DComTalk.com

kmatpral@yahoo.com · Guest

Hello -
I am using a 1000B-T Copper to 1000B-FX(15kms ) media converter. Is
there any limit to the number of these media converters that I can
cascade with out impacting the following
1) Propagation delay
2) latency
The media converters are purely PHY devices with SERDES and so I
presume that there is no buffering in them.

Could the experts comment on this ? Is there a spec which limits the
numbers of such devices that can de cascaded ?

What would be the propgation delay introduced due to the medium of
Fiber( 300m per microsecond) and due to the medium of Copper ?
Thanks
Kott

Walter Roberson · Guest

In article <1105419376.692941.170790@c13g2000cwb.googlegroups.com>,
kmatpral@yahoo.com <kmatpral@yahoo.com> wrote:
:I am using a 1000B-T Copper to 1000B-FX(15kms ) media converter.

Would that be something similar to the Allied Telesyn AT-MC1005/2 ?
The installation manuals for that specifically show a pair
of them being used in series, but I see no information at all
about latency or propagation delay in them. Nothing in the
FibroLan GSM1012 literature either. The Fiberdyne FTX-210M
page specifically says "with no performance degradation"

:What would be the propgation delay introduced due to the medium of
:Fiber( 300m per microsecond) and due to the medium of Copper ?

Interesting. I found one site offering a specific cat5e cable that
indicates that the propagation delay is

534 + 36/sqroot(f) ns/100m maximum

where f is the frequency in MHz from .772 to 100.

I see other sites saying 4.7 ns/m to 5.2 ns/m -- it appears that
doesn't take into account the connectors or whatever other fixed
factor is at work.

If I read the sites correctly, ~540 ns is a common fixed-delay figure
independant of cable length, and that one must then add the 4.7-ish
ns/m for distance. That's like saying that each 100 m copper
patch cable has a propagation delay equivilent of about 210 m
worth of wire.

So, how many can you cascade without affecting the propagation
delay? None! The convertors are useless without a copper patch cable
to connect them to something (if only another convertor), and
that cable is going to have a big delay no matter how short it is.

Note: I have little experience in the physics of cables. I am
*not* an expert in this area: I'm just good at google'ing and
interpreting what I find there.
--
We don't need no side effect-ing
We don't need no scope control
No global variables for execution
Hey! Did you leave those args alone? -- decvax!utzoo!utcsrgv!roderick

	DComTalk.com Forum Index -> Ethernet	All times are GMT
Page 1 of 1