On Thu, 13 Jun 2002 00:47:17 -0400, Anthony DeRobertis wrote: > >On Tuesday, June 11, 2002, at 02:14 , faisal gillani wrote: >> >> well i have 2 ethernet networks running which i want >> to connect but the distance between them is above 400 >> meters .. so this is way beyond the normal lan >> hardware .. > snip >The only thing that will >stop you[1] is if the signal is not strong enough on the other >end, or the noise has won. Amps might help, as would lower gauge >wire. I'd be interested in hearing how how far you can get away >with cat5.
This strikes me as an RF transmission line problem. Twisted pairs (cat5) are a reasonably low loss transmission line at audio and low rf. At 1 gHz they become very lossy. (I know, we're not talking more than 100 mb/s, but these are square waves which have high value odd harmonic components---3rd, 5th, 7th, 9th, and etc overtones.) Since these harmonics are attenuated more than the fundamental (a transmission line is a low pass filter), the effect is to slow the rise and fall times of the wave form. This may cause the detector to mis-read the signal. Unless active devices are used to regenerate the wave form within acceptable loss-distance, _very_ high quality (read air dielectric coax (hardline), or balanced twin lines) transmission lines are necessary. snip >[1] Before someone starts yelling collision domain, remember I said > 'full duplex'. There should be no collisions, and hence no > problematic late collisions. Thus, dual transmission lines, native to cat5. -- gt Everything here could be wrong--Messiah's Handbook--Bach -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
