[Beowulf] propagation velocity in cables

Lux, Jim (337C) james.p.lux at jpl.nasa.gov
Wed Sep 5 16:52:54 PDT 2012

The electrons move very slowly.. much more slowly than the electromagnetic wave propagates.   (I'm thinking a few mm/sec for some reason.. it's real slow, in any case)

Undersea cables these days are all fiber optic.  So about 2/3 free space.

Microwave terrestrial links are at a variety of frequencies. You trade it off between antenna gain (higher frequency has higher gain for a given size antenna) and atmospheric losses (rain, mostly).  There are bands everywhere from 2.6 GHz on up to 110 GHz and higher.  Long haul tends to be lower frequency, short haul is higher.

Yes, there are laser comm systems, but generally used only for short distances (<10km) like between two buildings. Advantage is that you don't need much power, it's really wide bandwidth (Gbps is no problem) and it's difficult to snoop on (beams are very narrow with minimal spillover).  OTOH, you also have to have active pointing, typically.. the building sway in the wind is enough to break the link.  And, of course rain is a problem.

(there are long range laser comm systems used for things like, oh, satellite to submarine, at very low data rates.. all you have to do is send a few bits: "Launch Missiles to target N")

Jim Lux

-----Original Message-----
From: beowulf-bounces at beowulf.org [mailto:beowulf-bounces at beowulf.org] On Behalf Of Vincent Diepeveen
Sent: Wednesday, September 05, 2012 4:44 PM
To: Beowulf Mailing List
Subject: Re: [Beowulf] propagation velocity in cables

hi Jim,

Thanks for a more technical explanation. Difficult to interpret for me though.
The 1/3 speed someone gave to me some years ago.

Litterally quote : "electrons move at 1/3 of the lightspeed through copper"

Maybe signal speed is not the same like electron speed?

Would be very interesting if he said that wrong, as he has a degree here, Phd cum laude in Natural Science and he improved something or better disproved something that has to do with quantum mechanics or something, basically he calculates everything about particles such as 'electrons' (my words), so i didn't verify all this, yet assumed what he said.

The most important question: what is the effective speed of signals through ocean cables?

We can safely assume they use fastest speed for this.

When i drove past that financial industry bunker in Land's End it seemed pretty solid to me.

The interesting question is that big satellite disk on the roof as well.
Microwave is simply civil allowed frequency i was told. Around 2800Mhz or so?
How fast is such communication?

Using a powerful laser would it somehow be possible to rival speeds of these cables?

On Sep 6, 2012, at 1:18 AM, Lux, Jim (337C) wrote:

> Propagation speed in copper is not 1/3 freespace.  More like 2/3 or 
> 80%, depending on the effective permittivity of the dielectric.  (c = 
> c0/sqrt(epsilonr))  Cat 5 is typically about 68% of free space.
> Foam dielectric coaxial cable is typically 81%  LMR-400 coaxial cable 
> is 85%.  There are air filled coax which get up to 90-92%, and open 
> wire line (like old telephone cables on poles) is around 95%
> For single mode optical fiber, n=1.62 for the core (1.52 for 
> cladding), so propagation speed is c0/1.62 or about 62% free space.
> Note that copper is faster than optical.
> People are setting up freespace microwave links for just this reason.
> Jim Lux
> C = 300k kilometers an hour.
> 300k / 3.6 meters a second
> 300 / 3.6 meters a millisecond
> 0.3 / 3.6 meters a microsecond
> 12 meters a microsecond
> Now i don't know whether signals travel through fiber at the full 
> lightspeed, as that's pretty important.
> In Copper it's roughly 1/3 lightspeed or so, so that's out of the 
> question to use.
> So saving some distance knowing 12 meters is a microsecond, that's 
> worth it.

Beowulf mailing list, Beowulf at beowulf.org sponsored by Penguin Computing To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf

More information about the Beowulf mailing list