PoE pi

I know there are PoE hats available for the 3B+ but I only have a 3B. When buying some 12V PoE splitters for CCTV cameras, I noticed this ?USB splitter, which arrived today, and works very nicely.

Reply to
Andy Burns
Loading thread data ...

Thanks for that, and I have just ordered one.

Reply to
RobH

I've seen these devices advertised before but what do they do? Power-over-Ethernet? What's 48 volts to do with it?

Can someone put me out of my ignorance?

Richard

Reply to
Richard Ashbery

Richard,

They combine eternet signals and power, and put it onto the same ethernet cable. The idea behind that is that you than can use that to feed (small) devices that do not have a powersource nearby.

You could ofcourse do the same by putting an internet as well as a powercable down, but that could be costly on longer distances.

That is the voltage that will be put onto the ethernet cable, and which the module on the other side which seperates the combined power and ethernet signals into seperate ones again needs to be able to handle (and convert into the 5 volts the Raspberry needs).

The voltage is that high because that way the voltage at the end of the (long?) ethernet cable will be high enough to be still usable (the voltage drops because of the internal restance of the cable - the longer the cable, the further the voltage goes down).

Hope that clarifies it (a bit).

Regards, Rudy Wieser

Reply to
R.Wieser

I have a PoE switch, that can provide power as well as data on half of its ports, some devices (e.g. access points, voip phones) can directly draw their power over a single ethernet cable.

A PoE splitter allows powering other devices from the switch, e.g a CCTV camera with 12V barrel connector, or a pi with 5V microUSB.

If you don't have a PoE switch, you can use a PoE injector at the other end of the cable, but it's not quite so convenient.

That's the 'phantom' voltage provided differentially over the pairs, the device converts down to what it requires, or for splitters they convert it down to common 5V or 12V supplies.

There is also a non-standard passive way of providing e.g. 12V over the unused pins (4,5,7 and 8) of an 10/100Mb ethernet cable.

Reply to
Andy Burns

Its more a case that a higher voltage needs less current for the same power.

I.e the drop at constant power is less absoluetley at higher voltages, not just less in proportion to the voltage

Power loss is I^2.R and power is V.I fraction lost is IR/V

So I = P/V so loss is P^2.R/V^2

Now the 132KV lines make sense...

Power loss is proportional to power SQUARED divided by voltage SQUARED.

But

--
I would rather have questions that cannot be answered... 
...than to have answers that cannot be questioned 
 Click to see the full signature
Reply to
The Natural Philosopher

The Natural Philosopher,

True. But you need to be able to convert that power to a lower voltage to again become usable. Where with AC that can be easily managed with a simple transformer, that doesn't work as well for DC.

If it would be DC, not really. :-)

Regards, Rudy Wieser

Reply to
R.Wieser

I would be a little concerned about loading my ethernet cable with say a Raspberry Pi but I guess some of you are finding it works OK.

Thanks for the explanation guys :-)

Richard

Reply to
Richard Ashbery

DC-DC converters are more complex than transformers being essentially switch mode power supplies but these days for smallish power levels they are cheap and very efficient with the complexity buried in an off-the-shelf chip.

--
Steve O'Hara-Smith                          |   Directable Mirror Arrays 
C:\>WIN                                     | A better way to focus the sun 
 Click to see the full signature
Reply to
Ahem A Rivet's Shot

People load the lines with all sorts of stuff. But anyway "do the math" as our US friends say.

48V in to 5V out. So the voltage in:out ratio is 48/5 is 9.6. It will provide 2.5A at 5V so will need 2.5/9.6 approx 260mA at 48V. Those are perfect figures. In reality the PoE adapter will be be 80-95% efficient, so the current will be higher.

But as long as you can pull 300mA say over your long length of CAT5/6 cable you'll be laughing. CAT5 is 24AWG ISTR wire and that has 5Ohms resistance over 60m length. You have 2 lengths in use so 10Ohms. Ohm says V=IR so 10 x .3 = 3V.

So 48V in should give 45V at the far end. Plus or minus. Looks like it may work as that unit is stated as 44-57V input. Probably we should make the limit 50m to give a bit more margin.

I may have dropped a googly in all those calculations, but I'm not proud so feel free to correct any mistakes!

Reply to
mm0fmf

Transformers are now being replaced by active rectifier/inverter devices, even in utility power systems.

High voltage semiconductors and high frequency switching are changing the landscape of power transmission and conversion. The ?smart grid? of the future will have a lot less iron cores. ;-)

--
-michael - NadaNet 3.1 and AppleCrate II:  http://michaeljmahon.com
Reply to
Michael J. Mahon

Very high power under water links are often DC viz:

formatting link

--
Steve O'Hara-Smith                          |   Directable Mirror Arrays 
C:\>WIN                                     | A better way to focus the sun 
 Click to see the full signature
Reply to
Ahem A Rivet's Shot

Steve,

Yep. Not because of it being a generic usable method, but because the need for DC - not influencing magnetic compasses - overruled the cost for it.

But I must say I'm rather surprised to see that new ones are being laid even now, in 2018 ... I though that magnetic compasses where all but obsoleted, especially in this day-and-age where even a $30 phone has a GPS reciever.

Regards, Rudy Wieser

Reply to
R.Wieser

With proper PoE the full 48 V is not present on the cable until the sending end has had a chat with the far end to see if it wants power. It's perfectly safe to plug anything onto a proper PoE supplied cable.

This is not the case with passive PoE, where the full power is present all the time.

--
Cheers 
Dave.
Reply to
Dave Liquorice

How does an unpowered peripheral talk or listen?

--




/ \  Mail | -- No unannounced, large, binary attachments, please! --
Reply to
Axel Berger

Erm no that's not the reason for HVDC. For long distance cable runs under water the main reason is lower losses to cable capacitance (about

30% lower losses), another use is connecting grids without needing to synchronise them.

It was never to do with magnetic compasses, they work just fine with overhead AC cabling why would running the cable under water make a difference ?

--
Steve O'Hara-Smith                          |   Directable Mirror Arrays 
C:\>WIN                                     | A better way to focus the sun 
 Click to see the full signature
Reply to
Ahem A Rivet's Shot

You're confusing me here. AC may make a needle tremble a bit in spite of being dampened but has no effect on the averge field. DC current on the other hand, even if coaxial or otherwise partially shielded, can make the weak terrestrial field go completely haywire.

Are you sure about your facts?

--




/ \  Mail | -- No unannounced, large, binary attachments, please! --
Reply to
Axel Berger

with some very clever electronics. the exact details of which are beyond my level of expertise.

formatting link
should provide the details if you really want to know

--
Mail server hit by UniSpammer.
Reply to
Alister

In PoE, the feed is between the transmit and receive pair centerpoints.

A proper Ethernet interface has transformers on both pairs, so supply feed in a normal network card is not an issue.

--

-TV
Reply to
Tauno Voipio

it's only type A where the power is between the 1,2 and 3,6 pairs

type B the power is directly on the 4,5 and 7,8 pairs, but there's still magnetics on those pins (if they're connected at all) so shouldn't come to harm.

Reply to
Andy Burns

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.