Placement of PoE splitter

[Plarsson]

Hi,

I may need to use a PoE splitter close to one of my cams to add an additional IR light. For instance Amazon.com: Urban Security Group PoE Splitter : Power Non-PoE Cameras :"Splits" Incoming PoE Signal to 12V DC Power + RJ45 Ethernet Video, 10/100Mbps, 15.4W Per Port, IEEE 802.3af, True Plug-&-Play: Industrial & Scientific
How do you usually install it? Looking at measurements, it doesnt fit in the camera junction box (I have the pfa130 from dahua). Best would be a completely hidden install, but I dont think it is possible for me at the position I want the camera? I could possibly have the splitter say 10m from camera and IR light, but then i have 3 cables instead of 1 at these 10 m.

Thanks for advice!

 

[Teken]

You'll also need this female to male (Y splitter) if you intend to add PIR:

https://www.amazon.com/Splitter-Adapter-Y-Cable-Security-Parking/dp/B07J28SX6Y/ref=sr_1_7?dchild=1&keywords=12+volt+splitter&qid=1611613398&sr=8-7

Besides that you will need to confirm your POE Switch / POE Injector can provide the needed current to run both the camera and IR light. Keeping in mind this also assumes you're using solid 23 AWG (copper) ethernet cable. If in doubt of the ampacity and the cable in place its best to run a 22-4 / 18-2 cable to power the new IR light.

You don't want to run into a situation where there is a voltage drop and the current goes up. As that will smoke what ever is either powering the camera / IR / Both. Never mind melt the wiring and posing a fire risk within the homes superstructure . . .

 

[Plarsson]

You'll also need this female to male (Y splitter) if you intend to add PIR:

https://www.amazon.com/Splitter-Adapter-Y-Cable-Security-Parking/dp/B07J28SX6Y/ref=sr_1_7?dchild=1&keywords=12+volt+splitter&qid=1611613398&sr=8-7

Besides that you will need to confirm your POE Switch / POE Injector can provide the needed current to run both the camera and IR light. Keeping in mind this also assumes you're using solid 23 AWG (copper) ethernet cable. If in doubt of the ampacity and the cable in place its best to run a 22-4 / 18-2 cable to power the new IR light.

You don't want to run into a situation where there is a voltage drop and the current goes up. As that will smoke what ever is either powering the camera / IR / Both. Never mind melt the wiring and posing a fire risk within the homes superstructure . . .

Hi,

Thanks for your comments.
Yes, of course, I will use a Y splitter as well

I have GS1005HP 5-Port GbE Unmanaged PoE Switch | Zyxel, which should be able to run both camera and IR light. It is a PoE+ and it has 50W in total, shared on 4 ports, max is 30W per port. Im running Dahua cameras on all four ports, and they are max 7-8W, i.e., in total 32W maximum. The univivi ir illuminator that is often suggested here is at 8W. All in all I think I will be OK.

Im not using solid copper ethernet cable, Im using the kind where each signal cable in the network cable is made up by many small copper threads. (Dont know the name of this type of cable in English, sorry). Is this a big problem? I will only run the cable max 20m. Looking at my Dahua cameras and suitable IR illuminators, they dont have solid copper cables.

 

[TonyR]

You don't want to run into a situation where there is a voltage drop and the current goes up. As that will smoke what ever is either powering the camera / IR / Both.

Which version of Ohm's Law are you applying here?

 

[Teken]

Which version of Ohm's Law are you applying here?

I X V = P (Watts)

Keep in mind this condition happens more often in AC circuits, think what a brown out (sag) in line voltage has on modern electronics.

In DC circuits where there are transformers like there are in many POE splitters and modern power supplies within a POE Switch / POE injector.

The first half cycle of a 120 VAC sine wave continues to be present due to CEMF. Also keep in mind almost every circuit has fixed resistance and sometimes that same resistance changes due to an increase in heat which directly translates to increase in resistance and thus more current draw.

A perfect example is a emergency strobe light which is life safety device which must operate in all electrical conditions.

Here’s a snippet from a friend that explains just one of hundreds of working examples as to why current May increase.

Why does Current Go Up as Voltage Goes Down in a Fire Alarm Strobe?

The strobe flashes at a constant rate, and the brightness of the flashes doesn't get dimmer as the strobe's input voltage decreases. In spite of differences in voltage on its input terminals, when flashing, a strobe is always putting out the same amount of light power. To keep a constant light...

www.douglaskrantz.com

In short if a piece of hardware draws constant power because it’s highly regulated this condition will be present.

Here is a nice example with respect to motors and everyday IT Equipment:

Voltage drop, amperage increase?

We had an electrical incident where I work a couple of nights ago, and I'm trying to wrap my brain around it. Our house engineers have maintained systems at 150 amps on legs with 200 amp fuses....

ask.metafilter.com

Bottom line, complex loads that are highly regulated may cause this condition to be present. Which also affirms the importance of a AVR & UPS in use for critical loads.

 

[tigerwillow1]

I'm using about a half dozen POE splitters and have them all in a generic waterproof plastic box close to the camera and external IR illuminator. The boxes are often called project boxes. All of the interconnections are DC circuits so you don't have to be concerned with reactive power or any of the other AC circuit gobbledygook. The "many small copper threads" is referring to stranded cable (vs. solid cable). A stranded cable and solid cable of the same wire gauge have the same current carrying capacity. Permanently installed wiring generally uses solid cable. It's easier to manufacture and all things being equal should cost less. I'm not aware of any disadvantage of using stranded network cable other than it will sag more between supporting points. All of my external IR lights are constant power devices and the cameras might be too (I haven't measured them), so it is true that as the resistance of the cabling goes up, so will the current in the circuit. If the load on a port exceeds 30 watts, or about 0.6 amps, the switch should cut it off. The current load is carried by 2 parallel conductors, so the maximum current per conductor should be about 0.3 amps. There's no hard and fast rule about how much current is safe to carry in a wire because there are so many variables. I'll just say that my opinion is that there's a pretty nice cushion of how much current a 23AWG copper network cable can carry over how much a 30 watt POE port can supply. If there was a simultaneous failure of the port's current limiting, and an excessive load on the far end, then something bad could happen. I've bought a bunch of the Y-cables and noticed that some of them have pathetically small conductors.

There's a more square form factor for splitters, for instance this one Gigabit PoE Splitter - TRENDnet TPE-104GS . In some spaces it will fit better than the long, narrow shapes. There's also a more expensive splitter that re-inserts the power on the splitter's network output, letting you run just one cable to the camera instead of two.

 

[Teken]

These are primary difference between stranded vs solid cable.

- Transmits electrical signals better over longer distances. In the field this is the big difference in seeing 720, 1080P, 4K.

- Lower attenuation (resistance). As noted up above when long runs or maximum technical distances must be used. An increase in resistance results in a reduction of supplied voltage.

That is the difference of something working fine over the long term or something not and unreliable causing video drop outs, power failures, and ultimately hardware failure due to short cycling.

When you add extreme weather from hot to cold. The drop in voltage will cause an increase in current because almost all modern PSU’s use a wide regulated power supply.

This is why you see multi voltage PSU’s that can operate from 90 - 240 VAC. Yet still provide the same regulated 1 amp DC over that wide range.

In modern UPS’s many offer trim / boost (AVR) to provide a regulated voltage to the attached load when the line is high / low.

Despite popular belief a voltage sag / lull is more harmful then a voltage rise / surge. Because current engineers are more focused on surge / spikes as this is the primary failure mode in electronics besides extreme temperatures like (heat) which causes dielectric capacitors to dry out.

- Lower Cost: Just easier to manufacture

Ultimately people will do what ever they want regardless of industry standards or best practices. My view is do it right the first time as this saves time and money in the long run. Everyday on this forum and others it’s the race to the bottom in terms of price.

That’s just the real world we all live in. But when you’re literally using all of the cheapest hardware and wiring in the market. It’s imperative to follow basic electrical safety and ampacity that allows You a margin of error.

I have yet to see a single person in my time cut open a Y adaptor to confirm the gauge of wire being used. In turn I have never seen a single person cut open a POE splitter to see the same.

Given all else being equal there is a reason why X brand cost more vs Y brand. Because that product actually uses the correct sized copper (gauge) of wire and not some hair thin CCA wire inside!

Short runs that have low current rarely causes problems because there is such a wide margin.

Regardless, I don’t know of anyone who enjoys scaling a 25 foot ladder running cable once, twice, never mind 3 times!

 

[TonyR]

Also keep in mind almost every circuit has fixed resistance and sometimes that same resistance changes due to an increase in heat which directly translates to increase in resistance and thus more current draw.

Nope...increase the resistance in a series circuit and the current draw DECREASES, not increases.

Ohm's law clearly states that the current flowing in a circuit is directly proportional to the applied voltage and inversely proportional to the resistance of the circuit,

 

[Teken]

Nope...increase the resistance in a series circuit and the current draw DECREASES, not increases.

Ohm's law clearly states that the current flowing in a circuit is directly proportional to the applied voltage and inversely proportional to the resistance of the circuit,

This assumes a simple resistive circuit. This is not the case in a inductive / reactive circuit as I outlined and provided links of examples.

I should have probably offered more clear insight with respect to my comments about resistance. As noted a standard multi voltage PSU that everyone uses can supply voltage from 90 - 240 VAC while at the same time output a steady 1 amp DC.

In basic ohms law how could that be possible if line voltage dropped. Why wouldn’t current drop as well from the PSU?!?

Because there are components like a transformer, capacitor, that allow the PSU to output a steady 1 amp DC. Now, if you have a strobe light as I provided as an example it will draw a steady amount of current regardless of the supplied voltage because it’s designed to within its operating range.

Now, if something can and will continue to suck the same amount of current or more while the voltage drops. What can happen to a improper system using improper wire gauge / type of cable?!?

Almost every cheaply made POE Spitter, POE Injector uses a transformer. Because of this depending upon how the circuit is designed the CEMF will continue to operate said device from seconds to minutes.

During this low voltage period if caused by a brown out (voltage sag / lull) that camera / strobe / insert what ever device will continue to draw the same current or more because the PSU inside of the device was designed to in hopes of reducing electronic failure.

So what normally is seen is with equipment that is highly regulated but begins to fail is this:

Voltage is low from the source doesn’t matter why. The wire begins to heat up this causes more resistance and more voltage drop.

Regulated device (Insert whatever device) doesn’t care but draws the same amount of current which causes more heat in the wire at which point the insulation opens causing arcing or the wire breaks open.

Depending upon how good or bad the design is for said device it will either stop operating (ideal case) , continue consuming the same power (This is a compromise), or increase in current draw (This is dangerous yet happens everyday in cheap electronics based on price).

The links I have provided above offer many examples as to why. Applying Basic ohms law isn’t something you do when your circuit in question isn’t a plain resistive load.

Everyday I see someone using stranded CCA wire in walls for POE. These are the same people who also purchased the $0.99 POE Splitter, coupled with the $35.00 POE Switch.

Now, what are the odds of that going wrong?!? When you add on someone who believes it’s OK to run more power through the same subpar infrastructure???

 

[pozzello]

back to the OP's question, I've used these to split off a 12V device from the PoE link:

https://www.amazon.com/gp/product/B07RB26YX6

not the cheapest, but relatively small form factor and works well...

 

[TonyR]

Holy crap...forget it, man..

 

[tigerwillow1]

I really think it's a disservice to bring AC circuit considerations into a discussion about DC wiring interconnects. The voltage drop in the wiring is based on how much current is pulled by the load at the end. To the wiring it's a simple DC load. It just so happens that most of the IR illuminators, and possibly the cameras, are constant power devices. They do not present a fixed resistive load. They do not present a fixed current load. Within their input voltage spec they present a fixed load in watts. As the voltage goes down, they draw more current, and vice-versa. To the interconnect wiring they're just a black box that draws a certain number of watts. Doesn't matter if there are transformers or AC circuits inside of them, they are DC-in and DC-out devices. Yes, the end result of lower voltage leading to higher current and higher wiring losses is correct, but understanding it doesn't need any of the complications of AC circuit considerations.