Steve, you ask: >So, does xDSL work on a "loaded" circuit? <
No, xDSL lines also call for non-loaded cable pairs.
>Our PairGain and Tut equipment calls for "unloaded" circuits - no sealing current - <
I'm not following you on this. Are you saying that PAIR does not employ sealing current, or that they stipulate that it cannot be used?
>Also, isn't it true that one finds sealing current mostly between central offices and NOT from CO to customer premise on a normal voice circuit?<
Sealing current may be a stand-alone remedy, or it may exist in the form of loop powering for those lines that require power from the central office, such as T1 CSUs and DDS units. In this case, it has the primary purpose of powering the line terminating equipment, with the secondary effects being that of "sealing" otherwise questionable splices and punch-downs. To answer your question on where it is used, it used to be common among metallic T-1s between central offices, like you suggest, but since fiber systems have now displaced most inter-office trunks, it is probably seldom used in that environment. Where it *is* primarily used is in the local loop to the business or to the home where pair gain systems (using the latter term is its generic form) are used. Standard POTS lines do not call for the addition of sealing current because they are already "wet," since in their normal state they employ constant -48 Vdc, and occasional bursts of ringing current.
>If there is so much variability in line quality how are the carriers going to sell service?<
One popular approach is to make the xDSL units "rate-adaptive" to the line quality and length of the loop. With decreasing quality (or increasing length) the unit will fall back to lower speeds. Your next question is more interesting, tho, and I hadn't considered it before:
>Will I get a reduced rate for reduced performance or are they going to "average" it out? Based on what I've read distance is a big factor.<
I presume that you are referring to the subscriber charges here, when you refer to "reduced rate"? I have never heard of a formula that would address this issue, but I see your point. Maybe someone else can comment on this. I think that the LEC could rationalize a fixed rate, however, on the basis that even if your speed decreases with distance, you are being served by a longer and therefore more expensive section of loop. Anyone else on this?
>I won't quibble over a few 10Kb but ... <
I particularly like that comment, especially when I reflect that it wasn't long ago that 10 kbps was unachievable, in itself. Today we're talking about 10 kbps as a rounding off number. Jeeesh.
>How can performance be measured and who makes the tools that the carrier will test with?... Is there provision for in-band or out-of-band monitoring? Can the end user get the performance stats? <
Management capabilities are included in the embedded design of the field and central office units for the carriers' and ISPs' uses. I don't think that these are extendable to the user, yet. At least I haven't seen anything on this. I don't know of any xDSL-specific TE mfgrs but that would be a market to watch when one emerges. There is an almost standard lag time in deployments of next generation transmission technologies whereby testing and management platforms are afterthoughts, and can only be refined with initial use and experience with the platform. I've asked a similar question here in the past, but got no replies.
>Have tests shown that line quality can vary over given time periods (making measurement difficult)?<
With the exception of intermittent faults, or swinging conditions, typical variations occur gradually over long periods of time due to environmental factors such as temperature. These would only affect those services that were marginal, or hanging on threshold, to begin with.
Regards, Frank |
