Hello John.

14 Jul 04 19:58, you wrote to all:


 >>  JT>   I only wish I could have a portable version for use at
 >>  JT> client's offices,

 >> you could use a pocket sized oscilloscope, but it'd no longer be
 >> al-cheapo.

 JT>   That and, no-one uses coax ethernet anymore.

 >> the maximum cable lenght spec used for ertherenet isn't due to
 >> signal loss but due to having the cable short enough that a
 >> collision can be detected by all stations.

 JT>   Do you mean signal stregth for detection of the clash, or
 JT> propigation delays involved?

delays...

 JT>  With twisted pair, cable length limitations are based on
 JT> signal strength and quality by the time it gets to the end of the
 JT> run.

 JT> Propigation delays are only a problem if you're stacking hubs, up to
 JT> four deep is considered as far as you would want to go before delays
 JT> start causing weird problems.

is anyone still using hubs to extend their network when switches (which do
store'n'forward) are so cheap.

 JT>   Though this can be easily worked around by using stackable hubs,
 JT> where in large multi-user (in the hundreds) environments, is the way
 JT> you're supposed to do it anyway.

 >>  >> I wonder how much loss you get in 100m of that cable at 10Mhz.
 >>
 >>  JT>   Easily calculated.  Cable specs state loss at specified
 >>  JT> freqency use, you just add it up for your length of cable.
 >>
 >> I don't have any specs handy, do you?

 JT>   The following is from the back of an old DSE cattledog, though only
 JT> part of the list.  I've picked the two more common cable types.  Ask
 JT> if there are other cables types you're interested in, there are about
 JT> 15 different types listed, some 50 ohm, some 75 ohm, some 93 ohm.

 JT>                      Attenuation in dB/100metres
 JT> Type                50MHz  100MHz  200MHz  400MHz     Velocity factor
 JT> RG-58C/U  50ohm     10.8   16.1    23.9    37.7       66%

yeah, but how does it perform at between 5 and 10MHz... where coax ethernet speaks.

hmm looking up the specs for some of that cable at DC you've got about 5.5
ohms resistance over 100m  so driving a 50ohm load you'd be down 10% (on
voltage) that's 0.5 db loss

so the response at 10 Mhz is going to be somewhere between the 50Mhz
response and that.

 JT> RG-213/U  50ohm      5.2   7.2     10.5    15.4       66%

 JT>   Though you're not likely to use RG213 in network applications, it
 JT> is
 JT> in common use on long runs for radio transceivers. --

there was a type of "thick" ethernet that used a large diamerter
coax which was tapped using a special device that connected to the 15-pin
AUI socket on the network card. (rather than being cut and fed into a T
conector)

hmm, it could be that type that had the length due to propogation delays.
and the thin type due to attenuation like everyone else is telling me.

Jasen

--- GoldED+/LNX 1.1.4.7