Trigger file not working as expected.

Chriss

Hi *,

I have trouble with an trigger. I have this trigger configured:
(Mechanical push button on the input to GND to turn a led stripe on and off)

M950 J1 C"^0.io1.in"
M581 P1 T2

The trigger2.g

if fans[5].actualValue == 1
   M106 P5 S0
else
   M106 P5 S255

The idea is now to toggle between S0 and S255 for P5. That works from time to time. Mostly the led goes off when I release the push button, I have the feeling that this happens mostly when I do not press the push button not long enough.

I actually co not get it. I see that the "actualValue" toggles when I trigger the push button but falls back when I release it. That makes no sense to me.

I tried it with, and without the pullup resistor.
Should I compare "actualValue" and "requestedValue" to work around that?

Cheers, Chriss

fcwilt

@chriss

Hi what hardware and what firmware version?

Frederick

Chriss

@fcwilt

Sorry, have forgotten:

Mini5+
3.4.0beta6

fcwilt

@chriss

Well I cannot help with any firmware problems as I don't use any of the betas - I'm on 3.3.

Pull-ups are part of the hardware so you should not use the ! character.

Triggers are activated on the transition from one state to another so it's not a question of holding the button down.

By default they trigger on the rising edge which in your case would be when you release the button. You might want to add S0 to your M581 command to trigger when you press the button.

You might try adding a G4 P250 at the end of the trigger code.

Frederick

Chriss

@chriss

Hi Frederick,

I tried what you mentioned, that made it even worth. The push button acts now exactly as a push button. So the LEDs go one if I press the button and go of when I release it button.
The plan was to toggle between the states. So first push = on, second push = off.

Cheers, Chriss

fcwilt

@chriss said in Trigger file not working as expected.:

@chriss

Hi Frederick,

I tried what you mentioned, that made it even worth. The push button acts now exactly as a push button. So the LEDs go one if I press the button and go of when I release it button.
The plan was to toggle between the states. So first push = on, second push = off.

Cheers, Chriss

Mechanical switches don't change cleanly from one state to another - the contacts bounce and the state can change several times from one state to the other before the bouncing stops and the final state is obtained.

That may be what you are seeing.

Try upping the P parameter in the G4 to 1000 and see if that makes any difference.

I use push buttons for triggers BUT I am not trying to toggle a state, just set a state.

Frederick

Chriss

@fcwilt said in Trigger file not working as expected.:

That may be what you are seeing.
Nope, this is not the case. I see it clearly when I press the button and hold it. It goes on, and it goes off when I release it.

There is a other problem. The trigger2.g is triggered in both situations in my opinion. This explains perfectly what I see when I push and release the button. the actualValue and the requestedValue change accordingly int he Object Model.

Try upping the P parameter in the G4 to 1000 and see if that makes any difference.

I tried that, but this does not help. I understand what you mean, but this is not a problem with the bouncing of the switch. It is, in my opinion, the problem that the trigger get executed at the push and at the release.

Cheers, Chriss

fcwilt

@chriss said in Trigger file not working as expected.:

It is, in my opinion, the problem that the trigger get executed at the push and at the release.

Well, it is beta firmware so it could well be a bug. Or it could be a bug that has existed for some time but simply has gone unnoticed. As I mentioned it would not have effected they way I have used triggers.

If I get a chance I will see if I can toggle a value in 3.3.

Frederick

cosmowave

@chriss said in Trigger file not working as expected.:

It is, in my opinion, the problem that the trigger get executed at the push and at the release.

Normally this happens when the switch is bouncing...

Chriss

@cosmowave

Yes... I know what you mean... Anyway, I tried it now with:

config.g

M950 J1 C"^0.io1.in"
M581 P1 T2 S0

trigger2.g

if fans[5].actualValue == 1
   M106 P5 S0
   G4 P1000			; Wait 1000ms
else
   M106 P5 S255
   G4 P1000			; Wait 1000ms

G4 P1000			; Wait 1000ms

The LEDs go on now for two seconds. (What a surprise. )

I'm very sure that the trigger2.g gets executed twice, at the push and at the release.

Cheers, Chriss

Chriss

@chriss OK, we can close it now....

It works now... do not ask me why but it works. I rebooted the printer for some other changes and it seems to work now as expected..... :?

Anyway, thanks for your hep @cosmowave, your direction with the S0 did it at the end.

Cheers, Chriss

OwenD

@chriss
Try this
EDIT
You need to check the pin twice to ensure it's finished bouncing
I should have had my coffee before posting

if sensors.gpIn[1].value=1
   G4 P500
   if sensors.gpIn[1].value=1
      if fans[5].actualValue == 1
         M106 P5 S0
      else
         M106 P5 S255

You may need to increase P500 depending on how long your button is bouncing

fcwilt

@chriss said in Trigger file not working as expected.:

@chriss OK, we can close it now....

It works now... do not ask me why but it works. I rebooted the printer for some other changes and it seems to work now as expected..... :?

Anyway, thanks for your hep @cosmowave, your direction with the S0 did it at the end.

Cheers, Chriss

You sure?

I just tried it under firmware 3.3 and the results were inconsistent, sometimes worked, sometimes didn't. More often then not it would trigger on push and release.

I'm pretty sure it is contact bounce which is surprising because I would have thought the firmware code for triggers would have de-bounced the inputs.

Frederick

cosmowave

@fcwilt No. The firmware makes no debounce.
My flowsensor of the watercooling switchs a relay, which has bouncing contacts.
I use also a G4 in my trigger2 macro for software de-bounce.

Chriss

@fcwilt said in Trigger file not working as expected.:

You sure?

Not 100% about the S0, but sure that it works now.
I guess that your "G4 250" makes the de bouncing here. And the behavior is consistent.
The 2nd press of the switch will not do anything when I press the switch fast enough. (Below 250ms)
With the finally final settings is that gone:

if fans[5].actualValue == 1
   M106 P5 S0
   G4 P25			; Wait 25ms
else
   M106 P5 S255
   G4 P25			; Wait 25ms

G4 P10			; Wait 10ms

It is not easy to press the button in 35ms after the first press.

I guess that something in the live configuration was mixed up and and have forgotten to reboot to verify the config.

Cheers, Chriss