Page 1 of 1

Odd, just odd...

Posted: Wed Mar 18, 2009 5:09 pm
by Sabre
Well, we're moving offices here and I came across this. Why would one want a Faraday cage on a CAT5 cable? Looking up the electrical characteristics of the cable doesn't really show a need for it. Even with PoE, the current isn't high enough to justify it. Looks like it came from Compaq/HP servers.... Discuss...

Re: Odd, just odd...

Posted: Wed Mar 18, 2009 5:20 pm
by complacent
Those cables come with their (now defunct) Digital Sender 9100C. I never understood it myself...

Re: Odd, just odd...

Posted: Thu Mar 19, 2009 12:46 pm
by ElZorro
http://en.wikipedia.org/wiki/Ferrite_bead
Ferrite bead
From Wikipedia, the free encyclopedia


A ferrite bead is a passive electric component used to suppress high frequency noise in electronic circuits. Ferrite beads employ the mechanism of high dissipation of high frequency currents in a ferrite to build high frequency noise suppression devices. Ferrite beads may also be called ferrite cores, ferrite rings, ferrite EMI filters, or mistakenly as ferrous beads.[citation needed]

Overview

Ferrite beads are similar to inductors in some ways but work especially in an area that is parasitic for general purpose inductors. They essentially act as a high impedance, or "resistor" to high frequency EMI/RFI electronic noise. The absorbed energy is converted to heat and dissipated by the ferrite, but only in extreme cases will the heat be noticeable.
Ferrite beads are one of the simplest and least expensive types of interference filters to install on preexisting electronic cabling. For a simple ferrite ring, the wire is simply wrapped around the core through the center typically 5 or 7 times. Clamp-on cores are also available, which can be attached without wrapping the wire at all.
Basically it cuts down on noise, and works fine as long as your signal isn't in the same freq range as your noise. This is why you see it more so on 'low speed' stuff like USB, or CAT5 that they aren't pushing tons of data through.

And Julian, the dean called, he wants your degrees back. Both of them.

Re: Odd, just odd...

Posted: Thu Mar 19, 2009 2:18 pm
by zaxrex
But on twisted pair Cat5, the bead shouldn't be necessary as they are by design good at inductively coupled noise cancellation, especially for near-field hi freqs.

Still think it is overkill, but it could do some good. But why not have it at both ends, or just in the middle?

Re: Odd, just odd...

Posted: Thu Mar 19, 2009 2:49 pm
by ElZorro
zaxrex wrote:But on twisted pair Cat5, the bead shouldn't be necessary as they are by design good at inductively coupled noise cancellation, especially for near-field hi freqs.

Still think it is overkill, but it could do some good. But why not have it at both ends, or just in the middle?
Agreed, thats why they use pairs and twist them, but with a very noisy EMI/RF environment (like you can get an a server room with all the different power sources, fans, etc) a 6' long cable can become a pretty amazing antenna.

Don't know what to say about the second part. I've always seen them installed near an end and never more than one. :shrug:

Re: Odd, just odd...

Posted: Thu Mar 19, 2009 3:42 pm
by Libra Monkee
If I ever get to the point of nerdom where I'm geeking out over cables I find, shoot me.

Re: Odd, just odd...

Posted: Thu Mar 19, 2009 5:16 pm
by Sabre
ElZorro wrote:
zaxrex wrote:But on twisted pair Cat5, the bead shouldn't be necessary as they are by design good at inductively coupled noise cancellation, especially for near-field hi freqs.

Still think it is overkill, but it could do some good. But why not have it at both ends, or just in the middle?
Agreed, thats why they use pairs and twist them, but with a very noisy EMI/RF environment (like you can get an a server room with all the different power sources, fans, etc) a 6' long cable can become a pretty amazing antenna.

Don't know what to say about the second part. I've always seen them installed near an end and never more than one. :shrug:
lol, so you thought you would get me back from the spark plug thread, eh? EPIC FAIL! ;) Did you not see me looking up the electrical characteristics of CAT 5 to verify that it doesn't give off much EM? And as Zak pointed out (and I thought was understood and I didn't need to say) CAT5+ is twisted pair, so it should negate the need for one of them. Although more of the wires are used in 1000Mbps operation than 100Mbps. In any case, I just thought that it was really odd considering the current going through the wire is not high at all, which should make the cage a mute point. As some one who has designed stuff on the micron level, I understand all kinds of interactions ;)

Re: Odd, just odd...

Posted: Fri Mar 20, 2009 10:32 am
by ElZorro
Sabre wrote:electrical characteristics of CAT 5 to verify that it doesn't give off much EM?
Ahh, you're assuming they are concerned about the cable giving off signals. :) My point is the issue is the cable picking up signal/EMI from the environment and corrupting the data. Even with a twisted pair if the noise is great enough you can effect a very high frequency and very time sensitive signal.

Re: Odd, just odd...

Posted: Fri Mar 20, 2009 10:33 am
by ElZorro
Libra Monkee wrote:If I ever get to the point of nerdom where I'm geeking out over cables I find, shoot me.
Ok.

:gun:

Re: Odd, just odd...

Posted: Fri Mar 20, 2009 10:36 am
by Sabre
ElZorro wrote:
Sabre wrote:electrical characteristics of CAT 5 to verify that it doesn't give off much EM?
Ahh, you're assuming they are concerned about the cable giving off signals. :) My point is the issue is the cable picking up signal/EMI from the environment and corrupting the data. Even with a twisted pair if the noise is great enough you can effect a very high frequency and very time sensitive signal.
Good point. I would think that the shielding around the pairs (basically aluminum foil) would prevent this, but given significant strength and proximity, I suppose the wire could pick up "noise" ;)

Re: Odd, just odd...

Posted: Fri Mar 20, 2009 10:46 am
by ElZorro
Sabre wrote:I would think that the shielding around the pairs (basically aluminum foil) would prevent this
I bet if you rip that cable apart you'll see the foil is missing. :)

Re: Odd, just odd...

Posted: Fri Mar 20, 2009 10:50 am
by Sabre
Every since one that I have taken apart (5E or 6) has it!