Rotary dimmer switches

A client has indicated that they don't like pushbutton dimming, and would prefer a rotary dimmer switch.

Am I right that a rotary switch suited to 1-10v dimming would also be suitable for use with a general input unit (E5504GI)? (I understand (having never used the GI unit before) that I will need a PAC to map the values generated to group addresses for dimming?)

Alternative/better solutions welcomed!

 

The 5504GI will take a 0-10V input directly and you can create a custom curve (or in this case a line) to map 1-10V to 0-100% on the GA. No logic unit required.

 

Perfect, that's all I can hope for I reckon. Nick's point about slow updates via the PAC is a good one so I'd hope to avoid that.

I have ordered a 5504GI to test.

 

Hi more-solutions
I'm going to add a comment which may or may not be helpful.
The customer has expressed a desire for rotary dimming.
I'm not surprised.
Analogue controls are nice.
But is that really worth the trouble for you when push button dimming is the default and voice control is starting to become possible for CBus?
Do they really want to pay extra for a dimmer knob?
The analogue control is not something which will work nice or easily with anything remotely operated.
My suggestion is that you talk them out of it for your own sake.
I'd take voice controlled dimming over push button or dial any day.

 

In this application (it's not domestic) stuff like that is highly unlikely and would involve a lot more than just changing a couple of things. The requirements are that it will replace existing switches (which are currently retractive switches via an aux input) with minimal cable changes, so if a rotary switch works then it's no "worse" than what's there at the moment. There are other reasons they've given that kind of make sense in the context of the application but the main one is probably that they don't like (aka are bored with) what they have and want a change.

But the comments and suggestions are welcome nonetheless.

 

Only you can make the call if it is worth meeting the customer requirements.
I'd be trying to talk them out of it.
For your sake.

 

Well this is a trial anyway, so if it is a bad idea then the best way to talk them out of it will be to proceed with the trial!

 

Or maybe sell them an ELDT so you don't have to reinvent the wheel
As long as you are making money either way.
I think there is a definite disadvantage for you to sell to them analogue control.

 

You are missing the point guys - we have someone willing to try something out of the box and share the outcomes (hopefully) and we will all learn from it. Experience says it wont be a great solution, but it may yield some results that do lead to a solutions that is acceptable - it may inspire someone to come up with a solution.

Lets support this experiment as much as possible and see what comes of it. I the customer is willing to trial it and deal with the potential issues then its a win for us - not many would put up with the issues of a "non performing" system

 

Fair call Mick
I'd like to see a rotary switch for ceiling fan control just because it is easy to operate.
No argument from me a dial for dimming would be easy to use too.
But I think there is a definite disadvantage adding analogue control on a system like CBus where you have all sorts of remote control options.

 

It's got me thinking now. May just have to hook up a rotary encoder to an Arduino and Serial Interface Module and see how it goes. All the benefits of a dial without the inherent problems.

Or maybe a ESP32 wifi module that talks directly to a CNI. Hmmmmmm......

 

I was just thinking the same.
Encoder type device would be a good fit in the digital world

 

Voice control trumps it though in my opinion.

 

Personally I find voice control far too primitive at this time. Apart from the fact it gets it wrong so often, you have unacceptable delays, with a fixed mike like Google Home any background noise upsets it, and the idea of having to pick up a phone and turn it on to use it is just too annoying. And in the middle of a movie, do you really want people talking to your house. Not for me thanks. A dedicated remote control that is always on is simple, reliable, and anyone can just pick it up and use it.

 

@Ashley
I'm on the Apple platform with Homebridge.
I'm finding it astounding.
For a number of reasons which I wont explain right now I'd rather give CBus control of any smart things.
For voice control it's bloody awesome.

 

Getting back to the original question:

I have obtained a 1-10v dimmer switch (Aurora AU-DSPLED), and I'm trying to configure it with the 5504GI.

I'm not that familiar with the inner workings of 1-10v, but I understood that at its core the switch is just a potentiometer?

Wiring the switch as a resistor and configured using the 10k setting in TK I have working dimming (set to 100ms sample rate it's responsive enough) but I'm only getting dimming in the first (maybe) 50% of the dimmer range, at which point it hits 100% (255 shown in TK current status).

Any suggestions for what I should be doing here? It feels like a simple problem that I'm finding way too hard, maybe that's just a Monday thing..

Also: Using the switch I'll never be able to switch to off. This rotary switch as a push on/off that I could either wire the dimmer output through (that'll be messy in this environment), or maybe bring into a auxiliary input unit to control that way, but better suggestions welcomed. (I presume if I use a switch without the push on/off function I could set a suitable dimming curve so that the lower end of the range goes off.)

 

I like the idea of an encoder - but as a tangent to that, perhaps one with BCD or Grey code output straight onto a CBUS coupler?
Little bit of script to decode the 4 inputs to a percentage output.
Inputs -- Output %
0000 0%
1000 10%
0100 20 %
0010 30 %
0001 40 %
1100 50%
1111 100%

 

How does the dimmer currently work? Is it writing to the bus continuously while the button is being pressed?

I would look at replicating this - although a rotary know would be much faster than what we currently have.

10% increments may be too much, and it if it LED then the lower 20% is non existent. A curve would be better suited.

 

Agreed Mick - and I would speculate that the difference between 90% and 100% would be negligible also so have the light start at 30% which would leave 9 increments from there to 100%. Might have to test this one out for myself out of pure curiosity.
Edit: My memory of BCD was a little off. Truth table could be done like this:

Code:

Input -- Output
0000     0%
1000     25%
0100     30%
1100     35%
0010     40%
1010     45%
0110     50%
1110     55%
0001     60%
1001    65%
0101     70%
1101     75%
0011     80%
1011     85%
0111     90%
1111     100%