button that has push on/off

stilgar · 2012-12-23 10:12

I have searched, even looked at the obex. If I missed it, please point me to it.
But not found code to make a toggle method that 1 button is used to toggle a output.
Example, push it ounce and it turns on a led, push it again and it turns off the led.

thanks,
stilgar

Invent-O-Doc · 2012-12-23 11:19

Check out Learn.parallax.com for details on activating an led and reading a switch. Search debounce for good switch reading code.

kwinn · 2012-12-23 12:11

stilgar wrote: »

I have searched, even looked at the obex. If I missed it, please point me to it.
But not found code to make a toggle method that 1 button is used to toggle a output.
Example, push it ounce and it turns on a led, push it again and it turns off the led.

thanks,
stilgar

Each time the button is pressed you execute the line below.

	!outa[0]			' toggle pin 0

If the pin is high it will go low, if low it will go high

JonnyMac · 2012-12-24 15:20

Pretty easy, really, and no need for a full object. Debounce your input pin and then toggle the output. Easy peazy.

pub debounce(pin, state, ms) | mstix, cycles, t

  dira[pin] := 0                                                ' force to input

  if (state)
    state := 1                                                  ' force to 0 or 1

  mstix := clkfreq / 1000
    
  cycles := 0                                                   ' clear cycles count

  t := cnt                                                      ' sync timing
  repeat ms
    waitcnt(t += mstix)                                         ' wait 1ms
    if (ina[pin] == state)                                      ' in active state?
      ++cycles                                                  '  yes, bump counter
    else
      cycles := 0                                               '  no, clear counter

  if (cycles == ms)
    return true
  else
    return false

if (debounce(TRIGGER, 1, 100))
    !outa[LAMP]
    pause(150)

Harprit · 2012-12-29 12:00

Jon:
I have a question for you (again). We discussed this earlier but I need more assurance.
Does a switch not make multiple connections on both the make and break condition.
If this is the case don't we need two delays as follows:

detect make
wait
detect break
wait

to get it right

I am about done with the PASM book and would like to really clarify this, for the beginners, in the book.
Thanks
H
H

kwinn · 2012-12-29 12:15

Yes, a switch will make multiple makes/breaks each time it is pressed or released. The term is "contact bounce" and that is why you need a software routine such as your delay to ensure you are reading the switch state after it stops bouncing.

In both Spin and PASM I use a loop that reads the state of the switch and returns the state after it has remained the same x number of times through the loop.

Harprit · 2012-12-29 13:21

The problem is that the same switch is going to turn the led on and off on alternate presses.
I think this needs 4 total delays for one complete cycle
Am I right?

H

kwinn · 2012-12-29 13:45

Harprit wrote: »

The problem is that the same switch is going to turn the led on and off on alternate presses.
I think this needs 4 total delays for one complete cycle
Am I right?

H

Based on your first post and this question I am assuming you have:

- A push button on a prop pin that closes a contact when it is pressed and opens the contact when released.
- A led on another prop pin that each time the button is pressed you turn on if it was off, or off if it was on.

You only need to:

detect make
wait
detect break
wait
if led is on turn it off
else turn led on

Harprit · 2012-12-29 16:31

Now it is in my head. Solid.
Thanks
H

Mickster · 2012-12-29 19:41

My personal preference is to not allow such events to halt execution of the entire loop. the countdown value of OSAD must be determined based on requirements:

'PropBASIC code

DEVICE P8X32A, XTAL1, PLL16X
FREQ 80_000_000

LED PIN 1 OUTPUT
PB PIN 2 INPUT
OSAD VAR LONG = 0 'One Shot And Debounce

PROGRAM Start

Start:
Do
     
     If PB = 1 Then
         If OSAD = 0 Then
            Toggle LED      
         Endif
         OSAD = 1000000
     
     Else
         If OSAD >0 Then
             Dec OSAD
         Endif
     Endif
     
 'I can keep executing other stuff because the PB doesn't hold me up    
      
Loop
END

teganburns · 2012-12-30 00:31

Made a simple program in Spin called Push. Hope this helps! and if you are just beginning i would highly suggest the parallax official guide (http://www.parallax.com/Store/Books/Propeller/tabid/171/CategoryID/45/List/0/SortField/0/Level/a/ProductID/637/Default.aspx)

kwinn · 2012-12-30 05:33

Mickster wrote: »

My personal preference is to not allow such events to halt execution of the entire loop. the countdown value of OSAD must be determined based on requirements:

[/code]

I agree, not usually good to halt the loop. I did not want to make the answer any more complicated than necessary.

I have a small routine that reads in all the switches once each time through the loop, debounces them, and sets the switch variable to the proper state once they are debounced. Any routine that needs to check the switch status reads the appropriate bit of the variable for the state of that switch.

Mickster · 2012-12-30 06:32

teganburns wrote: »

Made a simple program in Spin called Push. Hope this helps! and if you are just beginning i would highly suggest the parallax official guide (http://www.parallax.com/Store/Books/Propeller/tabid/171/CategoryID/45/List/0/SortField/0/Level/a/ProductID/637/Default.aspx)

Unfortunately, this doesn't accomplish what the OP wants. The LED won't latch in this case. All it does is have the output equal the state of the input.

Mickster

Alexander (Sandy) Hapgood · 2013-01-14 18:22

Here's some pseudo code based on the code i use to debounce normally open switches. The code for 'Button1' will toggle an output.
This code has the debounce period at the end of the routine. Most programmers place the debounce period at the beginning of the routine. Normally open switches don't randomly push themselves down so if the button is sensed down it's because it was pressed down by the operator. The delay of 1/16 second will work for most switches and can be shortened or lengthened to suit to type of switch and your needs. Military specification switches are very reliable but they can bounce around for what seems like ages before finally settling down.

start         read system clock frequency

Button1       test for Button1 down
              if not down jump to Button2
              Output1 := not Output1
              debounce by waiting 1/16 of a second (clock frequency shift right 4)
              wait for button up
              debounce by waiting 1/16 of a second (clock frequency shift right 4)

Button2       test for Button2 down
              if not down jump to start
              complete some action
              debounce by waiting 1/16 of a second (clock frequency shift right 4)
              wait for button up
              debounce by waiting 1/16 of a second (clock frequency shift right 4)
              
              jump to start

cavelamb · 2013-01-15 12:12

This almost works right.
The problem is that it can repeat too quickly and look like it's not latching...
It's not really a debounce. Touchpads don't "bounce" inthe traditional sense.

A quarter second delay at the top of the loop helped settle it down.
That could be other code, of course, but for a demo (shrug)...

TOTO.SPIN

CON { Toto - Touch On/Touch Off }
  _CLKMODE=XTAL2 
  _xinfreq = 5_000_000
VAR
  LONG MS001  , Lstate[8]
  Byte Buttons
  
OBJ button: "QS_Button13"    ' touchpad driver

PUB Toto | B                      ' local variable B
  MS001 := CLKFREQ / 1_000      ' define 1 millisec 
  button.start( @Buttons )    ' send address of Buttons

  Repeat B from 0 to 7          ' initialize all off
     LState[B] := 0                  '    
                
  Repeat                                ' main loop
     waitMS(250)                    ' short delay to slow autorepeat
     Repeat B from 0 to 7       ' scan all buttons
       if Buttons & |< B            ' button pressed?
         if  LState[B] == 0         '  toggle associated LState 
           LState[B]   := 1  
         else                 
           LState[B]   := 0    
         '
          
'  next, display results
     Repeat B from 0 to 7       ' show states for all
       if Lstate[B] ==1       
          TurnON(B+16)        
       else                   
           TurnOFF(B+16)      
   
PUB WaitMS(W)                   ' wait for W milliseconds           
  WaitCNT (W*MS001+cnt)
  
PUB TurnON(pin)
    dira[pin] := 1
    outa[pin] := 1
    
PUB TurnOFF(pin)
    dira[pin] := 1
    outa[pin] := 0

Touchpad driver:

'CON { QS_Button13.spin }
'    { returns buttons packed bitwise in a single byte}
VAR                  
  LONG ButtonAdr, ButtonCog, decay
  LONG Stack[ 16 ]                 ' define my stack

PUB start( BAdr )                  ' start this in a new cog
  ButtonAdr := BAdr                ' save address of return byte
  if ButtonCog                     ' did the new cog start?
    cogstop(ButtonCog-1)           ' OOPS! no cog available
  ButtonCog := cognew(ButtonScan, @Stack) + 1
  decay := 4*(clkfreq/1000)
    
PUB ButtonScan | B, B1              ' local variables
  dira [0..7] := 111111          ' all pad pins outputs    
  outa [0..7] := 111111          ' all pad pins high   
  'main loop - scan the buttons   
  Repeat                            ' loop forever
     Repeat B from 0 to 7           ' QS LEDS are pins 0-7
       dira [B] := 1                ' make pin an output 
       dira [B] := 0                ' make pin an input
       waitcnt(decay+cnt)           ' short delay for some decay
       B1 := ina[B]                 ' read the pad
       if B1 == 0                   ' 0 here means pressed
          BYTE[ButtonAdr] |= |< B   ' set bit if pressed   
       else                    
          BYTE[ButtonAdr] &= !|< B  ' clear bit if not

button that has push on/off

Comments