14294 dimmer Ge zwave plus; considered instant status?

Edit: I could be mistaken. Does the behavior I am describing below match everyone else?

————-

14294 dimmer Ge zwave plus; considered instant status?

Not seeing instant status to hub, only instant status to zwave associated directly controlled devices.

Hello,

I just got one to test, with only a little while to actually play with it; this is what I am seeing.

Instant status when using zwave association, does not send instant status to hub.

When using it with zwave direct association, it does send instant status to the controlled devices.

So as I hold the paddle either up or down, the controlled devices dim with the switch.

However, it seems that status is only received by the hub when the paddle is let go.

Is this considered instant status?

This behavior is the same as previous generation of Ge dimmers.

Old behavior, after pressing and then letting go of the paddle, hub sees that the paddle was pushed, it then polls to get the light status.

In the thread discussion of the custom device handler I see some discussion about tweaking the delay which sounded like the delay for when the hub should poll the light dimmer for light status.

So for example, if using built in handler does not get instant status, but using custom does? If this is the case then I have something mis-configured as I get the same behavior with the custom handler as with the built-in “dimmer” handler.

“Instant status” is just a marketing term – – it doesn’t have any technical meaning, and different manufacturers use it in different ways. Most often it just means that no polling is required to get the status.

These days most Z wave device manufacturers use it for devices which use central scene commandsets to send information to the hub. But it’s not simultaneous notification. Particularly for devices which can process tap, double tap, and long hold, the device will have to decide which hold pattern applies and then send the appropriate scene number. But with zwave plus devices it’s also just used for the normal Association group one lifeline notifications.

So I’m not sure exactly what behavior you are expecting to see. The fact that it’s faster for association is expected because it doesn’t have to go to the hub, get processed there, and then get sent back to the other device. The association process does not typically use the term “instant status” at all. That’s not the trigger device telling the other device what the trigger device’s status is. That’s just the trigger device sending a “basic” (that has a very specific meaning in zwave) command directly to The target device.

I believe GE uses the same responsiveness standard that most Z wave manufacturers use, Which is anything up to two seconds.

Lutron, in contrast, sets a responsiveness standard of under half a second, but that’s much faster than most smart devices.

To answer your own question in more detail we would have to know exactly which DTH you are using. But the general answer is that direct association will almost always be faster than anything which goes through the hub.

1 Like

@JDRoberts

Thanks for the prompt response.

Let me try and clarify, I am making a stronger claim than just direct association is faster.

For dimmers, (switches this does NOT matter as much) the usability is impacted negatively for use in 3 ways if you cannot see the lights dimming as you interact with the switch.

What I am seeing status to hub is not sent until paddle is let go, regardless of how long paddle is pressed and held.

You can press and hold paddle for a long time and no updates are sent to hub.

This makes the dimmer NOT useable for a 3 way using the hub.

some background on my thinking on usability.

The minimum behavior needed for a switch to participate in a 3-way usefully/useable in contrast to a dimmer is much much less. Here is why,

case 1: Non-load 3 way switch, you click either up or down, let go of paddle, controlled switch then goes on or off. Interaction is done.

case 2: non load 3 way dimmer

Expected behavior: You press and hold paddle and light dims up or down as you hold and stops when you let go.

However, it seems dimmer is not sending status until paddle is let go. So, there is NO visual feedback in order to determine dim level.

So
#1) press and hold dimmer paddle and then guess when to let go.
#2) see if dim level is too high or low, if so go back to 1, potentially many times.

Furthermore the previous generation of Ge switch (NOT dimmer) was effectively instant status, so stating that new generation of Ge zwave switch is instant status does not add much

“Instant status” never applied to binary on/off switches– – the term was only applied to dimmers. That’s because Lutron had a patent on how dimmers reported to a hub.

I’m confused: when you say “seeing the lights dim as you hold the dimmer switch” are you talking about the lights which are controlled by The switch on their same circuit branch? That is, where that switch would control those lights even if you unplugged your hub altogether? Or are you just talking about an auxiliary switch where a message has to be sent to the master switch to do the actual dim adjustment?

Also… When people were talking about “delay” were they talking about the fade rate and ramp rate? That doesn’t have anything to do with the hub, it’s stored in the firmware of the device itself.

Yes I know instant status is a marketing term.

Either a device sends notifications to the hub or it does not. If it does NOT send status to the hub, then the hub must poll, then you run into the problem that the hub must poll the device often enough that it does not miss changes to status.

With the last generation of zwave switches, while they did not send the light status, they DID send press events, since the hub is notified that something happened, it can then request the light status. Which minimizes/eliminates the need to poll the device.

This does not work for dimmers because the light has not finished dimming at the time the paddle is let go, and not finished for some amount of time AFTER the paddle is let go.

While this is still better than needing to poll() in order to capture state changes, this is still a problem for a human who is interacting with the switch trying to get a specific dim level.

I imagine, (but have not actually used switches like the home seer switches), that with the leds on the switch itself showing dim level, this might be enough feedback to minimize this usability issue.

You are saying that the patent did NOT apply to switches, it only applies to dimmers.

As far as I know, this is effectively NOT true, does not matter to why, legal or not.

As far as I know those switches do not send on/off notification, only that the paddle was pressed, The hub asks the switch for status.after being told that paddle was pressed.

The use case I am talking about is when the switch is NOT load and controlling something else.

For the “normal” case with a switch/dimmer directly controlling the load, status to hub is far less important. You are there physically pushing the paddle and seeing the results.

I am not referring to ramp rate per se.

The issue is not the duration of the ramp rate per se, in this case any ramp rate greater than zero is a problem.

With any ramp rate greater than 0, if the hub asks the device for status too early, it will get too early a value.

You can see this sometimes when you dim a light to full brightness and SmartThings shows light on, but dim at for example 40% when the light is really at 100%. The hub is sent the paddle press event and requests light status, and since light has not fully ramped gets an intermediate result, for example 40% instead of 100%

This is not an issue for a switch, it is either on or off with an effective ramp rate of 0.

Yes, people always ask about ramp rate etc. but that is not what I was referring to.

I might be mistaken, but it seemed there was also an issue with making sure the hub had the correct status and adding between 100ms to 500ms or so made it work.

OK, but the hub itself doesn’t have anything to do with how a master switch controls the current to a fixture on the same circuit branch. The “instant status” is the switch reporting to the hub when something has been manually done at the wall switch, but that’s a completely separate operation from actually dimming the lights.

So I’m just a little confused about what you’re describing if you’re saying that this switch is wired in to directly control a fixture, that is it is replacing the dumb switch that was previously there, then are you saying that you are seeing a delay when you just press the rocker on that switch?

If so, my first step would be literally to unplug the hub and see if you get the same behavior.

If you do get the same behavior, it’s being caused by the ramp rate parameters that are stored in the firmware of the switch.

If the delay goes away when the hub is unplugged, on the other hand, then it’s something in the rules set up (or the associations or the DTH) for the switch.

While it may be okay to control an outlet with a switch, controlling an outlet with a dimmer is problematic.

There is nothing stopping a family member (might be the in-laws, not saying …lol :slight_smile: ) from unplugging the lamp in that outlet, even though there are plenty of other empty outlets to use and plug the vacuum cleaner into the outlet controlled by the dimmer. And then frying the dimmer switch.

So if I want to control some lights in a room, and there are no existing wired lamps, just wiring the dimmer to the outlet is not such a great idea.

Okay. I have not described my scenario well

Example; 3 ways. In my house, there are 3, 3 ways. There is a neutral in one box but not the other box of each 3 way.

So I repurposed the traveler to be neutral and use previous zwave generation linear aux switches to dim the lights.

Works great for several years.

Response is such that as you dim you cannot tell by dimming response whether you are at the physical switch or the aux switch.

I am looking “future proof” myself, if one of those aux switches fail, or for a similar scenario in which I want a dimmer switch to control other lights.

It seems that in terms of just replacing the linear switches, the new GEs will work, they can be setup to directly control the load using zwave association. And the behavior of the zwave association seems similar, as light dims status or “basic commands” or whatever is sent to the other device, this is before I have let go of the paddle.

However, it seems the switch status is NOT sent to the hub (or polled or whatever) until, and only until the paddle is let go.

The speed of communication of the hub is irrelevant. Communication time could be zero, faster than the speed of light.

This means that you have to let go of the paddle before the light has dimmed on or off, there is no feedback.

Is the 2 second “upper limit service contract” before or after the patent expired?

So with the patent, the dimmer can send its status to the hub, but must wait 2 seconds?

Right now, hub speed is irrelevant as something is broken, at least for me.

iOS SmartThings app (classic and new) is not updating status correctly.

Started last night or this morning.

Rebooted hub several times and rebuilt zwave several times.

Getting weird status delays in SmartThings app.

I’m sorry, I’m tired and I’m just not following. Especially I don’t understand at all what you’ve done with the auxiliaries.

And

A) However, it seems the switch status is NOT sent to the hub (or polled or whatever) until, and only until the paddle is let go.
.
The speed of communication of the hub is irrelevant. Communication time could be zero, faster than the speed of light.
.
B) This means that you have to let go of the paddle before the light has dimmed on or off, there is no feedback.

A) and B) nothing to do with each other if this is the master switch which controls the load. Because the hub is not involved in dimming the lights if this is the master switch which controls the load.

So anyway, I’m obviously very confused, so I’m going to leave this discussion for some of the wiring guys to look at.

Good luck! I’m sure somebody will be able to help you get it straightened out.

@jhamstead

As always thanks for feedback.

We all love solving problems! That is why we are here! Even if we sometimes may come across that we feel too strongly about a given topic. Lol :slight_smile:

1 Like

By the way, I am not saying that in the GENERAL case round trip delay to the hub is not important.

I was just trying to get the point across that since it appears in my testing that the status from this dimmer, is not sent or received or whatever mechanism, UNTIL the paddle is released, round trip delay to hub only makes this scenario worse.

This is not correct. The z wave plus switch sends unsolicited reports for on/off (binary switch report) and the dimmer reports level (SwitchMultiLevel report). There is not a two step process here. The use of unsolicited reports is sometimes described my marketing as “instant status” (in contrast to polling).

From what I understand, The older z wave (not z wave plus) dimmer required polling rather than unsolicited notifications (in part due to the Lutron patent which is now expired).

If I understand correctly, the behavior you are noticing is that while holding down the dimmer, say to dim from 10% up to 80%, the switch doesn’t send any report to the hub until the dimmer is released, at which point one report is sent with the new level.

Okay thanks for responding. Thanks for the clarification.

So, this is the behavior I think I am seeing, is there some setup I have missed?

——
Edit:

The less polling and more direct communication to hub is always appreciated.:slight_smile:

This switch does feel like hub updates status quicker with this new generation of dimmer switch than last generation of dimmer switch.

—-

Trial setup: I set up this new plus in wall dimmer to directly control a zwave plugin dimmer through zwave association; and through hub, smartlightning, mirror, a hue light.

To repeat:

  1. plugin dimmer/ zwave association
  2. hue light/ smart lighting, mirror

As I hold the dimmer and dim up the switch, the Ge plugin dimmer dims up, but the hue light does not go on until I let go of the paddle.