Future of lighting (residential)

[John B]

So, who else plays around with lighting in the house? One thing that I am constantly experimenting with in interior decor stuff is modernised lighting. To get a perspective, my house was built in the 70's and had the typical design idea: 1 or 2 ceiling mounted bulbs, incandescent, powered by mains from a switch.

This design philosophy in its original form has a few drawbacks. The lighting pattern creates bright spots and shadows in the room, and the incandescent bulbs were/are inefficient.

Modern CFL and LED retrofits seem to be a bandaid solution. They don't solve the lighting pattern problem, and concentrate heat in a small area where there is both the light source, as well as the driving circuitry. Consequently they are built to be disposable, as no matter what fails, the whole package is discarded into the landfill.

So, if you were to design a lighting system from scratch, what would be your philosophy? For me there would be many individual sources of light in the room, probably LED. This would give an even light pattern and they could be run at very low current to remain cool. The power supply would be a separate unit so you could optimise heat dissipation as well as include additional measures such as power factor correction. Switches in rooms would be more like controllers, sending low voltage control signals to the power supply. Alternate switches could easily be interfaced, such as motion sensors or timers. It would also get rid of those meddlesome switch mounted dimmer units, as dimming can all be handled by the power supply.

Also, with the addition of a battery, you could have emergency low level low power lighting if the power goes out.

Since the power supply would not be considered a disposable item, all components could be built to a higher quality and made to last. The LEDs themselves pose a problem with packaging. Do you have plug and play replacements, or perhaps replaceable heatsinks where an old/malfunctioning LED chip can be desoldered and reinstalled? Seems more responsible to only throw out the minimum.

Thus far its been more practical to just make freestanding custom lighting in the room. I'm almost certain our building codes forbid most things I've suggested.

[Brumby]

I don't think the building codes are as restrictive as you think on this front.

The only reservation I would have about putting in your own system of low voltage lighting would be the possible requirement of the cabling being done by someone with a licence to do so - and that does not necessarily mean an electrician.  There is a credential available for cablers running ethernet, for example, where they are trained to do so safely.  While this may sound OTT, it is meant to ensure safety when this cabling is run in areas that also have mains or other cabling of concern.

Aside from that, as far as I am aware, there is a lot that you can do.

[IanB]

You did not in your post seem to make any mention of aesthetics or decor? In so omitting them, I feel you have excluded at least 50% of the population (the female half). Lighting for the most part is decorative, not functional. In order to make your thesis work, I think you need to describe how to make low voltage LED lighting decorative and attractive.

[John B]

You can go nuts on the subjective artistic side, there's not much to argue about (at least here, no guarantees with your other half!). For me at least the considerations are more on the practical side. An even glow makes working much easier on the eyes, particularly when you're looking at a computer screen. The other issue for me is EMI. Most of my lighting is in a music studio, and so removing sources of EMI are a concern. Once I tracked down a CFL bulb in another room that was causing buzzing through my musical instruments' magnetic pickups. Also, I no longer build PWM controllers for the same reason. My last design was a linear one in fact, as it sits right next to audio equipment. It only has to drop around 1V @ 1.5A so the heat dissipation is minimal, and it's whisper quiet (both audible and radiated EMI, unlike PWM solutions). I'm currently working on a few switchmode designs too.

[cdev]

Residential lighting can also mean intimate lighting..

Liquids create an inherent sense of relaxation. Nothing is as relaxing as floating in water so if you can simulate that effect with lighting, people will relax.

If you look at some of the things people have done in hotels around the world, I have seen some very cool effects.

[steve30]

I like to play with different types of lighting.

I've attached a couple of pictures or a lighting experiment from a few years ago. I stuck a couple of 5" fluorescent tubes along two of the walls. That produces a very nicely diffused light throughout the whole room. As I could control what specific tubes I had, I used some with nice colour temperatures and good CRIs. I also used a dimmable electronic ballast to vary the brightness.

The particular setup in the photos is very extremely crude, but it lit the room up nicely.

I wouldn't mind playing with some metal halide lights some time.

[Corporate666]

This is my industry - I design and manufacture residential and commercial LED lighting.  The stuff I make is on the high end (it's the only way to make it succeed building products in the USA).  I also do a lot of one-off projects.

I completely agree that the existing solutions are complete band-aids.  There aren't many applications less tailored to the nature of LED lighting than existing home lighting.  What I mean is... you are starting with AC, LED's are DC.  You usually have central fixtures, LED's produce less light so you need many of them, ideally spread around.  The AC dimming solutions are sub-optimal (at best) for LED's.  And the thermal problems are tricky - bulbs take heat fine, so one big light high up in a room is no problem.  Not to mention the customer mindset is oriented towards bulbs... even when people are sold on the idea of LED's, they still think of light fixtures as being a single unit with a light source, a shade and a housing attached to the shade.  LED's open up so many opportunities for lighting that it's hard to educate the customer on what is possible.

Having said that, I disagree that LED's could or would be something you install once and they could last for decades.  From an electrical/performance standpoint, that's certainly possible.  But LED technology is changing rapidly.  CRI is getting better.  Thermal performance is getting way better. Lumens/watt is getting way better.  These advances open up new possibilities for the lighting designer, so things that were not possible 5 years ago can be possible now.  And as someone mentioned above, aesthetics is at *least* 50% of the equation.  If you look at a modern style home from the 70's, it looks so outdated today.  An 80's modern home looks dated too - just in a different way.  Same with a 90's modern home.  Lighting is probably the single biggest contributor to style - because it has a style of design as well as a style of performance.  In other words, you can really set the mood and style of a room depending on how you light it.  But what the latest-and-greatest is in 1997 is radically different than what it is in 2017.   

Someone above mentioned restrictiveness and rules... I personally feel it's highly restrictive, certainly compared to other markers I work in.  For automotive, there aren't many rules (at least in the USA), and what rules there are operate on a self-certification process.  So anyone can make anything for automotive use without onerous rules and testing requirements.  But for lighting, your product *must* be UL listed.  That costs a minimum of $10k for a single light fixture that uses off-the-shelf parts (off the shelf LED driver from Meanwell/GE/whoever) and an off-the-shelf light engine from Cree or similar.  If you want to design your own driver, you're looking at about $20k additional costs for certification.  And only that specific implementation is certified... if you take the same driver and light engine (which you already paid to be certified) and put them in a different light fixture (say, one made of bamboo and acrylic instead of steel and glass), you have to pay again to get that specific fixture certified.  And these are the costs when you are using LED's and it falls under the low-voltage directive.  It's more if you were making bulb lights!  If you are not UL/CE/whatever certified, no shop will carry your products, no home inspector will sign off on an installation, no insurance company will cover you, no electrician will install them, etc.  It's tantamount to a total blacklist from the industry if you aren't certified.

I type too fast for my own good - but the last thing I'd add is to answer the original question, the ideal solution would be if people were willing to pay more for lighting .  AC isn't actually too bad to work with, because it's easy to shift voltage with a transformer.  If you started with DC, you'd want it at a high enough level to eliminate the need for boost converters... so would 48V even be sufficient?  I am not sure it would be for someone who wanted to make a big light fixture with many LED's.  But even starting at 48V, if you want to use a single high power LED, you're seriously limited in the number of buck converters that will accept 48V on the input.  So cost is already a problem - may as well have just started with AC.  On the control, DMX, 0-10v, DSI, DALI and custom solutions all have their limitations and none are really perfectly suited to home/commercial lighting.  And now with WiFi and (more recently) BlueTooth and BLE, traditional dimming techniques are very much losing their monopoly on the market.  Which is another reason I wouldn't want a fixed system that is intended to last for decades... because I'll wager than even in 10 years, residential LED lighting will be a lot different than it is today.

[Brutte]

So, if you were to design a lighting system from scratch, what would be your philosophy?

I have been visiting TU/e (Eindhoven, Netherlands) several years ago and there is a whole department that tries to give answer to precisely this question. Most likely because Philips has its headquarters in Eindhoven.

I would not reinvent the wheel but try to follow those that are engaged in such development. These guys are light years ahead of your most demanding dreams.

[madires]

I've built some nice lighting using aluminium U channels and LED strips. You can also add cheap remote controlled dimmers or go RGB. The channels can be mounted everywhere and covered with clear or opaque acrylic glass.

[tszaboo]

Or maybe not...
In all seriousness, I believe future lighting will be some kind of "full top surface" illumination, with changeable color temperature. To simulate natural light. I also think, it will be a built in IR heat source, with smartness to follow you around.

[madires]

Of course, you can also overdo things.

[Red Squirrel]

I think LED strips will probably get more popular over time, you could do some pretty cool lighting effects with those in the right places.  RGB ones could be fun too for setting different themes like Christmas.   

[cdev]

I use LEDs at home a lot but unfortunately its still mostly in the form of screw in replacement bulbs.

I would love to use colored LEDs year-round in some settings in my home but fading over time and limited replace-ability of especially the blue component of RGB elements, which in my experience tend to fade faster,  is an issue that I wonder about.

If the blue LEDs were dimmed a bit - run a bit less brightly, or more care was used to heatsink them better, it likely would be less of an issue for me at least.

Bright blue LEDs sometimes hurt my eyes anyway.

[cdev]

As a society, what would be a way of lowering this cost of entry issue without sacrificing on safety?  Because I can see this frustrating almost all potential lighting designers's hopes to sell any innovative products completely.

Someone above mentioned restrictiveness and rules... I personally feel it's highly restrictive, certainly compared to other markers I work in.  For automotive, there aren't many rules (at least in the USA), and what rules there are operate on a self-certification process.  So anyone can make anything for automotive use without onerous rules and testing requirements.  But for lighting, your product *must* be UL listed.  That costs a minimum of $10k for a single light fixture that uses off-the-shelf parts (off the shelf LED driver from Meanwell/GE/whoever) and an off-the-shelf light engine from Cree or similar.  If you want to design your own driver, you're looking at about $20k additional costs for certification.  And only that specific implementation is certified... if you take the same driver and light engine (which you already paid to be certified) and put them in a different light fixture (say, one made of bamboo and acrylic instead of steel and glass), you have to pay again to get that specific fixture certified.  And these are the costs when you are using LED's and it falls under the low-voltage directive.  It's more if you were making bulb lights!  If you are not UL/CE/whatever certified, no shop will carry your products, no home inspector will sign off on an installation, no insurance company will cover you, no electrician will install them, etc.  It's tantamount to a total blacklist from the industry if you aren't certified.

[John B]

Or maybe not...

I like that, but why not just put it on the ceiling? It sounds like solar freakin' roadways all over again, except now its illuminated freakin' floorways. Don't spill your drink or bring dirty shoes into the house.

[blackbird]

Or maybe not...
In all seriousness, I believe future lighting will be some kind of "full top surface" illumination, with changeable color temperature. To simulate natural light. I also think, it will be a built in IR heat source, with smartness to follow you around.

Well, this could be the future. After all, earlier in this movie they showed those flat interactive display thingies we all use today. ;-)

[IanMacdonald]

The local uni put lights in the pavements. They are terrible, as they just blind you. Whole floor lighting is probably less of a glare but still not the best idea.

One issue with LEDs is that some flicker very badly, far worse than CFLs in fact. Price isn't always a guide on this either.

Though, one of the worst lighting developments IMHO, is the ceiling spot fad. All they do is put circles of over-bright light on the carpet, interspersed with gloom. Keeps the plasterers in business repairing the numerous holes, I guess.

[tszaboo]

I like that, but why not just put it on the ceiling? It sounds like solar freakin' roadways all over again, except now its illuminated freakin' floorways. Don't spill your drink or bring dirty shoes into the house.

It's from the movie 2001: A Space Odyssey. The film is from 1968, the title is 2001. It supposed to be a strange imaginary place.

I agree, on the ceiling it makes total sense. I hate glare. Take for example museums. It seems to be absolutely impossible to find one, which doesnt have glaring lights, which blind you. Its like everywhere you look, there is a glaring light.

I always say, this is the best way to describe the 21 century for future generations:

We had excellent lights and air conditioning, but nobody knew how to use it properly.
--NANDBlog

One issue with LEDs is that some flicker very badly, far worse than CFLs in fact. Price isn't always a guide on this either.

Flickering LED lights. I hate them. If they installed in a room, i literally get sick if I'm in that room for more than 10 minutes. And the owner doesn't even notice it.