How many times can I safely write parameters to the Red dimmer switch? If it can only handle something like 1,000 or 10,000 write cycles then I will need to reconsider my automation strategy, but something like 100k cycles would probably outlive the natural life of the switch for most use cases so I could continue doing what I am doing.
Maybe more generically stated: what is the expected lifetime of the switch, and what use case is that based off of?
I can’t find the post but I had a similar question buried in a thread about a year ago.
There was at least one person updating the color parameter of the switches several times a day without issues.
I’ve had my switches for two years now, running constant notifications that update each time a door opens, alarm is armed, etc and have not had any of my switches or individual LEDs fail. The switches run a red chase effect all night which I’m guessing is a constant workout. I personally don’t recall seeing a post in this forum from anyone having a switch just die, unless it was a short while wiring or self inflicted firmware brick. I’m at the point I’m stock piling switches for my future, larger home so everything matches lol
Notifications should go to ram which doesn’t have a limitation, but changing the stored parameters must be written to an eprom so the parameters are retained during power cycles. The actual manufacturer of the switches should be able to easily provide the number of write cycles of the eprom inside. @Eric_Inovelli or @EricM_Inovelli could ask them for the number of writes the memory can handle.
So, should be good. The parameter cell in memory should only be written when a user changes the color so there should be no need to worry about the chip writing to that cell for housekeeping.
If that parameter is stored in the extra MCU then the answer could change though. But then, it could change if they didn’t use memory that meets that spec.
To get certified they have to use spec components as far as I am aware.
Thanks everyone for the replies. Makes me feel a lot less worried. I am new to Z-Wave (and home automation, for that matter) so I didn’t know about the 1 million write endurance requirement.
I am doing something like @BuilderTroy where I am writing the default power on level based on time of day. Writing parameters 9 and 10 twice a day is 4 writes per day minimum, assuming nothing else. But I also currently have the “remember last state” set, so each time the switch turns on or off is another write.
Regardless, with 1 million writes being the required endurance, it seems that I have nothing to worry about. That’s 11.4 parameters written per hour for 10 years straight… and I am currently ~10 per day. Well, as long as they meet the requirements 
Bummer! Looks like the EPROM chip will only last 273 years with that write cycle. Time to refactor! haha! Go get em! 
I’ve written every integer percent change from 99 down to 1 with circadian rhythym integration every single day for almost 2 years at this point without a single noted failure.
Just for reference :D.
Wow. I had not thought about this problem. I have set up my LED bars to show the current temperature (mapped to color with blue being 30 degrees F and red being 90 degrees). So, it gets written about once an hour. When I saw this post, I went in and updated the code so that it will only update temperature during the hours I’m awake, which should cut the number of write cycles per day down to about 2/3 of that.
