I feel so sorry for you guys...having to settle for substandard devices to supply your sad sad addiction.
3 Likes
I ordered mine online from Home Depot. They were supposed to ship in 2 weeks, but came the next day. This was early this week.
2 Likes
3 more wallplates worth of switches upgraded this evening. 1 x pico, 3 x dimmers, 1 x switch, and 1 x fan controller.
Making decent progress. These devices are so much more responsive compared to my old GE switches (Z-Wave, but not Z-Wave Plus.)
10 down, 10 to go!
3 Likes
Pico button presses generate a lot of RF traffic. Because they're Tx only devices, they have no way to know that they've been "heard" by the relevant devices. So, when you press a button they actually transmit repeatedly (pretty much in sync with the LED flashes on the Pico). In the case of two Tx only devices talking at once, they just Tx several times over the course of a second or two and statistically at least one message from each makes it through the noise (a-la the concept behind CSMA/CD). Devices that are Tx / Rx capable (like the smart bridge), are presumably smart enough to "wait their turn."
One of the reasons Caseta works so reliably is the Picos talk directly to the dimmers, then the bridge gets updated after the fact.
https://www.reddit.com/r/homeautomation/comments/anb2ei/comment/efumoho
This means that we can never get the same responsiveness if button presses are routed through HE. Direct pairing (using the Lutron system) is the only way to get instant controls.
1 Like
Good info. Any idea if pairing a Pico to a dimmer through the Lutron app ends up directly associating the dimmer and Pico just like pairing them without a bridge?
Based on my discussions with Lutron support a few years ago, I was left with the impression that Caseta pairing was one to one. So essentially the pico can be paired to the hub or the light but not both.
The reason I even had this discussion was because I had the Lutron occupancy sensor connected to my caseta wall switch and could not also connect it to my smart bridge. Support advised that I would have to unpair the switch/sensor and then pair the switch to the bridge. This made the occupancy sensor useless if I wanted any smarts for my wall switch.
Unless I misunderstood what I was told or Lutron fundamentally changed their pairing process over the years....then the answer to @destructure00's question above is no.
Regardless, the response time of my picos when controlling caseta switches in pretty much instant. Milliseconds don't matter to me if they are not easily perceivable.
My experience so far is that if I pair a Pico to a dimmer through the Lutron app (like in a 3-way setup), I get instant response. If I route the automation through Hubitat (using Button Controller or similar), I get a 1-2 second delay.
I would guess that happens because the commands have to traverse more hops.
Pico->Bridge->Hubitat->Bridge->Switch (routing through RF and Telent as needed)
When configured through the Lutron App
Pico->Bridge->Light (all through RF)
Could be. Here's my comparison scenario:
I have a Pico (using Fast Pico driver) set up through button controller to use the middle button to toggle my fireplace on and off. The fireplace is controlled by another Caseta on/off switch. I also have the on/off switch on a dashboard.
If I tap the Pico button, I get the delay. If I tap the dashboard tile, the switch is on instantly before I can pull my finger (hee hee) away from the tablet screen.
If it was solely a "hops" issue, I'd think there would still be a short delay using the dashboard. Makes me think that Hubitat sending commands to Lutron bridge is immediate, but maybe there's a delay in Hubitat receiving info (i.e. the button press) from the Lutron bridge. But this can't be right...if I watch the Hubitat logs, the button press is registered immediately and the on or off command is sent immediately, and then a short delay happens before the switch actually changes. I don't understand. But I don't really care much either, it's not enough of a delay to cause a problem.
With the dashboard press, the bridge would only receive one telnet command to turn on the light.
With the pico involved, the bridge would have to send out a telnet command (button pushed) and then immediately recieve a telnet command (turn on the light). Maybe there is a telnet queuing issue on the bridge when it sends and receives telnet communications almost immediately. I'm far from an expert on any of this.....just spitballing.
1 Like
Whether it is multiple hops or Pico RF bottlenecks, there is a performance (delay) issue for Lutron devices controlled via Picos programmed in HE.
As a result, I've changed any automation that is scene-like (multiple devices). Those scenes should be programmed on the Lutron bridge/main repeater. And then layer on non-Lutron devices in HE. The Lutron scenes are called via button pushes in HE (the bridge has 47 user configurable scenes (buttons)).
This is more work, but it is definitely faster and provides backup for Lutron devices if HE is not in the system.
1 Like
I've been doing the same thing for the same reasons. As you say, it's more work but the end result is more pleasing.
I completed my initial ground floor installation of Lutron Caseta Switches, Dimmers, and Pico remotes. Everything is working very smoothly.
Total device counts
- 8 x Switches
- 8 x Dimmers
- 4 x Pico remotes
- 1 x Fan controller
- 10 x Claro wall plates
I just ordered some more dimmers from eBay which should allow me to remove the remaining 4 x GE Z-Wave switches from my house. I love the clean look of these Caseta switches and how responsive they are using the Telnet interface. I am finally able to reliably "Automate" my home lighting using door contact sensors (HubDuino) and motion sensors (Iris Gen 2), without worrying about failures. I also love how trivial it is to setup 3-way light switches using Pico remotes.
3 Likes
The three-way setup using the Pico is just fabulous!
Agreed! And four-way, and one Pico controlling two or more dimmers... All working instantly even if you don't have a hub.
The ONLY thing that bugs me about the Caseta stuff is that the diagonal across the dimmer buttons should be a "backslash" (top left to bottom right), not a slash, so the button on the right goes up, not down. I can (and do) live with it, but it still bugs me sometimes after almost a year! It would also be nice if you could buy Picos colored like the buttons on the plug-in dimmers, so you could tell which side is which without marking them with a sharpie.
Are you referring to something like this?
http://bim.lutron.com/item/sub-controls/pico-remote-control/pj2-2brl-gwg-l01
And some of them are actually available on eBay...
Interesting that these don't have the round button in the center. Also, is there any difference between one for shades and one for lights?
I really don't know what is different aside from the silk-screening. I do know that some users have tried to associate a typical 5-button "lighting" Pico with the new Caseta Fan controller with the Lutron App - this did not work. However, a direct association between the two did. I am guessing the idea here is that if you have lights, shades, and fans in the same room, adding picos to that same room would cause confusion if one pico tried to command all three classes of devices. So, a lighting pico would associate with switches/dimmers... whereas a fan pico would associate with fan controllers, etc...
I am pretty new to the whole Lutron ecosystem, so I am really just taking an educated guess at this time. I am sure Lutron support would be happy to answer questions. Others have stated that Lutron support is very good.
None, aside from the engraving on the buttons.
Is this some sort of international standard I'm not familiar with? The placement as-is seems just as intuitive to me, which is less than a regular switch but among the best for smart switches that don't look like traditional ones. 
1 Like
Probably not an "international standard", but left or down to decrease, and right or up to increase seems pretty standard in the U.S., and I suspect all/most cultures with left to right writing systems would be the same way, but I'm not certain.
