Posts: 2,339
Threads: 104
Joined: Apr 2009
So this is RGBW so we should be able to dim color and White, right?
Yeah, I had to flip my wiring on my unit to work properly for red and green.
Posts: 40,483
Threads: 491
Joined: Aug 2002
06-01-2017, 10:31 AM
(This post was last modified: 06-01-2017, 10:32 AM by Dean Roddey.)
I'm not sure what the White bit would do, or how one would get to it via the Z-Wave interfaces it supports. If you just set the HSV values properly you'll get white. Just lower the Saturation component to zero and it effectively becomes white. And you can dim it via the Value component (or using the regular dimmer field.)
Dean Roddey
Explorans limites defectum
Posts: 2,339
Threads: 104
Joined: Apr 2009
Sorry, I was not clear, for RGBW the White is handled separately.  You can also generate "white" with the RGB but on an RGBW LED strip the RGB is typically 1 LED and the W or WW (warm white) is a separate LED
If not done properly the W can turn on unintentionally along with the color. Essentially each channel should be explicitly set separately for the desired result. FWIW here is a link to a version of the Fibaro RGBW controller for SmartThings.
Posts: 40,483
Threads: 491
Joined: Aug 2002
The SmartThings stuff is mostly Greek to me. I'm not sure what channels you are referring to? The Z-Wave color interface provides access to separate color components, and I'm setting the red, green and blue separately. It also provides access to a 'warm white' and 'cold white' component, so perhaps the white LED is one of those. I can try it and see.
Still, we have to be able to control it via the RGB components or there's no way to make it work with the V2 colored light device class. I can provide separate switch/dimmer fields for the white I guess, which just look like a separate light. Effectively it sort of is anyway it looks like.
As usual with this Z-Wave stuff, it's all so ad hoc and inconsistent that it's difficult to make it work within a generic framework. At some point I may have to just give up on even trying to do that and just write separate code for every module type and be done with it.
Dean Roddey
Explorans limites defectum
Posts: 40,483
Threads: 491
Joined: Aug 2002
OK, I just did a quick test and set the color values to all off, and set the 'warm white' to full on and I have all just white LEDs. So I'll make that look like a separate dimmer field.
Dean Roddey
Explorans limites defectum
Posts: 2,339
Threads: 104
Joined: Apr 2009
Sounds good Dean, as long as there are separate controls are there then it should be good. So that would include color intensity as well as brightness for the white, correct?
Try testing unit on and off (if there is such a thing) with the white off and then with it on along with some blended color like purple or something. If white is off it should remain off when the colors are turned on.
I'm traveling this weekend but if you have something by Sunday I can test, I've already got mine configured to the VRC0P. Controller manages my deck rail lights. I'll post some pictures when I get a chance.
Posts: 40,483
Threads: 491
Joined: Aug 2002
I'm really struggling to understand this thing in terms of how it reports changes. It makes no sense. The manual is pretty poor, and really doesn't tell you much. There are configuration parameters related to these things but they seem to do nothing.
I get three separate multi-level switch reports, which presumably are the three color levels, but there's nothing to distinguish which is which. I'm configuring it to send me color command class reports, which I could make sense of, but it doesn't seem to care to do that.
There are five association groups. The fifth would seem to be the correct one for the VRC0P, but it's set in the 4th and I can't seem to remove it. And teh auto-config is setting it to set it into the 5th but it's being ignored. The configuration parameters that control what type of reports are sent say nothing about what association groups they are relevant to.
Dean Roddey
Explorans limites defectum
Posts: 745
Threads: 33
Joined: Oct 2008
They seem to have robust support channels. Would it make it easier to send off a few questions to them about the Zwave issues?
https://www.fibaro.com/en/support/
support@fibaro.com
https://developer.fibaro.com/
_______________
Denon 3808ci, 2112ci ,Sonos, NoVo Grand Concerto, Z-Wave(Lights,Locks), Hue, SmartThings,
iPads,Tivo,Hikvision,Elk-M1,TED5000,Somfy RTS blinds+ZRTSI, Amazon Echos+Dots, Polk XRT12,
Honeywell Wi-Fi 9000, Caleo Wi-Fi Thermostats, Rainmachine
Posts: 40,483
Threads: 491
Joined: Aug 2002
I semi-sorta figured out the issue with the reporting. If you query a color component (you can only query one at a time) it sends a color report, which just has that one component. If there is a change to the color from the outside, it sends a color set command instead and I wasn't looking for those. And it sends multiple color components all at once.
But, it only reported them for a while and then it stopped again. I've power cycled it and it still doesn't. The only way I can change the colors externally is by just pressing the button, which toggles the color temporarily. For a while it reported those via the multi-value set command, but it's stopped doing it again.
It also doesn't send any reports if the change is made via the Z-Wave interface, which means we'd never see any changes made by another controller or software. If so, then the association based reporting is really utterly useless.
Dean Roddey
Explorans limites defectum
Posts: 40,483
Threads: 491
Joined: Aug 2002
Well, I really don't know what to make of this thing. It just doesn't seem to be right and I'm spending way too much time on this. I would just say screw it and provide some non-V2 write only fields for it, but those fields are created by the (newly added) color switch ZW command class, and it has to create V2 compatible fields since this is a V2 driver and that ZW class maps to the CQC Colored Light device class which requires V2 fields.
I'm going to get back to other things for a bit and see if any new information comes to light.
Dean Roddey
Explorans limites defectum
|