atcomsystems.ca/forum Forums Wiring - Telecom Carriers - Trunks & Services T1's, CSU's and DSU's Extending AT&T T1 Demark

There isn’t a hard and fast rule that it HAS to be something to do conditioning there Bill… If the loop lengths are short enough to the exchange (or remote) and the signal is DSX1 format as it get’s to the C.O; then it technically CAN be put out JUST on cable pairs and termed to a 66-block or biscuit jack (as long as it’s an 8-pin.) That being said… I can’t see WHY anyone would not install some kind of NCTE (Network Channel Terminating Equipment) on a special circuit. The loss of not having the conditioning, isolation and most importantly “remote test access” at the customer premise does not make any sense to me. I’ll ALWAYS spec a NCTE device of some kind on all circuits I design.

There’s also a possibility in a multi-tenant building that a terminating device might be located somewhere else besides your customers space. In other words just because you do not SEE a smart-jack doesn’t necessarily mean it’s not there.

Bill is correct that the lowest pair in the count SHOULD be the receive from the customer location (overall transmit) … One trick I learned way back when I was working as a field installer is this… Take your butt-set and clip to each pair. You SHOULD be able to hear a VERY HIGH frequency squeal or ringing on the pair you should be receiving on.

if I can summarize my understanding of all the above posts:

1. Use an RJ48X since it has a shorting bar to avoid the need for a loopback plug.

2. RJ48X's are wired as follows:
Pins 1/2 -transmit from CO to CPE; same as receive at the CPE from the CO. This is usually the top pair of the punch block

Pins 4/5 -transmit from CPE to CO; same as receive at CO from CPE.

3. Hook your butt set to a set of pairs and when you hear a high pitched sound, that confirms this is the receive pair, which will go to pins 1/2.

Is everything correct?

Looks like you got it to me Steve.

Bryan, thanks for the explanation. Guess I never had one that close to the source. I agree with everything you said about why there should be an NCTE on it.

It may be an HDSL circuit which only uses one pair. If it is, then you have to use a smart jack before it can be passed off to you. The HRU (smart jack) changes it from 1 pair back to 2 pair. Bad thing about HDSL is it runs somewhere around 187 volts versus 120+ for LDSL circuits, so it degrades the copper faster. Got that one from a 30 year BS DLC tech. And yes AT&T is getting bad, not from laziness so much as pressure from their management. They have cut their install and repair times so the techs have to cut corners and they are pushing for more cost savings. I have several acquaintances who are planning on retiring very soon from the "new" AT&T because of all the issues.

I heard that most T1s are now being delivered via 2 wire HDSL because it's cheaper? (btw- what protocol did they use before?)

Does HDSL use something called PairGain?

The cheapest T1 tester is a LED. Place the LED on pins 1 and 2 and if it lights up that is your RECEIVE. If they demarc on a R66 or BIX again put the LED on a pair and if it light up thats your RECEIVE (ATT Transmit)

I heard that most T1s are now being delivered via 2 wire HDSL ... TRUE!

because it's cheaper? Yes and no... The saving of one pair (over 4-wire) is the biggest gain... which can be a $$$ savings to us Telco’s. Honestly the BIGGER reason is 2-wire can actually be MORE resilient.

btw- what protocol did they use before?
(I'm assuming you mean to deliver T1 signal)

(in chronological order of being implemented)
* DSX1 format … This is true T1 signal. Rock solid but didn’t reach anywhere for crap.
* HDSL… 4-wire HDSL with a transmit and receive.
* HDSL2 … 2-wire and actually LONGER reaching than the 1st generation of HDSL and should not be repeatered.
* HDSL4 … 4-wire that uses similar technology to HDSL2 except it splits the bandwidth to two full-duplex paths between the two pairs… and it CAN be repeatered.

CnGRacin

Say you're saying that 2 wire HDSL, which is supposed to be used for T1 delivery, doesn't use repeaters? I thought all T1 signaling is repeated? Which is what makes it different from ADSL and more available than ADSL?

That’s what I was saying, yes… There were at one time repeaters made for 2-wire HDSL (HDSL2._ It was discovered that they caused interference with ADSL signal within the same binder group. Once ADSL got more prevalent the idea of puttering repeaters HDSL2 was abandon. (or at least SHOULD have been)

Since even the 1st generation of HDSL has been deployed there are a lot more T1 services being delivered over copper that are non-repeatered. Once upon a time before HDSL a T1 signal couldn’t make it more than about a mile in perfect conditions so nearly all T1’s were repeatered but that’s no longer the case.

There are many similarities between ADSL and HDSL but they are really two different animals. HDSL is merely a technology to reformat a DS1/T1/E1 signal for more efficient transport over copper pairs. The T1/E1 is still the service being delivered; Which is sinked and timed at 1.544 Kb per second, no more, no less. ADSL is a service offering that connects an end-user to a DSLAM (Digital Service Loop Access Module) that ultimately connects to an ISP (inter-net service provider.)

Good question, local_host.