General Discussion
  >> Fibre Broadband


Register (or login) on our website and you will not see this ad.


Pages in this thread: 1 | [2] | (show all)   Print Thread
Standard User deleted
(deleted) Mon 01-Oct-12 18:23:40
Print Post

Re: Infinity installation -cabling question.


[re: Anonymous] [link to this post]
 
Could you run 12v power over a spare pair in the cat5? (draw, length and gauge permitting)

cheers, a
Standard User deleted
(deleted) Mon 01-Oct-12 19:24:15
Print Post

Re: Infinity installation -cabling question.


[re: deleted] [link to this post]
 
Power over Ethernet (PoE) standards seem to be more like 44-50 volts. If you wanted to supply 12V, I imagine it would be possible.

I wonder why such a high voltage? Perhaps because they need the current to be low? Perhaps because it is congruent with the standard Telco power of -50V?
Standard User deleted
(deleted) Tue 02-Oct-12 01:21:39
Print Post

Re: Infinity installation -cabling question.


[re: deleted] [link to this post]
 
The higher the supply voltage, the lower the wire loss. There are several online wire loss calculators. Most are US-based with imperial measurements just to complicate smile

This wire loss calculator is nice though. All the maths is done in Javascript:

http://www.nooutage.com/vdrop.htm

It only goes down to 14AWG, but the values can be extrapolated to fit the common gauges of comms cables (24AWG and 26AWG).

Good quality cat5/5e cable has a conductor diameter of 0.500mm (about the same as 24AWG). And 24AWG solid copper apparently has a DC resistance of 5 ohms/100metre at 25degC. Whereas 26AWG (0.400mm diameter copper core) apparently has a DC resistance of 8 ohm/100m at 25degC. The DCRs for solid aluminium 24AWG and 26AWG are given as 9 ohms/100m and 14 ohms/100m respectively (at 25degC).

The load of the Huawei HG612 is roughly 600mA @ 12vDC.

According to that calculator, if a spare pair from a 30 metre length of solid copper AWG24 (cat5) cable is used to power the Huawei, there will be a voltage drop of 13% (down to 10.4 volts) at the modem.

And the higher the load, or the longer the cable run, or the finer the wire gauge, the worse that voltage drop will become.

The alternative VDSL2 modem from Openreach, the ECI B-FOCuS, draws a higher current of some 700mA. Powered over a spare pair in the same 30m length of 24AWG cat5 cable, that would see an even higher voltage drop of ~15%.

At some point, when that wire loss gets too great, the voltage regulator in the modem is going to give up and it will fail to energise the device.

Using a 'spare pair' in a cat5 cable to directly supply 12 volts DC to power an average DSL CPE looks to be limited to domestic cable runs of ~20 metres. Though if there's a second unused pair in the cable for 'doubling up' to carry DC power, then the effective maximum length could be doubled too.

It wouldn't be hard or expensive to do this quite tidily using standard LJ6C modules. An off-the-shelf module for RJ45. And the DC power socket could be fashioned out of a module blanking plate and a standard barrel-type chassis socket like this:

http://www4.picturepush.com/photo/a/10514432/img/pow...

Caution! All back of envelope stuff, not to be relied upon!

cheers, a


Register (or login) on our website and you will not see this ad.

Standard User deleted
(deleted) Tue 02-Oct-12 09:28:32
Print Post

Re: Infinity installation -cabling question.


[re: deleted] [link to this post]
 
Basically true but over a short distance the loss is insignificant and can be ignored.

It does become importent with power at high currents hence the use of very high voltages on the National grid. It is partly as well down to needing to keep the weight down for overhead lines so they tend to use Ali which is not the best conductor
Pages in this thread: 1 | [2] | (show all)   Print Thread

Jump to