MDAR
(Stuart Hanlon, UK importer of Velbus hardware)
24
Hi
I haven’t packaged them up or actively promoted them yet, in fact I was only having a conversation with the chap who designed and populates the PCBs last night.
(There is a box of 80 under my desk)
What I can do…
Is give you a link to a YouTube video of them on my desk.
I’ve got some more above my desk, so now that I’ve got QLCplus up and running, I’ll try to do something funky with openHAB2.
Then work out how to demonstrate it, including linking them to Velbus button actions.
And…
I’m happy to loan you one or two to play with to see if they’ll be what you’re looking for. (Along with a simple Art-NET / sACN LAN adapter so you can try them with openHAB2 directly, or with QLCplus, remotely controlled by openHAB2.
Other than what @Dome and I have put on this forum, I’m not aware of anything.
Maybe Dome & I could put something together.
The various Thing & Item files that I uploaded will get you a way to issue remote controls to QLCplus or Jinx!
@Dome What do you think?
Should we work at putting a guide together?
I think I’ve cracked the remote control side, including sending openHAB2 RGB values, so that QLCplus can use them in functions.
(I particularly like how it converts RGB colour picker data onto something that works for RGBW lamps, that’s something that expensive software sometimes struggles with)
Those down lights are nice. Is it too much to hope that they fit into a standard down light fitting for an MR16 halogen/led bulb?
What hardware do you use to power them and control them?
Could they be used in outdoor down light fittings under eaves?
MDAR
(Stuart Hanlon, UK importer of Velbus hardware)
26
What?
As if they are mounted in a Par16 style lamp house or something?
Obviously the MR16 pin grip won’t work, but the lamp body will fit the standard space that a GU10 / MR16 based Par16 will go into.
(50mm diameter front face, roughly 47mm deep / tall)
Or we’ve created our own style of downlight housing, both of these are shown in the video.
If it’s not clear, let me know and I’ll post some more pictures or try to create a 360° view somehow.
We haven’t officially created an external grade housing, but I have filled some with silicone and mounted them into Up / Down external fittings.
I’ll get around to fixing them to the wall soon for a soak test.
Okay…
Any type of SPi driver will run them, as they are essentially SK6812 RGBW chips.
We use the Smart-Show Art-NET / sACN LAN adapters.
The concept is…
Master unit at the start of the run contains
Mains power inlet
RJ45 socket
12Vdc PSU suitable for ~30 lamps
Smart-Show LAN adapter (with 5V dropper)
3 pin outlet with 12V power and SPi data (up to 170 lamps)
The 3 pin connection is daisy chained along all the lamps, each taking 12 control channels.
If you need to inject more power, a slave PSU would contain
mains power in
3 pin inlet (that ignores the 12V+, but takes the remaining data and the 0V reference)
3 pin outlet
PSU for ~30 lamps
These lamps aren’t the brightness, because they were designed to be used in large numbers, to increase the flexibility.
We’re toying with using ultra bright addressable chips, but that’s still on the drawing board.
At the moment, these lamps need to have a retail price of about £20, until we can get the PCBs and lamp houses made in large numbers.
Our original design concept was to install them at ~700mm intervals in ceilings.
When I get to refurbishing another room in my house, I intend to install them and create some videos.
These sound pretty awesome. I’m planing a family room renovation and I think these would work really well. I also could use them in my bathrooms.
1 Like
MDAR
(Stuart Hanlon, UK importer of Velbus hardware)
29
Hi
If anyone is interested, I’ve worked out how to run QLCplus on 64 bit Debian Linux, in an X session with a VNC remote server.
(I.E. No desktop manager)
I’m trying to connect OpenHab2 to QLC+ using the DMX binding as described above. For now I just want to click a button in OpenHab and turn on a DMX RGBW spotlight that is connected to QLC+ (tested and working from QLC+).
QLC+ is installed on a RPi at 192.168.171.21 with bitwizard DMX interface
OpenHab is installed on Windows 10 at 192.167.171.146
I used the DMX Binding to created a sACN/E1.31 Bridge and set the Receiver Address to 192.168.171.21 and DMX Universe to 1. The Status says ONLINE.
I created a DMX Color (RGB) Dimmer thing and set DMX Channel Configuration to “1,2,3”, fade and dim times to 1000.
I’m not sure what to do at the QLC+ end.
From the config in the web page at http://192.168.171.21:9999/config I set the output for universe 1 to “UART 1"ttyAMA0” (the bitwizard DMX interface) and I can control the lights from simple desk no problem.
I don’t know how to set the input to the OpenHab binding. The drop down shows [E1.31] options for 127.0.0.1 and 192.168.171.21. I tried “[E1.31] 192.168.171.21” but tinkering with the dimmer does not make the lights go.
How do I make the connection? Will it work if OpenHab and QLC are on different machines?
Thanks!
Phil
MDAR
(Stuart Hanlon, UK importer of Velbus hardware)
31
Hey Phil
Great to hear you are making progress.
It’s a little tricky to get your head around to start with, but once you get the drop of it, you’ll find it’s quite straight forward.
Whether you go down the DMX / Art-Net / sACN route for remote control, or you take @Dome approach with WebSockets.
The trick is to get the Virtual Console up and running in QLCplus, then control the elements of that with openHAB2.
For example.
Let’s say you have an Animation widget running in the Virtual Console.
This haa a fader that controls the intensity of the effect (or whatever you have it set to) - That will need an input DMX channel assigned to it so that openHAB2 can influence it.
Here you can see that it’s using QLCplus universe 4 as an Input, which is an sACN / E1.31 stream, and using channel 2
You might find it easier to assign a single channel DMX dimmer (for each channel) in openHAB2 if you want to use the Auto Detect option, as QLCplus won’t be able to tell you which of the three channels to assign to each element. (then delete it / them later, FYI you can create them and use the same channels as the RGB Dimmer, it won’t do any harm)
After that, all you need to do is put QLCplus into “operate” mode and openHAB2 should be able to control the effect.
You can also add remote control to Presets in an animation widget too, which can be fun.
I do intend to make a video of how this is done, once I get some time.
(I’ve got a couple of projects on the go that are demanding my time)
Good luck
Stuart
FYI
These are the openHAB2 Things and Item files I’m working with
Thanks so much for all the detail Stuart.
I’ll start with a simple button that runs a chaser and work my way up to matrix animations.
On the virtual console in QLC+ from my laptop, I’ve defined a button and set the input like this:
I load the project via the web interface in QLC+ on the RPi and then set the output for universe 1 to UART and the input for universe 4 to E1.31 for my RPi. When I click the button my lights go.
In OpenHab I have a DMX Chaser thing with these settings:
When I click on the Chaser thing the Status is OFFLINE - CONFIGURATION_ERROR, with this error: chaser configuration malformed.
Any ideas what I’ve got wrong?
Thanks,
Phil
MDAR
(Stuart Hanlon, UK importer of Velbus hardware)
33
You need to access the full Input Output section of QLCplus UI (I use a VNC session, check out the QLCplus forum to see how I’ve done that) to setup the universes, because you’ll need to set the incoming universe of sACN.
(I’m not sure what the purpose of the web interface universe setup, it’s lacking something)
That’s exactly what it means.
The chaser isn’t set correctly.
Or…
Try a different version of the DMX binding.
It shouldn’t make any difference, although I’ve never run QLCplus on the same machine as openHAB2.
I’d dedicate a Pi just to QLC+. I currently use a Pi 4 (4 Gb) for light browsing and QLC+. Sometimes when I interact with a web page on the Pi I get weird behavior from QLC+. It jumps to a different scene without being commanded to. I’m going to spin up a Pi 3 with the custom QLC+ image and see if that’s enough horsepower.
When editing the universes via RPi directly, I can untick Multicast and the IP address changes to 192.168.171.21 (for unicast I assume), the port = 5568, E1.31 Universe = 4 and transmission mode and priority are blank.
I’ve tried a couple of combinations:
Universe 1 output = UART and universe 4 input = E1.31 in QLC+, and the sACN/E1.31 Bridge in OpenHAB set to 192.168.171.21 and universe 4
Universe 4 output = UART and universe 4 input = E1.31 in QLC+, and the sACN/E1.31 Bridge in OpenHAB set to 192.168.171.21 and universe 4
In OpenHAB both combinations show the bridge as ONLINE.
I created a single dimmer set to use the bridge and DMX channel 1 (also tried 2 or 3 or 1/512).
None of these combinations turn a light on when I click the dimmer buttons in OpenHAB but the lights always work from simple desk in QLC+.
Any ideas what I’ve got wrong?
Thanks,
Phil
MDAR
(Stuart Hanlon, UK importer of Velbus hardware)
36
Okay…
That looks about right.
You can completely ignore the Web interface DMX config.
I’m not entirely sure why it’s there, because it’s missing that vital ability to select the exact universe within the NIC. (You can only select pre-configured universes, which are ONLY available via the X session. Have you added the Tiger-Scrapping-VNC-Server to your Pi? https://www.qlcplus.org/forum/viewtopic.php?f=30&t=14384
)
This shows my Input (of QLCplus) to the 192.168.178.24 (It’s own address) as sACN, and listen to the transmitted universe 1, and mapping it to QLCplus (internal of the software) Universe 4
Meaning, openHAB2 is unicasting sACN to the IP address on QLCplus machine, using sACN universe 1.
QLCplus’ internal Universe 4 is mapped to listen to the NIC for Unicasted sACN, specifically sACN Universe 1.
{So using Universe 4 of QLCplus, is not the same as Universe 4 of the sACN Stream, unless it is mapped that way}
FYI
sACN supports >60,000 universes, while Art-Net supports 16 x 16 x 16 universes (4096 universes)
Whereas, this web interface Configuration doesn’t allow for selecting the sACN universe that is available via the selected NIC / Port / Device
Have you setup a Function in QLCplus to control your lights?
Then, have you linked a Channel of the universe coming out of openHAB2 to one of the effects / sliders on the Virtual Console in QLCplus? (as per the pictures I have posted?)
IE ~~
A Function in QLCplus sends a pattern to your lights
The function is controlled in QLCplus via the Virtual Console (Ether a simple slider or Animation widget)
Single channel of the outgoing openHAB2 sACN is then mapped to the Slider within the Virtual Console.
The advantage of this is that you can see the element of the Virtual console following the openHAB2 universe.
Thanks so much for your patience. It’s ugly, but it’s working.
Being new to DMX/QLC+ as well as OpenHAB and connecting OpenHAB to QLC+, I was confused by all the detail and settings in your animation example, partly because I didn’t have a sequence or matrix to connect the animation to.
Hopefully this simplified description will help another noob…
As I understand it, the closest thing to a button for turning on a DMX light switch in OpenHAB is a DMX Dimmer.
As you said, the first bit I missed was that the sACN input universe (4) in QLC+, should have the E1.31 universe set to the universe (1) used by the sACN bridge in the OpenHAB DMX binding. Makes sense now that you have to tell each of them how to point to the other system.
I set up a simple slider in the QLC virtual console and set it’s Input universe to the QLC sACN input universe (4: E1.31), gave it a channel (2) and then set up a DMX Dimmer thing with the same DMX Channel Configuration (2) in OpenHAB.
Setting a function in the Playback tab for the slider in QLC+ makes the light go.
Now to see how these web sockets work…
Thanks again,
Phil
1 Like
MDAR
(Stuart Hanlon, UK importer of Velbus hardware)
38
Hi
I’ve just taken a closer look at the configuration for that chaser.
Try something like this instead, to fade a single / all channel.
Wanting to follow Dome’s node red instructions and I tried to install node red on my RPi (it’s the QLC+ package with mosquito added) and it failed due to disk filling up. I tried a second time without the Pi-specific nodes and got the same result.
Does node red have to be on the same machine as QLC+ (my RPi) or would it be fine on the OpenHAB machine? (Currently windows 10)
I’d recommend putting QLC on its own Pi. NodeRED should be fine on your Windows machine alongside openHAB, although I’ve personally only run it under Linux.
What size SD card are you using? Is it potentially a corrupted card? SD card issues are common with Raspberry Pis and can be minimized by getting larger, quality cards.
I just spun up the QLC+ RPi image on a Raspberry Pi 3. So far, so good, though I haven’t had a chance to really stress it yet.
I’m running the full version of QLC+ on my desktop Mac, and save the config files on this computer using the full version. This works great as I keep the Pi fully headless.
When I want to open this full version, I go into the config manager of the QLC+ Pi (at qlc.local:9999 on my local network) and open a blank config to keep the two instances from interfering with each other. Then, I open the config file using the full version and make my tweaks. When I’m done, I then reload the config file and go into the system tab to apply it as the startup default.
Thanks Dome.
I switched the card from a 16GB Sandisk Ultra to 64 GB Sandisk High Endurance and put the same image on it, then ran df -h and realised it was only allocating 1.7G to /dev/root.
My windows brain struggle with allocating more disk space until I found FAQ #3 in Massimo’s guide, which gives instructions to allocate the rest of the disk space:
sudo raspi-config
7 Advanced Options
Expand Filesystem