That will not work. The /90 here applies only to the minutes part of any time of day, going from 0 to 59. So this will match every hour on the hour.
I proposed PR #38282 which will make the above configuration invalid. Things can still get confusing though as e.g. minutes: /25 will match xx:00, xx:25 and xx:50 … so it waits for 25 minutes two times and then for 10 minutes.
(I don’t know of a simple way to fire every 90 minutes.)
Can you please try this? You need to setup an automation which has trigger type Template. Please adjust timeout_in_minutes variable and let me know.
Information: this will check when specific automation had run last time and if X minutes had passed. If the automation has never run yet, it will take last updated time instead of last triggered. So, it is not going to be in 3:00 - 4:30 - 6:00, it will try to execute in every 90 minutes from a starting point.
PS: if you are looking for guaranteed solution, this cannot be the one please discard this possible solution, other way around, this is best intention and good will. Feel free to discard or give it a try!
You have posted a link to the Time Pattern Trigger. Did you see amelchio’s post (above) explaining why a Time Pattern Trigger cannot be used for a 90-minute interval?
This is a community platform, not a customer service, we are doing our best to help. So, I am sharing my personal experience and my thought, so, if it works perfect, if not, i will learn something new. I do not have any obligation to be sure 100% correct/valid/tested other than good intentions. If you are looking for answers which are tested/validated, I will not be able to share them and you are free to discard my messages.
Your reply only needed to be “No, didn’t test it.” However, that was evident from the fact you proposed using the Time Pattern Trigger when amelchio, one of Home Assistant’s developers, already explained it couldn’t be used to repeat every 90 minutes. Yet you proposed it anyway (10 months later).
Nowhere in the documentation does it suggest the options are additive.
The “danger” is that people will stumble across your recommendations and waste time trying to get them to work. If I ever suggest something that turns out to be wrong (not unknown ) I go back and edit the post to clarify my error (recent example).