Just a thought... POE Lighting?

Generally speaking, PoE is a terrible topology for lighting. It would be moving a basis of system design from daisy chaining fixtures to a home-run-per-fixture cabling scheme. There are also problems with melting cables when you bundle lots of PoE cables together that this last cycle of the NEC has begun to address -- the kind of bundling you might see out to a lighting batten. It's not bad for decor lighting and certain architectural products but it doesn't scale up well to the be the principle form of lighting for a theater.

I think we're more likely to see a homogenized cable standard with power and data that falls somewhere between today's scroller cables and Lex's new hybrid PowerCon/DMX cables. Preferably with a sister standard that allows pass-through power with built-in per fixture power relays so we can stop dropping these large DMX relay panels into every project that serve little purpose beyond extending fixture longevity because manufacturers don't care to figure this out.

Whether that standard incorporates DMX, sACN, honest ACN, or whatever leapfrogs ACN for taking so long to come to market is up to probably ESTA and three hundred manufacturers to argue about before having mixed success bringing it to market.

What glue does the stage bring to make POE a success story here?

Common use of Cat5 to lighting positions?
Widespread installed base of POE ethernet switches?
A desire to accept less interesting and less bright lights for the simplicity of a single cable?
Availability of 50v LED tape?
Electrical/fire rules that encourage the use of data cable for power in public spaces?
Opportunity to replace many lights and controllers all-at-once?
Never needing to deal with visiting staff or rental gear?

Stage lighting is the last bastion of RS485 because the circumstances are so stacked against modernisation.

Lyle Williams said:

Stage lighting is the last bastion of RS485 because the circumstances are so stacked against modernisation.

Click to expand...

RS-485 is common for architectural lighting, too!

Lots of building control was RS485, but that changed starting about ten years ago. Architectural lighting applications are changing, and the technology has a greater chance of changing because buildings get built, or refurbished, as large one-off projects. There aren't interoperability needs outside what the architect has specified.

Lyle Williams said:

Architectural lighting applications are changing, and the technology has a greater chance of changing because buildings get built, or refurbished, as large one-off projects.

Click to expand...

This is becoming obvious in larger architectural retrofits where DMX as we currently know it is being pushed to the brink of death. Some feature that was installed 10 years ago with only DMX cables out to it that spans the entire facade of the skyscraper is being replaced with something that needs more universes of control for increased pixel, fixture, or parameter density. Of course, adding CATx or fiber cables to the exterior face of the building is insurmountable and grossly expensive.

Wouldn't you love to be the guy dangling from a climber's rope when it comes time to add some DMX universes to the Burj Khalifa in a few years for the next big architectural lighting widget?

MNicolai said:

This is becoming obvious in larger architectural retrofits where DMX as we currently know it is being pushed to the brink of death. Some feature that was installed 10 years ago with only DMX cables out to it that spans the entire facade of the skyscraper is being replaced with something that needs more universes of control for increased pixel, fixture, or parameter density. Of course, adding CATx or fiber cables to the exterior face of the building is insurmountable and grossly expensive.

Wouldn't you love to be the guy dangling from a climber's rope when it comes time to add some DMX universes to the Burj Khalifa in a few years for the next big architectural lighting widget?

Click to expand...

Rope Access Technicians, RAT's, DESPISE being referred to as "Rats"
Toodleoo!
Ron Hebbard.

High-total-power switches/injectors aren't that hard to come by or -- on stage lighting price scales -- especially expensive.

Anyone who was going to sell luminaires that could run off PoE on the sACN cabling input would certainly have China building them a switch that had enough power to drive every output at full power.

And yes, I also know of no special reason you can't daisychain them with a copper-thru jack.

Jay Ashworth said:

And yes, I also know of no special reason you can't daisychain them with a copper-thru jack.

Click to expand...

An ETC Colorsource Spot has 28A of inrush at 120V. Not sure how that corresponds to 48V for PoE, but each fixture would draw 3.1A at 48V -- granted, that current is split across multiple 23AWG pairs, but it still produces heat. Trying to power just one average fixture on your cable will warm it up beyond acceptable levels. Bundle a few cables together and you'll melt them in no time.

It's *that* high? Damn.

My suspicion is, still, though, that there are lots of places in a theatre that *aren't* the electrics or FOH bridge, where it would be useful to be able to run a single Cat cable to light up a few small fixtures or LED tape or what have you; everything doesn't need to (be able to) run the same way.

Also, you're suggesting that the 5 or 6 I can chain together on powercons on a single 20A are going to have almost 150A inrush; that's going to pop breakers on *real* power cabling, no?

This is a good article by ETC describing the problem and how they limit the number of devices on a circuit to accommodate inrush. May not pop the breaker on a given circuit but if you turned on a loaded up IQ48 relay panel in one fell swoop you could send a whole lot of inrush upstream to the transformer. That's why it's good form on sizable deployments to sequence the startup process just like you would striking arc's on a bunch of movers. I just finished an arena with 94 active speaker cabinets and spec'd sequenced panelboards to avoid 1880 amps of inrush upstream -- even with soft-start speaker cabinets (~20A/cabinet inrush).

As always, power efficiency will increase over time as will brightness and the ability to soft-start without significant inrush, so it's not outside of the realm of possibility that this will happen eventually in some form. I'd expect to see a wireless mesh control topology take over first though before the world goes through another reinvention of light sources.

RS-485 and other serial data communications are still being used in hobby electronics at least. I see the like of Ben Heck and others touching on it occasionally in their projects. Serial data is dead stupid simple and reliable when bandwidth isn't a huge factor.

So what I've seen is that this conversation has been directed to Power over Ethernet. What if you flip that idea and instead think of it as Ethernet over Power, or more correctly data over power. They've been doing that since the early 1900's. Granted I think the Ethernet-based solutions are going to win out over anything as it's very simple to add Ethernet to an existing building. However, it would be something to see an instance where your data and power is in the same cable and the fixture can demodulate the data stream from the mains connector. Something like that could be integrated into a building's dimmer rack, adding data wherever you've got existing cabling. Pre-2000 installs would definitely benefit from something like that.

Ethernet over Power has some technical challenges in multi-phase installations and with the harmonics produced by dimmers and power supplies. That said, I tried to get ETC interested in an EOP relay for Sensor racks but they have opted to pursue wireless solutions instead because it has a bigger addressable market. FWIW, I first suggested it here in 2011:

• https://www.controlbooth.com/threads/dmx-rack-or-dmx-pack.23104/page-2#post-204506
• https://www.controlbooth.com/thread...up-some-led-fixtures.27028/page-2#post-240816
• https://www.controlbooth.com/thread...t-cue-etc-gio-led-s4.24725/page-5#post-222404

sk8rsdad said:

Ethernet over Power has some technical challenges in multi-phase installations and with the harmonics produced by dimmers and power supplies.

Click to expand...

That's a good point. Most of the power-line comm applications I've seen implemented rely on the AC oscillating at 50Hz or 60Hz. I'm guessing it changes things a bit when you start chopping on it. I think you're right though, while an interesting thought experiment, the industry is full-steam toward IP-based and wireless models.

MNicolai said:

An ETC Colorsource Spot has 28A of inrush at 120V. Not sure how that corresponds to 48V for PoE, but each fixture would draw 3.1A at 48V -- granted, that current is split across multiple 23AWG pairs, but it still produces heat. Trying to power just one average fixture on your cable will warm it up beyond acceptable levels. Bundle a few cables together and you'll melt them in no time.

Click to expand...

My idea would to have a inverter in the fixture itself and only push 12-48v from the POE to the built in inverter which steps up thto voltage in the fixture to power the rest of it. Then you just daisy chain the rest of the fixtures as you would normally.

For older fixtures you just have an external version if you so desire.

Amiers said:

My idea would to have a inverter in the fixture itself and only push 12-48v from the POE to the built in inverter which steps up thto voltage in the fixture to power the rest of it. Then you just daisy chain the rest of the fixtures as you would normally.

For older fixtures you just have an external version if you so desire.

Click to expand...

@Amiers Perhaps I'm missing your point. Power's power whether lower voltage at higher current or higher voltage at lower current. Either way, you're looking at supplying / consuming the same amount of power give or take the various trade-offs and losses of the various methods. This hearkens back to why high tension lines between generating stations and end users normally operate at higher voltages over lengthy distances and are transformed down to safer voltages for city streets and lower again for residential blocks and end users. In a sense there are no 'free lunches', only different costs for losses, such as heat, along the way. It matters little whether you're dealing with AC and transformers or DC and DC to DC conversions. Perhaps @ship can invoke the spirits of Edison and Tesla?
Toodleoo!
Ron

@Amiers LED sources want DC power so feeding purpose-build fixtures DC would go a long way to reducing inrush current since the need for big honking capacitors is reduced. It would be sub optimal to use an inverter to convert to AC so the fixture can convert back to DC internally.

sk8rsdad said:

@Amiers LED sources want DC power so feeding purpose-build fixtures DC would go a long way to reducing inrush current since the need for big honking capacitors is reduced. It would be sub optimal to use an inverter to convert to AC so the fixture can convert back to DC internally.

Click to expand...

Yeah but in theory it would work.

I hate dragging long DMX cables, so I went with wireless DMX. There is now only one cable run; power.

@sk8rsdad hits a glancing blow on inrush, but let me clarify.

Inrush is a power supply problem and so outside of the POE issue. Small power supplies are fairly inefficient so add to the line voltage draw. Going POE put all of this in that nice dimmer err, rack room along with the multi-phase 'bridge' device. The "last mile" gets much easier especially if we move up to a 18gu wire or something. Yes it's entirely feasible, if you rework all the gear we currently use, except for the console.

I'm not betting on it becoming available anytime soon.