Attenuation

Drakonblayde

Recently in class, we got into a debate regarding a question of attenuation. The question was a multiple choice asking basically which did not affect attenuation on copper media. Two of the available answers were immedietly obvious as being factors of attenuation, so were immedietly thrown out. The remaining choices were bad connectors and low signal frequencies.

I'll explain the debate in class, but first I'm curious what the rest of you think the proper answer should be and whether or not the same debate develops here

Webmaster

Difficult question...

Attenuation is typically greater at higher frequencies, so I tend to say the answer "low signal frequencies" is correct. But if the question is exactly "which did not affect attenuation on copper media"... using low signal frequencies means less attenuation than with higher frequencies so in that way low frequencies do affect attenuation.

So I think bad connectors is the correct answer. It is a weird answer anyway because in most cases the connectors are already attached to the media, or it is obvious which connectors should be used. (i.e. you won't be able to attach a BNC connector to an UTP cable, or RJ-45 to a coax cable. So I guess it would be very hard to 'use bad connectors', if bad connectors refers to 'wrong type of connectors'. In case bad connector refers to a 'broken' connector and 'affect attenuation' refers to an increase of attenuation the low frequency answer might be good afterall.

In case I got this question on the exam (with the above comment about wording in mind) I would choose bad connectors as the correct answer.

Sartan

Bad connectors will introduce something called near end cross talk, in which the wires on the rj45 will communicate with each other if they don't retain their twisted pair.

If it's a weak signal, however, outside white noise will also attenuate the cable. That's PSNEXT - power sum near end cross talk, or could also be classified as Far end cross talk - FEXT.

Not many cable testers are able to test PSNEXT. ANSI TIA/EIA does not yet have a standard for PSNEXT.

STP or ScTP cabling will prevent PSNEXT, and better connectors / crimping will prevent NEXT.

(Shielded twisted pair / Screened twisted pair ~ for more info on the diff just reply)

Drakonblayde

While it wasn't specifically defined, by bad connectors, we were all assuming RJ-45's that were untwisted too far, or crimped on the connector with too much wire showing, that kind of thing.

Webmaster hit the crux of the debate on the head. The teacher insisted that the answer was bad connectors, but the answer key said low signal frequencies.

The main reason I disagreed with the teacher (and consequently agreed with the answer key) is because, on Cat5 at least, an improperly installed connector will cause an impedence mismatch. That will cause attentuation by reflection, better known as jitter, and so therefore a bad connector will affect attenuation on copper media.

Unfortunetly, this is an instructor who dislikes being corrected, so I had to be careful in how far I pushed my conclusion. Personally, I think that's a really crappy question, and I hope to bog I won't ever see something like that on the real CCNA