Topic: Do you only have two wire connected from the Telco?

[CnGracin]

Is there something wrong when you only see two wires connected to your NID (Network interface device, aka Smart Jack) from the telephone companies T1 or DDS (digital data service) circuits? Nope not at all.

So how is it that a four-wire digital connection works on 2-wires then? There have been technologies developed for both DS1 circuits and DS0 DDS services that allow for the Telco to deliver to the end-user/customer over one copper pair. Total Reach for DS0 and HDSL2 for DS1 were both pioneered by Adtran and are widely deployed by Telco’s today.

Only one-pair (two-wires) coming in from the Telco is nothing to worry about and isn’t something to worry about when/if installing or shooting trouble on a digital circuit. 

[MacGyver]

I'm really glad you posted that Bryan.  I've been trying to understand that one for a while.  At three of my locations including my house I have a full T-1 all from the same provider terminating at an Adtran.  Then one Christmas night I believe, we had one go down and at about 2 a.m. the tech said he switched it to a two wire circuit.  He idd it because it was a holiday I thought and wanted to just get home.  That really concerned me.  To  me a T-1 was by definition 4 wires.  One pair for transmit and one pair for receive.  To me this was more of DSL.  

Last week the provider upgraded the circuit at one of my other locations using that same provider and equipment and I asked if it was one or two pair.  The tech said they replaced it with another 4 wire circuit just like the one that was already connected.

I do notice a little more overhead on the 2 wire circuit, but I'm not sure that's the provider's issue.  I think I'm probably running more VPN tunnels out of that location.

[CnGracin]

Then one Christmas night I believe, we had one go down and at about 2 a.m. the tech said he switched it to a two wire circuit.  He idd it because it was a holiday I thought and wanted to just get home.

Maintenance cut-over’s on Special Service Circuits typically are done on nights, weekends and holidays. Since nearly all these types of services are for business, the thinking is the majority of people are at home. Most people aren’t like you and have a T1 to the house.   

To me a T-1 was by definition 4 wires.  One pair for transmit and one pair for receive.  To me this was more of DSL. 

A T1 IS a 4-wire circuit. There’s a term we use on this side of the demarc of “effective 4-wire” that used in these cases. Some really smart folks have figured out to make these digital signals full-duplex over the 2-wires.

The backbone of HDSL (Hi-bit-rate Digital Subscriber Line) technology is a 2B1Q (2 binary, 1 quaternary) line coding. It confounds me a bit to even think about it at the bit level, but to put in layman’s terms they figured out a way squash the binary data stream in a very compressed form and then uncompress it at the other end. This greatly reduces the attenuation loss over the copper pairs. Allowing for longer reach over the copper as well as full-duplex over just one-pair.

I do notice a little more overhead on the 2 wire circuit, but I'm not sure that's the provider's issue.  I think I'm probably running more VPN tunnels out of that location.

There’s absolutely no difference in bandwidth on a HDSL2 (2-wire) T1…. Conventional T1 (DSX-1,) HDSL (1st generation,) HDSL2 (2-wire,) and HDSL4 (4-wire, current generation) are all 1.544 Mb/s end-to-end all day every day. (Well, unless some poor tech is cutting it over on Christmas.  )

Due to the way these circuits are timed and error checked anything less would cause the circuit to go down hard.




Another advantage of 2-wire HDSL delivery no one very often considers is the less risk of failure. Copper pairs can and do go bad on rare occasions. Talk to splicers and it’s always in the middle of the night in a snow storm  but anyway… One less pair (2 less conductors) on a 2-wire circuit is one less pair with a probability of failure.

[MacGyver]

So if I'm understanding you correctly Bryan, what you're saying is that even though we're billed for a Full T-1 Circuit, you see no downside to one of these locations being run across a single pair as opposed to a 4 wire drop.  Is that correct?

[CnGracin]

Correct… As long as the loop-length is within allowable range for HDSL2 to work, there’s not only NO downside to using it… There’s an extra upside gained with 2-wire, in that there is less copper to potentially fail.

And btw, the billing for 4-wire again goes back to the “effective 4-wire” terminology. There’s a four-wire interface on each end regardless of the last mile copper delivery…. Think about what happens to your hi-cap circuit as it gets shoved up on an OC-48 for interoffice transport along with 1,343 other DS1’s. That could be over just two fiber strands. There will never be a end-to-end separate transmit and receive path belonging to just one T1’s.

[MacGyver]

There will never be a end-to-end separate transmit and receive path belonging to just one T1’s.

Good point.  I never really thought about it that way.   

[MrGemini]

Bryan, probably not the correct terminology but I've heard the 2 wire T1s referred to as "high voltage" T1s.
One thing I found out the hard way is that you don't want to have a 2 wire T1 in the same bundle as DSL Ckts.
It plays hell on the DSL. 4 wire T1 didn't seem to effect it.
Why is that?
----------------------------
Thanks, C I K

[CnGracin]

Hey there Wayne and thanks for the follow up question  

You’re actually on the right track but with the terminology but slightly askew… “High Voltage” on HDSL would actually be referring to HDSL (of any flavor) loops that have repeaters in them. The extra voltage is present to power the line repeaters along the spans.

Repeatered HDSL (1st generation) and HDSL2 equipment is no longer available and should not be deployed any longer. Why is that? It messes with ADSL signals in the same binders.   The signal format for HDSL and HDSL2, ended up being too close in spectrum to a ADSL signal and when pushed up to the levels that repeater power does. It would knock down ADSL in the same binders, especially if the loops were a bit on the long side.

There’s no concern with putting ADSL and non-repeatered HDSL2 in the same pair count.

[MacGyver]

So it has to do with the frequency spectrum and not the voltage.  Correct?

For instance we recently went out on a site where trunks were in the same conduit as an Inter-tel Premier station lines.  Whenever ring current came in over the trunk, it would choke out the station line and drop the call, but that was voltage related.  If I'm understanding you correctly, the frequency spectrum of the SDSL lines *terminology* will alter that of the ADSL.  Am I getting it or did I just take a left turn at Albuquerque?

[CnGracin]

So it has to do with the frequency spectrum and not the voltage.  Correct?

Not ‘sactly… Seems the two (HDSL signal and repeater power) have to go hand –n- hand to effect ADSL.

ADSL will work in the same count as HDSL2, no worries.

ADSL will work in the same count as other digital formatted signals that require being repeatered, no worries.

When it was attempted to use repeaters on HDSL2, it would knock down ADSL loops in the same binders. Especially if the ADSL signal was marginal due to longer loops. I’m in the arena of an “edumacated guess” now, but I’m assuming there was a shift in relative frequency due the two voltages (AC payload and DC power) being superimposed on one another that was the cause.

Am I getting it or did I just take a left turn at Albuquerque?

Trace, I think you zig-ed when ya should have zag-ed. 

[MacGyver]

Trace, I think you zig-ed when ya should have zag-ed. 

AH!  Ok. 

You're correct.  I did.  Somehow I missed the part about the repeaters.   

Thanks Bryan.