I’m sorry, but there is a language barrier after all. now you know what you think
that doesn’t work either
don’t know why the scene would repeat every 1 second
as i say if R5 is used in the scene scene 2 will not perform.
Have you checked the scene logs to see if it is being executed?
as i say if R5 is used in the scene scene 2 will not perform. because it won’t start according to When using the zb button in scene 1b, the scene is also started according to the protocol.
Set the scene delay to 10 seconds.
*)do scene 1A.
wait 30 seconds.
*)Perform scene 1B.
wait 30 seconds.
*)take a picture of the scene log.
It doesn’t work that way. Your Scene 2 doesn’t continuously monitor anything. Scenes do not monitor. There are simple automations of the “if-then” type. In your example, it goes like this: if lamps change from off to on, then wait 20 min. There’s no next step defined. This scene waits 20 min. and do nothing else. On the “Komplet foto” screenshot the situation is different, because the next step is defined. It means that after 20 min. the lamps go to “off” state. No other conditions are set. The scene is executed despite of anything else. If you switch the lamps off manually, the scene will go on anyway. When you switch the lamps off and on again during the 20 min period, another instance of your Scene 2 is fired.
As pointed out before, scenes in the eWeLink cloud platform are simple automations that are not capable of performing complex scenarios, including monitoring. That’s why you need to turn to something else if you want to achieve your goal. Getting angry and grumbling won’t help or change anything. This is how it was conceived.
To put it simply, the creation of complex automations (scenes) requires precision to achieve the desired results. The way in which the implementation of the next steps takes place is also a difficulty, especially when the states are to be monitored. What is supposed to happen when conditions change? Is the automation to be continued, restarted, queued?
you obviously don’t understand the reason for this thread. the problem is not in scene 2, it behaves as expected, but in the fact that if the light is activated via R5 or s-mate. so the scene will not be executed. and if it is activated via the application or by pressing the button, everything will work as expected. I expect that sonoff and ewelink cooperate to such an extent that the three sonoff buttons will perform the same action in the ewelink application and not that the ewelink remote technology behaves differently than zb.
I am very sorry but you are the one who does not understand. Let me repeat myself: Getting angry and grumbling won’t help or change anything.
As an aside, there is no compulsion to use eWeLink solutions if you feel they do not meet your expectations. Happy automating 
No comment.
Have you checked the scene log?
In the “test 5” scene, there is only one condition option and no more can be added. is this the reason why the subsequent behavior of the scenes is different?
for example, I changed the shutdown time to 30s.
S-Mate and R5 devices as well, do not keep state, they just trigger events. If you want to do something with switch 2 based on smate switch 1, you have to ask the status of the device controller by SMATE switch 1, not the status of the SMATE switch itself.
1 Like
maybe a language barrier what do you mean? in scenes it ensures the v/status of the smat sensor but the result of the scene it performs.
We tried to explain this earlier in a more generalized way. @mario.bianchi gave a specific reason. So understand that you need to revise your approach to automation ideas. And no, I don’t think this is a language barrier problem.
generally i could watch youtube. the answers are as if my question was not understood.
I’ll ask another way
Model 1
1.
scene 1.a
if : zb click button
then: manual scene “reverse” (switches the state of the light)
Do you agree that this is okay?
2
scene 2
if: light 1 on or light 2 on
then: 30S delay
manual scenes “off”
do you agree that it is okay?
do you agree that scene 2 will be done based on the state of the light on?
Model 2
scene 1.a
if : s-mate click
then: manual scene “reverse” (switches the state of the light)
Do you agree that this is okay?
2
scene 2
if: light 1 on or light 2 on
then: 30S delay
manual scenes “off”
do you agree that it is okay?
do you agree that scene 2 will be done based on the state of the light on?
if I want to better understand how scenes work, a general answer is not enough.
my question is
why “model 1” works
why "model 2 " does not work