible to solve it by software so a small circuit should be done. For more detail on how the debouncing should be implemented and why this occur, take a look to the following pdf: http://www.eng.utah.edu/~cs5780/debouncing.pdf

So im confused, you think debouncing will be an issue here, or the opposite?

Hi Mark,

Enrico is suggesting try it with a resistor, see if it works for you.

If you observe some behavior that is undesirable (e.g. having to hold the button pressed for a long time) then it can be considered further and you then may need a more complex circuit.

His linked article is extensive, and provides much detail on how different debouncing circuitry works.

Whether you actually need delays, debouncing, etc., can be answered in detail either by Gizmosphere (since they designed the board), or a detailed examination of the circuit and the power management IC user guide (which would require some time to investigate). An alternative approach (as suggested by Enrico) is to just try it, observe the results and see if it fits your required button-press behavior.