Showing results for 
Search instead for 
Did you mean: 
Create Post

A Dim View of PoE LED Lighting

I’m probably going to get some heat for this, but I have to get something off my chest. At Cisco Live this year, I saw a technology that was really flexible, with amazing controllability potential, and just cool: PoE-based LED lighting. Rather than connecting light fixtures to mains power and controlling them via a separate control network, it’s all one cable. Network and power, with the efficiency of solid-state LED lighting, with only one connection. However, after several vendor conversations, I can’t escape the conclusion that the idea is inherently, well… dumb.

Okay, Not Dumb, Just Math

Before Cree®, Philips®, or any of the other great companies with clever tech in the Cisco® Digital Celling Pavilion get out their pitchforks, I have to offer a disclaimer: this is just my opinion. But it is the opinion of an IT engineer who also does lots of electrical work at home, automation, and, in a former life, network consulting for a commercial facilities department. I admit I may be biased, and I’m not doing justice to features like occupancy and efficiency analytics, but the problem I can’t get past is the high cost of PoE lighting. It’s a regression to copper cable, and worse, at least as shown at Cisco Live, ridiculous switch overprovisioning.


First, the obvious: the cost of pulling copper. We’re aggressively moving clients to ever-faster WLANs both to increase flexibility and decrease network wiring costs. With PoE lighting, each and every fixture and bulb has its own dedicated CAT-3+ cable running hub-and-spoke back to an IT closet. Ask yourself this question: do you have more workers or bulbs in your environment? Exactly. Anyone want to go back to the days of thousands of cables in dozens of thick bundles?  (Image right: The aftermath of only two dozen fixtures.)

Second, and I’m not picking on Cisco here, is the per port cost of using enterprise switches as wall plugs. UPNP is a marvelous thing. A thousand-plus watts per switch is remarkable, and switch stacking makes everything harmonious and redundant. Everyone gets a different price of course, but the demo switch at Cisco Live was a Catalyst 3850 48 Port UPOE, and at ~$7,000, that’s $145/port. Even a 3650 at ~ $4000 comes to $84 to connect a single light fixture.

It’s not that there’s anything inherently wrong with this idea, and I would love to have more Energy Wise Catalysts in my lab, but this is overkill. Cisco access switches are about bandwidth, and PoE LEDs need little. As one vendor in the pavilion put it, “… and bandwidth for these fixtures and sensors is stupid simple. It could work over dial-up, no problem.” It’s going to be tough to sell IT budget managers enterprise-grade stackable switches with multi-100 gig backplanes for that.

And $84/port is just a SWAG at hardware costs. Are you going to put a rack of a dozen Catalysts directly on mains power? Of course not. You’re going to add in UPS to protect your enterprise investment. (One of the touted benefits of PoE lighting is stand-by.) The stated goal by most of the vendors was to keep costs under $100/port, and that’s going to be a challenge when you include cable runs, IT closets, switches, and UPS. Even then, $100/port?

Other Considerations

There are a couple of other considerations, like Cat 3+ efficiency at high power. As you push more power over tiny network cables it becomes less efficient, and at a certain output per port, overall PoE system efficiency drops and becomes less efficient than AC LEDs. There’s also an IPAM management issue, with each fixture getting its own IP. That ads DHCP, and more subnets to wrangle without adding much in terms of management. Regardless of how you reach each fixture you’ll still have to name, organize, and otherwise mange how they’re addressed. Do you really care if it’s by IP you manage or a self-managing low-power mesh?

DC Bus for the Rest of Us

What this initiative really highlights is that just as we’re in the last gasps of switched mobile carrier networks, and cable television provided in bundles via RF, we need to move past the most basic concept of AC mains lighting to the real opportunity of DC lighting. Instead of separate Ethernet runs, or hub-and-spoke routed 120VAC Romex, the solution for lighting is low voltage DC busses with an overlay control network. It’s the low voltage and efficient common DC transformation that’s the real draw.

Lighting would evolve into universally powered, addressable nodes, daisy-chained together with a tap-able cable supplying 24-48VDC from common power supplies. In a perfect world, the lighting bus would also support a data channel, but then you get into the kind of protectionist vendor shenanigans that stall interoperability. What seems to be working for lighting or IoT in general is more future-proof and replaceable control systems, like wireless IPv6 networks today, then whatever comes next later.

Of course, on the other hand, if a manufacturer starts shipping nearly disposable white-label PoE switches that aren’t much smarter than mid-spans, mated to shockingly inexpensive and thin cables, then maybe PoE lightening has a brighter future.

What do you think? Besides “shockingly” not being the worst illumination pun in this post?


Wow..just think of the voltage drop over the 22g or smaller wires.  While LED's are getting more efficient, fluorescent is still a better bang for the buck at this point in time.

The higher output LED's do put out a fair amount of heat. If you ever look at some of the higher output LED's, they have heatsinks and airflow requirements.

Here is just one example



so now you have other heat sources to content with.

Yes LED's are great for longevity and such but they still require power and give off heat.

Running them of POE ?  I think not.

I agree! I see this as a demonstration of "we do this because we can!" I am not an OSHA expert but I have been in an office long enough to know that emergency "always on" lights are an absolute. So there will still be traditional electrical lights installed right along with the POE lights. IMHO lighting should be like analog, Always On.

Let me apologize in advance for what's coming.  Patrick, you've hit a nerve.  And it wasn't even you--it was the folks you're talking about.  But I'm pushing back against this due to the ridiculous cost it brings to networking.

Let's start by me admitting I understand the rationale for "Convergence."   I've seen the pretty documents that persuade people they will save a lot of money by having to cable only one network instead of three when they combine video networks and telephone networks and data networks together.  They ARE persuasive.

But they don't tell even half of the real cost of convergence.

Now someone else wants to jump on that convergence band wagon--electrical lighting.

Who else is already riding that wagon?  Security cameras.  Wireless Access points using Power Over Ethernet, and everything riding on those AP's, from guests streaming Pandora or Netflix to mission critical applications on internal corporate SSID's.

Let's take a typical high visibility deployment with increased demand for POE from a switch.

And just for fun, let's make it in a bank, or in a data center, or in a hospital, or in any location where your environment and customers demand 7x24 availability with Five 9's of uptime or better.

As a result of this you have deployed chassis switches with dual supervisors that support In-Service-Software-Upgrades (ISSU) so you can upgrade them without a hit to your customers.  Further, your chassis switches use dual power supplies, each with dual cables to different circuits--one pair to a UPS that is generator-protected, another to building power, perhaps without UPS or without generator protection.  Again, this is designed and paid for and done with the intention of 7x24 uptime of Five 9's or better.  You can lose building power, or lose UPS, and the switches will operate perfectly on just one power supply (which has two cables, to two different circuits).

Your chassis switch has 384 Gig Ethernet patch cables plugged into it--and all of its ports are active, supporting edge devices ranging from Thin Clients to Security Cameras to AP's to printers, and more.

You have a dozen of these chassis switches in your building, and your users have enjoyed years of no down time due to the great technology, and to your excellent advice and design skills, and due to the organization trusting you and funding your project appropriately.

Your Telecom group knows your switch offers POE, and they make plans to retire a twenty-five-year-old POTS phone switch and they'll replace it with modern VoIP phones and Call Gateways.  It makes sense to them--the old phone switch should be retired, and they've gotten their VoIP toes wet in this site by adding a VoIP phone here and there in it.  Now's the time they've decided to replace all 4,000 of the building's old POTS hand sets and wall phones.

They get the budget and make plans.  They contact you as a courtesy, and you think about the challenge.  You start to do the math.  4,000 VoIP phones spread across a dozen of your chassis switches is 333 phones per switch.  Each phone demands 7 Watts, meaning every switch has to come up with an additional 2.33 KW of POE above & beyond what it's already providing.

You planned ahead and bought dual 6500 Watt power supplies that are auto-sensing, and they can do 110V or 240V without an issue.  But they only put out 6500 Watts when they're hooked up to 208V.  The 110V in your network rooms only lets them do 3250 Watts, and a lot of that is used by the switch itself.

You warned your boss that one day the building's old 110V circuits wouldn't be sufficient to supply the demand from POE devices like AP's and cameras and Phones.  He said "We'll cross that bridge when we come to it."

You remote into one of your chasses and issue the "show power" command, and you see it has only  233 Watts available to share out, due to its own demand, and to other POE devices already using most of its 110V power budget.  There's no way it can support more than 33 more VoIP phones on that switch without upgrading building power from 120V to 208V. And you might need to support 384 phones on that one switch.

Worse, adding more power demand without increasing power supplied from 120V to 208V will reduce your power resiliency--you'll have to use both power supplies instead of having an N+1 solution.  This means if power to one of the four power cables is lost, the entire chassis will go down.

Similarly, if one of the switch's power supplies fails, everything's down.

You're no longer able to weather a hit to normal building power with just your UPS.  Your Five 9's are at risk, as are your customers and your business.

So you contact Facilities/Maintenance:  "I need a quote from your electrician to bring two dedicated 208V power circuits into each of twelve network rooms".

Maintenance gets hedgy:  "I can get you a quote like that in two or three months--our electricians are swamped.  Will you need anything else?"

You think for a while . . .  "Yes, we'll need additional cooling for the additional BTU's our switches will put out when they start providing that power.  Oh, and we'll want that power to be conditioned and protected by UPS and Generator.  And obviously the room chillers will have to be generated protected, too.  Hmm.  I think we'll need the room lights on Generator as well."

Silence . . .

Facilities:  "Let us start working on that; we'll get a team together to start working with you on BTU output and timing and generators and demand."

For your Hospital Grade / Banking Grade / Data Center Grade / Highly Available Network, Maintenance recommends a new A-B generator solution, with A-B building UPS protection.  This will allow them to do maintenance on either building UPS and on either of two new Generators without risking loss of service.  It requires all new electrical gear, room remodeling, and new breaker panels and new feeds from the electric company. 

Did I mention there are no 208V runs anywhere near your network rooms?

Total cost for that additional power:  $1.5 Million.  For one building.  Because 4000 VoIP phones are coming in, and they want to save money through Convergence, dropping their old phone systems for something running on the network, taking power from your switches.

Sooooo . . . .

You go back to Telecom and share the news.  They're dismayed.  "We already applied for the budget--and it was approved."

You can respond in several ways:

•  "You can put in the phones, but you'll need to provide a power brick for each of them.  4000 phones X  $75 for individual power bricks is $300,000 in power.  Plus the additional electrical outlets you'll need the electricians to install.  But that's a lot less expensive than $1.5 Million."
•  "Sorry, I don't have power for that many VoIP phones.  Please go back and resubmit a budget for the building infrastructure upgrades to support that kind of power demand.   Have you considered going back to a traditional phone switch?  Your system already has all that power allocated, ready to use for the phone handsets and dial tone, and your system will work in a power outage, while the network VoIP will die without big UPS's and additional generators."

•  "Your project will have to go on hold until Maintenance/Facilities can submit a budget request for the additional power needs, and until that request gets approved."

•  "Did I ever tell you about the hidden costs of POE . . .?"

•  "Oh, has anyone ever noted that we only provide our switches with enough UPS power to cover fifteen minutes of power loss?  It's too expensive to provide bigger building UPS's that can keep them up longer, so that means your phones won't work after fifteen minutes of down time.  Not like the POTS that stayed up and running no matter what the building power or network was doing.  But 15 minutes should be enough to call 911 and then get out of the building, right?  In a blizzard or tornado or security event?"

This is only one example of how Convergence isn't everything its advocates say.  It cost this one building an extra $1.5 Million just in electrical charges.  And you haven't even touched UPOE levels yet.  Or running the lighting system off the switch.

Convergence advocates don't talk about how the POTS was rock-solid, bullet-proof, tried-and-true technology that had no outages, and that provided its own power to phones, and has been working well for the last 50 years.  They just know the old phone switch has to go because it's old and not shiny, maybe it can't be supported anymore, maybe it doesn't have the cool new features that VoIP can offer, and replacing it with a new phone switch is $1M or more.  And they don't sell POTS switches, they sell VoIP hand sets and Call Managers, and they want your money.

Convergence advocates don't ask people to realize they're putting all their eggs in one basket.  When a network switch goes down or must be replaced, there goes your phones, your wireless, your security, your P.C.'s, etc.

And now you want to power your building's interior lights with that same network switch?

I'm not done yet.

Think about your network rooms.  UPOE is going to affect more than just your switches and their power supplies.

Maybe there are 2000 - 3000 CAT5 cables coming into any one of them, maybe in two or three big bundles in some spots, passing through floors or through walls, aggregating on top of racks.  And there are bundles of hundreds of cables running vertically through your network racks between the patch panels and your switches.  Maybe you had to do Core Drilling to get huge amounts of cables from one floor to another.  And maybe you have pre-stressed tension concrete that requires X-Rays of the floors before you can do your core drilling, to ensure you don't hit one of those tension cables hidden within the concrete when you drill.

UPOE supports up to 60 Watts, and you network was designed for 4 Watts to 30 Watts.  What will you do when some department purchasers new equipment that needs all 60 Watts from every port on your 384-port switch?

You'll upgrade your switches' power supplies from dual 6500 Watt units to dual 9000 Watt models, and upgrade your UPS's and Generators again.  And upgrade your air conditioning solution again.

Remember those big bundles of cables?  UPOE's 60 watts will make them hot.  And you haven't addressed that cable heat issue.  Big bundles of network cables are no longer accepted for electrical and fire codes.

It turns out the IEEE says that if you're ever going to run UPOE in your cables, you'll run into cables over-heating when they don't get enough exposure to air.  All cable bundles must be divided up into groups no larger than 100 cables, or you might worry about a fire.

How are you going to do that where the cables pass through those narrow holes from one floor to another, or through walls?  You'll have to cut the cables at the back of the patch panels, pull them back upstairs, make NEW Core Drillings, and feed the cables back down to the patch cables--but in bundles of no more than 100 cables per hole.  And then reterminate the cables & test them at the patch panels.

You might as well ask for budget to replace all the CAT5 with CAT6, and toss the CAT 5 out as an investment in future proofing.

So, now you're looking at recabling your building and going with the latest flavor of CAT6.  Those 4000 cables are only a fraction of the CAT5 cables in the walls; maybe there are 8000 or 10000 in a big network room. and you've got twelve network rooms.

It will cost you $550 in materials and labor for each new cable run to be installed by your favorite certified vendor.  That price doesn't cover removal of the old cable and old patch panel / wall plate terminations.

"But WAIT--That's Not ALL!" 

Your local cablers are all Low-Voltage Certified, and that's OK for 5 Volts or 30 Watts on a data cable installation.  However, UPOE cables that will run 60 Watts can't be installed with that certification; you've now got to hire full electricians to get the job done legally.  At a higher rate of pay.

Be VERY CAREFUL before you allow your organization to go down this path.  Make sure they KNOW the true cost of convergence, of providing UPOE to lights and phones and Cameras (some of which require all 60 Watts due to having tilt/pan motors AND heaters to keep things working on external cameras in the winter) and everything else.

Carefully explain to them how putting all their eggs in one basket will mean everything in one switch could be down for an extended period (including lights, telephones, computers, security badge readers and security cameras, panic bars, etc.) during a power event or during a switch replacement.  Plan on a minimum of four hours of down time for that switch replacement--assuming you've got all the right power and bundle sizes.  Shorter down times are an impossibility in an environment where there's insufficient space for a second chassis along side the first one, to which you could migrate patches one at a time.

There go your Five 9's.  99.999% uptime only allows five minutes of downtime per year.  Maybe you're lucky, and you only measure "unscheduled" down time, and this maintenance would be scheduled.  Still, you don't want to go there.

Don't be caught unaware of these limitations, don't let your customers and Facilities/Maintenance/Telecom/Security/IT Teams assume they can keep adding more and newer POE devices without a cost.  Convergence has a large hidden cost.

I am with you. We are building at least 6 new buildings in the next 5 years, and likely renovating a significant number of other spaces. Our facility people love the bells and whistles of PoE LEDs. So far the list of reasons not to use this technology have been enough to keep it at bay, but someday Information Services will lose the argument.

My next statement was going to be things that I hope get better about the technology, but  I won't hold my breath.

Level 9

Superb description of the hidden costs of "just adding a few VoIP phones" and a great tale about clicking past existing capacity and how much it really costs when you are close to the limit.  I have a few tales of my own ("we just need 4 more 110VAC outlets in the office"), but none so comprehensive.

All it really takes is great planning and great information sharing.  If you can eliminate the assumptions (there'll always be enough power to the switch, there won't ever be issues with our existing cable bundles and heat, UPS and Generators will always be sufficient and never need to be off line, etc.), an organization could talk rationally about the options.

But just because you can steer a car with your feet doesn't mean you SHOULD do it.  It's an interesting trick, but benefits no one with arms and hands in the long run.

So, too, must we treat consuming limited network switch ports and cabling for building lighting.  Folks will proclaim how much money you'll save, and you have to have the information about the REAL costs of it all.

And let's put on our paranoid hats, shall we?  If/when someone hacks your network, now they can shut off your lights.  Just the kind of annoyance a 6th grade script kiddie would think is funny, even if he leaves your data alone.  Now imagine what a creative virus writer or zombie master can do with hacked access to your lights . . .   Any coolness or savings from LED lighting isn't worth the risk and trouble and real expense of providing it with the Network.  Let your switches do what they do best--forward data.  Give lighting back to the electricians and save some hassle.

Oh, go for it--lay out your stories for us.  We're all professionals here (I hope!).

Level 10

As a guy who worked as an electrician prior to being IT, I've been thinking about this quite a bit as well.

The reality is that anytime you use DC, voltage drop becomes a real issue, so even using 48 VDC from a dedicated controller will have significant distance limitations and inherent loss. That is the reason most industrial and corporate fluorescent lights are 208VAC.

Heat doesn't become an issue with LED's until you begin to overdrive them, and as LED technology continues to evolve, it will become much easier to provide the required lumens of high quality light without overdriving those circuits. No matter how you slice it, LED's are still at least 2X as efficient as fluorescent, which tells you EXACTLY how much less heat is produced by LED's over Fluorescent Light Bulbs (FL for simplification). ANY power that does not get converted to light is waste heat. End of story. Many people seem to mistake the fact that FL's are cool to the touch, and thus don't produce much heat. Those same people have never burned themselves on the ballast the the FL's require to operate.

Let's talk about the goals of a connected ceiling. If every light is individually controlled and LED based, our control of the light becomes less about ON/OFF, and more about dimming from 0-10. We can also easily provide employee access to control the lights in their office via PC, smart device, or even (dare I say it) an app built into their LAN connected phone.

Is POE lighting the best solution for this? Absolutely not! You said it yourself, tiny amounts of data, large (in comparison) amounts of power. Cisco has a hammer, and they are looking for a nail!

There are two options that make much more sense.
1) mains connected devices controlled via wireless (preferably 900Mhz). This is a good solution, but as a WiFi guy, it make me cringe a little. (a lot)

2) mains connected devices with onboard powerline technology. NOW we are talking! Let's use the existing infrastructure. Replace ballast with a controller/stepdown module and tubes with LED's. Connect one IP connected controller to the mains power. Greenfield or brownfield/upfit both could easily work.

Sadly, that doesn't sell more switchports, so...

Level 12

"Hold on everyone, we're upgrading firmware on the switches and the lights will be out for the next 5 minutes!"  Sorry, but just don't see PoE lighting for general deployment as a useful thing.  Maybe in spot or niche usages, such as the dim recesses of a network closet, but that's about it.

Actually, we don't even want to consider POE LEDs in network rooms (don't say "data closet", or a closet is what they'll give you) because hey, if you've got a problem bad enough to require your presence in a network room, do you want to be working in the dark on it?

Level 12

I quite literally laughed out loud when I read this. Luckily I just put my water down, or I might be out a monitor right now!

I think POE needs to stay limited to devices that NEED network connectivity to do their job. Otherwise your going to have someone that wants to run the office Kuerig off the damn thing so they can turn it on and off from their phone.


the idea of spending Cisco money on base level appliance lighting is crazy - as you say, a white box switch is fine, but not Cisco thanks.

As rschroeder​ has already mentioned. Once you're powering all these things from the data closets, all that power has to come from somewhere and most data closets just don't have either the power or heat load capacity.

You end up with massive UPS's in places designed for small switches.

That and the Cisco POE switch is double the cost of a none-POE


it puts my network guy into the path of being on the hook when bits of the building lighting start misbehaving.

All of you swtich maintenance are now building maintenance

Yeah - this is a truly horrible idea.


I always enjoyed the HTCPCP (Hyper Text Coffee Pot Control Protocol) enabled coffee machine that would send a trap when the brew cycle is finished...

Then there is the COFFEE-POT-MIB

Level 14

As I was reading this I picturing myself going off on management for wanting to implement it.  Thank you for providing your experience.

I've gone through this multiple times.  With any luck, some folks might read some or most of my diatribe and carry away the important parts that will help their organizations avoid the pitfalls and traps inherent in non-Network people making decisions based on bad information and assumptions about a network's capabilities.

Level 14

Outstanding post rschroeder​ !

Powering lights via POE sounds more like a "life hack" than a sound solution..

Living in the northeast... I wonder how long it would take for someone to come up with a ceramic heater that fits under the desk that ran off POE.

Perhaps a wall clock or pencil sharpener.... oh well off to file my patents!!!

Level 10

I really can't believe that this is a thing... i really can't..What kinda of IT Department would put this into there IT Strategy and support it?

I'm ok with VOIP, POE Cameras and heck even Access-Control proximity sensors, but lights? yeah maybe when that technology Li-Fi becomes a thing, this might be a viable trade off.. -> Li-Fi - Wikipedia, the free encyclopedia

Whilst it's an impressive concept, I can't see anyone, bar the most irresponsible CIO, recommending this at that unit cost. It's bonkers to use enterprise-grade switching equipment to keep the lights on!

There may be some places where this will be used, some ultra secure environments where every single wire in a wall has to have known function, and known to be secure, but for the great unwashed this is like using the afterburner from a military jet engine to light your campfire...


Its about money...another way to get more cash from the customers.

It is all about marketing.  Some shops like the one stop approach..everything in one basket.

This way customers need more ports...which means more money for more equipment.

Then you need more power...bigger power supplies = more money.


Ok, in general I agree, that running PoE lightbulbs seems a bit on the crazy side.  But, maybe we should step back and think a bit on it rather than simply dismissing it.  I mean, maybe this is a gateway to potentially solving other problems.   I mean what else could it be used for?   And I'm not necessarily advocating for these types of things, but maybe by thinking past the simple lightbulb we can potentially solve other problems..


To me this could potentially be the biggest application.   I mean think about it, as Patrick said "We’re aggressively moving clients to ever-faster WLANs both to increase flexibility and decrease network wiring costs."   The problem is getting sufficient coverage where it is needed and backhauling all that traffic back to the switches.   If you could replace an LED lightbulb with one that was also an AP, you could target specific areas that were having problems.   Or, rather than having a number of high-capacity AP's dispersed throughout your environment to cover larger areas, what if you went with the approach cell towers are going with where you have lots of lower powered cells that serve a smaller geographic area rather than bigger higher powered cells that serve larger geographic areas.   Don't get me wrong, I don't think there is a one-size-fits-all approach, but for some areas in companies this might be a better approach?

Maybe similar to this product, but where this is a booster and somewhat part of a mesh-like solution, if you had wifi lights with PoE back-haul, that eliminates much of theproblems associated with a mesh-based approach.


This could definitely increase the whole "Big Brother is watching you" type approach, which is needed in some applications.  What if you had LED lights that could do things like motion sensing, security cameras, reading RFID tags or other things?   While a person is walking through an area you could track where they're going and do things like if they don't have a security badge on with an RFID tag, send a security guard to meet them.   Or if you're wondering where a key piece of equipment is that is tagged with an RFID, be able to locate it quickly or maybe even see where it was last moving and with whom (ie: which two RFID tags were moving together).   Or even better yet, if an RFID tagged piece of equipment is moving and there is no other RFID tag moving with it (ie: an untagged person is trying to leave with it) send a guard out.   Heck, you could even combine the RFID tracking with facial recognition to see if the person who is walking with the RFID tagged credentials is the person who own those credentials.   Lots of applications here I could think of...

This company makes a lot of stuff!   Once again, with these guys they're doing a normal lightbulb that has built-in wifi to backhaul this type of stuff potentially, if you instead had a cabled backhaul it makes it that much better.   Not to mention if you're in a high-security government facility, its possible sending this data over WIFI is prohibited.


So, in our company we have tons of temperature sensors all over and basically the HVAC system has its own proprietary network of sorts that takes all this information back and processes it and attempts to keep our areas cool or warm based on the season.  What if these things were integrated into a facilities network where the temp sensors were part of the lightbulbs.   I'm thinking you could even potentially go with PoE based systems that open and close HVAC vents based on how hot or cool a specific area is, and admittedly this probably wouldn't be integrating into the lighting itself, but maybe the lighting is a gateway into doing something like this.   I would think this would allow for much more granular control of the HVAC system to potentially eliminate the hot and cool spots.   For instance, where I'm located they had done some post-construction wall building and messed up the HVAC to where my area either gets super-cool or quite warm depending on the season.   If all they had to do was replace a lightbulb in my area with a temperature sensing one and add it in their facilities map and associate which HVAC vents were associated with my area as opposed to the area they walled off, they could easily and cheaply remedy this situation.


I'm sure there is more stuff that could potentially be done.    I would think its cheaper and easier to run a CAT5 cable than it is to run AC power cabling, or HVAC cabling and such.   And if the resultant lighting was better than the glaring fluorescent lighting I'm under I'd be all for it!  And especially if the mindset was to replace all cabling with one unified system, that this could really start making sense somehow.   Just trying to think outside the box...

Anyone else have ideas for how this could potentially make sense?

Level 20

It's another dumb idea really... like was stated a PoE port plus most likely not Cat3 cable more likely same as everything else Cat 6e or 7e or something... can't be cost effective at all. I'm sure the cabling companies LOVE THE IDEA and of course the switch makers... waste PoE port for a light... what???


in some circumstances it could make sense - but would require NEW networks, not piggy backed onto a real data network.

My biggest concern isn't the wiring or technology - it's that the network team becomes part of facilites and massively changes how building management is done.

That is fine for green field sites, but retrofitting it all would be one nightmare after another with no payback at the end


I'll bet patrick.hubbard​ is sitting back and grinning about now.

Talk about "light blue touchpaper and retire"


I'll bet the unions would have a fit !

talk about "Holy Separation of Duties Batman".

I'd rather folks sat on their PC's to warm themselves than someone developing a ceramic heater that ran off POE.  Heat is (perhaps) the worst, least efficient use of electricity.

The security camera / motion sensor floodlight expands the idea of "What can we do with people who have too much money?  I know!  Let's build a light bulb with a security camera in it, and have it connect via (slow, unreliable, insecure) wireless.  We'll charge $150 for every bulb, and then REALLY sock it to the consumer's wallet when they want to access the light bulb's camera records via our Cloud service"

Level 14

On it!!!! A subscription service.....

Level 11

At that cost, I think not.


I do agree with that sentiment, the stuff people are putting on wireless these days is bordering on absurd.  I'm just waiting for someone to publish a probably-bogus paper on how "WiFi is unhealthy" and seeing everyone rush to remove it from everywhere.   Places like Starbucks advertising "WiFi Free" instead of "Free WiFi" and such!  🙂

That being said, having come from a super-high security enclave, I could see them buying into running PoE LED lighting and embedding security-like devices into them though.  Something they wouldn't do if the backhaul was WiFi.   Not to mention companies with a Big Brother mentality. 

Level 12

Boy I am glad its Friday because my brain just went poof. I went to that "url" and it didn't load so I got annoyed and started digging into the web filter to figure out why. 10 min later it finally hits me. Derp.....

Level 15

Remeber events in BR. trash.

Level 16

Copper is getting expensive and the trend was to use less and less of it. That was the whole idea behind the IP phone and plugging your PC into it - 1 wire to the desk.

Now you want to fill the ceiling up with thousands of cables just for lighting? No Cisco I don't think so....

Level 13

He he.  I went back and forth on it on the flight back.  These comments are amazing and I'm glad to open it up.

About the Author
I'm the Head Geek and technical marketing director at SolarWinds, (which basically means I'm an mature geek in the services of the product team). When I say geek I mean Geek, with extreme prejudice. I started writing assembly on my Apple II, got a BITNET email account in 1984, ran a BBS @ 300 baud, survived X.25, abused Token Ring, got some JavaScript award love in '96, and my hack flight notification service still backs Along the way in various jobs I’ve been a developer, SE, PM, PMM, and now principal evangelist. (Let us all join hands around the server.) Over 10 years at SolarWinds I’ve hatched our online live demo systems, managed the SolarWinds Certified Professional program, launched the Head Geek program, helmed SolarWinds Lab and THWACKcamp, and these days I’m focused on the hairball that is Hybrid IT, Cloud, DevOps and helping IT admins learn new skills not just to manage increasing complexity, but accelerate their careers. I’m always looking for new and more fiendish ways to use our products- just like our customers. And when I have a few spare minutes I fly a little when the weather is good.