Power over Ethernet (PoE)

This topic is to discuss the following lesson:

Guys, lessons is very cool (I didn’t knew about the most recent type3/type4 standards, so it was a very good lesson to keep myself updated).

But I guess there is an issue:

1.2. Midspan

The PoE midspan method uses PoE injectors […]. These are intelligent devices, and like a PoE switch, they detect whether the PD requires power and, if so, enables power.

There are (rackmount) midspans that can power many cables

[PHOTO]

Or small PoE injectors you can use to power a single device

[PHOTO]

Then there is this image here, a single-device Ubiquity PoE injector.

The problem is that most single-device Ubiquity PoE injectors are not intelligent PoE devices (at least it was true at the time when I worked with them). They just add power on the wire, and they could “burn” a non-PoE device connected by mistake.

The explanation may make people misunderstand that the device on the photo is intelligent, although it is not.

If this is true, maybe you should update the explanation accordingly. Something like:

The PoE midspan method uses PoE injectors […].

There are (rackmount) midspans that can power many cables

[PHOTO]

These are intelligent devices, and like a PoE switch, they detect whether the PD requires power and, if so, enables power.

Or small PoE injectors you can use to power a single device

[PHOTO]

Some small PoE injectors, like the one from the image above, are not intelligent devices and just add power on the wire without previously detecting whether the connected device requires power. If not, the device may be damaged.

Does it make sense?

Update: I saw that, at some point in the lesson, you state:

There is also something called Passive PoE. This is not based on any standard and basically, this is a PoE injector that adds power immediately without negotiation. Because power is always on, it’s possible to burn out devices that don’t support PoE.

So, the explanation about Passive PoE is there. So maybe a single-line change that you can do to make the lesson accurate is:

The PoE midspan method uses PoE injectors […]. These usually are intelligent devices, and like a PoE switch, they detect whether the PD requires power and, if so, enables power.

1 Like

Hello Rarylson

Thanks for the feedback! I’ll let Rene know to take a look and consider making changes for clarity.

Thanks again!

Laz

Hi @rarylson

Thanks for the suggestion. I made some changes and moved the text around to make it clear.

Rene

Hi
the lessons are awsome
in the real world which POE we use ? different scenarios ?

is there any list of POE switch

thank you

Hello Abdul

In the past, many IP cameras, IP Phones, wireless access points would come with their own passive PoE injector. This was primarily done because such a passive injector is cheaper and more readily deployable. However, in recent years, more and more network device vendors have been incorporating active PoE using the IEEE 802.1af and other related standards, so today you will find that most good quality network devices will support this.

It is best practice to use the IEEE standards with switches because you can provide uninterruptible power to a switch, and have that switch in turn provide uninterruptible power to all of the devices it powers. This is more difficult to achieve using passive PoE injectors.

Switch vendors, including Cisco sell switches that support the IEEE PoE standards. As far as Cisco is concerned, each line of switches comes with different models, and some models support PoE. When looking at the datasheet for the 9300 series switches, for example, you can see the list of models available and which ones are PoE capable.

Switches that support passive PoE are rare, and typically not used in high-end reliable networks.

I hope this has been helpful!

Laz

Hello Laz,
image
In the above diagram, what is the difference between Maximum Power per PoE port and Maximum Power to PD? To my understanding, Maximum Power per PoE port means the amount power a switchport is capable of providing to a device and Maximum Power to PD means the least amount of guaranteed power a device can receive from a switchport. For example, for PoE+, a device (PD) can get up to 30W, but it is guaranteed that the device (PD) will at least get 25.5W of power because of cable loss. Is this correct?

Hello Azm

You’re on the right track with your understanding, but there are a few details that I would like to add. The Maximum Power per PoE port is indeed the maximum power that each switch port can supply to a device. If you measure the wattage at the port itself, you will be able to get up to this value. This is also the value that is subtracted from the total power budget of the switch itself, when this wattage is drawn from this port.

On the other hand, Maximum Power to PD refers to the maximum power that can be delivered to the Powered Device PD after considering power loss due to cable length and other factors. This assumes that cabling follows all the correct standards (i.e. 100 meter maximum cable runs, use of correct Cat5/Cat6 cabling, connectors, plugs, and terminations etc). The power loss over the cable is due to resistance, which leads to a decrease in power as the length of the cable increases. So, if you were to measure the wattage received by a PD, the maximum you would receive is that value.

So, using your PoE+ example, the switch port can supply up to 30W of power, but due to power loss, heat dissipation, and other factors, the PD will receive a maximum of 25.5W.

I hope this has been helpful!

Laz

Hey Laz,
Thanks for your reply, but I am still confused. According to your explanation, Maximum Power to PD refers to the maximum power that can be delivered to the Power Device (PD) after considering power loss due to cable factors. So now if we apply that statement to our example, PD can receive maximum 25.5 W (it could be less than 25.5 W since it says maximum 25.5 W and it will never exceed 25.5 W) from a 30 W switchport.
However, in your example, in the last paragraph, it says, the PD will receive at least 25.5 W out of a 30 W switchport which means the PD may receive more than 25.5 W, but it will never be less than 25.5 W. I guess if it was Minimum Power to PD, then it would make more sense to me. Please shed some light.

Thanks a lot.

Hello Azm

You are correct, I apologize for my misleading terminology. The maximum power delivered by the PSE is indeed 30W at the source. That is what it is rated at. The maximum power that will reach the PD is 25.5W. My use of “at least” was incorrect. However, the difference between the two is indeed due to power loss over the cable, and dissipation into heat.

I have corrected my previous post, and I have created a NetworkLessons note on the topic as well.

I hope this has been helpful!

Laz