TheBackShed.com

Forum Index : Microcontroller and PC projects : Software debounce

Author

Message

palcal

Guru

Joined: 12/10/2011
Location: Australia
Posts: 1993

Posted: 04:04am 14 Mar 2024

Copy link to clipboard

Print this post

@ Disco4now
Back in Jan 2023 you posted some code for debounce,

Quote Here is what I used for the tipping bucket to de-bounce it. I probably got it from TBS somewhere.

DIM INTEGER RainPin=15,rain_count,rain_last

Setpin RainPin, Intl,rainguageClick,Pullup
.
.

'rainguage sensor interrupt
Sub rainguageClick
local thistime
thistime=Timer-rain_last
rain_last=Timer
If thistime>10 Then
rain_count=rain_count+1
Print "Rain click "+str$(rain_count)
End If

End Sub

I can't get my head around what is happening here.
Can you or someone explain to me how this is debouncing the switch.

"It is better to be ignorant and ask a stupid question than to be plain Stupid and not ask at all"

phil99

Guru

Joined: 11/02/2018
Location: Australia
Posts: 2640

Posted: 04:17am 14 Mar 2024

Copy link to clipboard

Print this post

The way I read it is if the interrupts are more than 10mS apart it is a valid transition, if less it is a bounce and ignored.

palcal

Guru

Joined: 12/10/2011
Location: Australia
Posts: 1993

Posted: 04:19am 14 Mar 2024

Copy link to clipboard

Print this post

Yes that makes sense, thanks.

"It is better to be ignorant and ask a stupid question than to be plain Stupid and not ask at all"

PeteCotton

Guru

Joined: 13/08/2020
Location: Canada
Posts: 543

Posted: 04:16pm 14 Mar 2024

Copy link to clipboard

Print this post

DIM INTEGER RainPin=15,rain_count,rain_last

Setpin RainPin, Intl,rainguageClick,Pullup
.
.

'rainguage sensor interrupt
Sub rainguageClick
local thistime
thistime=Timer-rain_last
rain_last=Timer
If thistime>10 Then
rain_count=rain_count+1
Print "Rain click "+str$(rain_count)
End If
End Sub

Also, if you change the variable thisTime to millisecondsSinceLastClick, and rain_last to timeOfLastClick then it might be a bit easier to read.

DIM INTEGER RainPin=15,rain_count,timeOfLastClick

Setpin RainPin, Intl,rainguageClick,Pullup
.
.

'rainguage sensor interrupt
Sub rainguageClick
local millisecondsSinceLastClick
millisecondsSinceLastClick =Timer-timeOfLastClick
timeOfLastClick=Timer
If millisecondsSinceLastClick >10 Then
rain_count=rain_count+1
Print "Rain click "+str$(rain_count)
End If
End Sub

But to be really picky - I think this code resets timeOfLastClick every time you click - i.e. if you kept spamming the click faster than 10ms it would always reset timeOfLastClick and never increment rain_count. So personally I would make one last change and rename timeOfLastClick to timeOfLastValidClick and move it into the IF statement. But that is being really really picky. It almost certainly wouldn't make a difference in the real world. However it does allow you to standardise on your maximum measurement speed which might be useful. i.e. The code below will always max out at 100Hz. If you increase the number in the IF statement to 25, then you move to 40Hz maximum etc.

DIM INTEGER RainPin=15,rain_count,timeOfLastValidClick

Setpin RainPin, Intl,rainguageClick,Pullup
.
.

'rainguage sensor interrupt
Sub rainguageClick
local millisecondsSinceLastClick
millisecondsSinceLastClick =Timer-timeOfLastValidClick
If millisecondsSinceLastClick >10 Then
timeOfLastValidClick =Timer
rain_count=rain_count+1
Print "Rain click "+str$(rain_count)
End If
End Sub

Edited 2024-03-15 02:22 by PeteCotton

TassyJim

Guru

Joined: 07/08/2011
Location: Australia
Posts: 6283

Posted: 08:32pm 14 Mar 2024

Copy link to clipboard

Print this post

I have used variations of the rainguageClick for many years.

When I changed my weather station to emulate a Davis station, I changed to hardware debounce. I read the rain counter every two seconds and if the count has advanced 5 or more clicks, assume a false reading and ignore, otherwise treat the count as valid.
This allows for a downpour and reduces false reads cause by wind etc.

Jim

VK7JH
MMedit

palcal

Guru

Joined: 12/10/2011
Location: Australia
Posts: 1993

Posted: 08:56pm 14 Mar 2024

Copy link to clipboard

Print this post

Thanks for all of that, I totally understand what is happening now.

"It is better to be ignorant and ask a stupid question than to be plain Stupid and not ask at all"

PhenixRising
Guru

Joined: 07/11/2023
Location: United Kingdom
Posts: 1383

Posted: 07:26am 15 Mar 2024

Copy link to clipboard

Print this post

FWIW: I have some cases where I need to be absolutely sure that I have a valid signal:

Inspired by a FOTS example: 40ms of steady state.

pintst% = 19
retcode% = 0
retmem% = 0
ctr% = 0

Do:Loop While Inkey$ <>""
SetTick 5, scanin

Do
ctr% = 0

Do : Loop Until ctr% = 8
If retcode% <> retmem% Then
If retcode% = &HFF Then
Print "Down", Hex$(retcode%)
EndIf
If retcode% = 0 Then
Print "Up", Hex$(retcode%)
EndIf
retmem% = retcode%
EndIf
Loop

Sub scanin

Static accum%

Select Case ctr%

Case 0
accum% = Pin(pintst%)<<7
Inc ctr%
Case 1
accum% = accum% + (Pin(pintst%)<<6)
Inc ctr%
Case 2
accum% = accum% + (Pin(pintst%)<<5)
Inc ctr%
Case 3
accum% = accum% + (Pin(pintst%)<<4)
Inc ctr%
Case 4
accum% = accum% + (Pin(pintst%)<<3)
Inc ctr%
Case 5
accum% = accum% + (Pin(pintst%)<<2)
Inc ctr%
Case 6
accum% = accum% + (Pin(pintst%)<<1)
Inc ctr%
Case 7
retcode% = accum% + Pin(pintst%)
ctr% = 8
Case 8

End Select

End Sub

Print this page

To reply to this topic, you need to log in.