Quote:
Originally Posted by I.C.D
Hi All I read the other day that any USB cable over 2 m should not be any smaller then 25AWG ,as anything smaller will slowdown any information on that cable,is this true
Ian
|
Hi Ian,
The real problem, as I say to fellow electrical engineering colleagues,
is that when it comes to the speed of light, God is a real underachiever.
Forget cable AWG's and all that gobbledegook for one moment.
We regard USB cables as what we formally term "transmission lines".
As you are aware, digital signals are transmitted as pulse waveforms.
In a vacuum, light takes about 1 nanosecond to propagate one foot.
In a cable, a signal travels a little slower.
Depending upon their surrounding medium (dielectric constant of
insulation and so on), a signal will take about 1.4 nanoseconds to travel
one foot.
When you consider today's high speed circuits, that is pathetically slow.
I mean, what was God thinking when he came up with 3x10e8 m/s
for being the fastest speed that can be? It's a real problem these days.
In particular, the problem for electrical engineers comes about when
we have signals whose rise times (that is the time it takes to transition
from a zero to a one or a one to a zero) becomes similar in magnitude
to the propagation delay time of a cable.
Man wants to change the signal and routinely builds circuits easily
capable of doing so, but God is still taking his time propagating an earlier
pulse down the cable.
It's not just the cables resistance that comes into play, but its
impedance which includes inductive and capacitive effects.
So it's also not the cables current carrying capacity that is the limitation.
It is the fact that it acts as a transmission line. Eventually you can
push the transmission like both mathematically and physically to the point
that the signal at the receiving end is so distorted and has so much
interfered with itself as to become unrecoverable.
That's why USB cables are only so long before needing repeaters.
Best Regards
Gary Kopff
Member Institute of Electrical & Electronics Engineers (IEEE) 42 years