Hi Guys,

I have AT&T installing a T1 (PRI) for customer. The original install was done by a tech who just slammed it in and left without telling anyone he came. It was installed in the MPOE and the customer is on the 8th floor about 700 ft away. I spoke to the AT&T test board who told me the distance was to great to work with a good consistent signal, so I requested AT&T to move the Shelf and smart card. It's been 8 months now (my last AT&T T1 install took them over 2 years and they still botched it) and they still haven't found anyone who knows how to do this. They know how to install a jack (I can do that) but that's it. One problem is terminology- the terms used by different techs have been- "Shelf, Smart Card, NI card" Which is correct? I believe NI and smart card are the same and they plug into a shelf. Is this correct?

Now after 8 months the latest tech says that no one can move this card. It goes in the MPOE even if customer is a mile away. If the distance is too great, this is the customer's problem and customer needs to purchase a repeater or signal booster. This is not the same phone company I worked for. Does anyone know the distance limitation between the smart card and customer jack and if this smart card (or whatever the correct term is) can be moved?

Thank you

655 feet maximum with proper LBO settings on the card.

Yes, it can be moved by the provider of the circuit by just extending the street side pair(s) via house cable to the location. They just don't want to do that because it limits their access to it.

Ed is right on the money for proper distance limitation for T-1/PRI.
But I will go on to say this, a little over a year ago we extended one that was a little over 1400ft (Yes 1400ft). And it has worked like a charm from day one without any problems.
So you may want to do some testing and see what happens.

Jjimm,
If you could imagine this the telco will do it if you order it and pay them… It’s just a simple IW (inside wiring) request. Not a clue what at&t chares in your area but I’d be sure it’s expensive.

Ed’s number of 655ft is spot on but if I were you, I’d give Mike’s suggestion a try first as well. I’ve seen those NI cards reach way beyond their expected reach more than once. You’d just need to be sure LBO (line build out) is set to the max.

Your thinking is correct… but those terms and many others are often used interchangeably.

JJimm -

I'm with the guys here. You might be able to pull this out. Set the LBO to Max and then the install: Run two shielded Cat 3 cables (one for xmit, one for rcv) from the Smart jack and terminate them on an RJ48x. Then test the circuit with a T-Berd or Sunset. Do a thorough test (5 minutes of all 1's, 15 minutes of all 0's, 15 minutes of 3-24 and then 24 hours of QRS. If you get less then 5 errors I would say you'd be good to go.

If it doesn't work, then you're going to have get a repeater - but worry about that if and when you have to.

Sam

Thanks guys. I realise it might work -most times, but, 1. it's a gamble, 2. customer wants to be assured of constant good service (at least feel like he has this)3. after all the hassle with AT&T about this it would be better business for me to just move it, and 4, customer is a lawyer and is irate at AT&T.
I have a new sub who is PRI experienced. If he knows how, is there any legal reason why we can't just move it ourselves? If absolutely necessary I can probably get a release from AT&t. I'm sure they would be glad to get customer off their back.

Yeah, that’s not a legal maneuver. The NCTE (network channel terminating equipment, commonly called the smart jack) is telco property and serves as the demarcation point for the regulated service. We can’t advise you to move it.

If I were in your position I’d get a LoA (letter of agency) from your customer and put in a service order yourself with at&t to request a demarc move with inside wiring charges approved, to have the point of demarcation moved to inside your customers suite. Could be pricey but would be the best and most reliable way.

That was the official answer… Sometimes smart jacks do unexplainably move, but you now know that’s an illegal solution.

I wouldn't advise it either. Your best bet is to extend the customer output via pairs in two separate cables (one for transmit and one for receive). If using existing house cable, use pairs in separate binders. This will allow you to get more distance and reliability.

AT&T VP just called to cover his *ss and said "AT&T only provides the NI at the MPOE. From there it is the customer's responsibility no matter what the distance." He had no grasp of how this circuit works and was repeating what someone had told him.
I offered to have my tech with T1 expertise move the card but want a release so that AT&T still supports the circuit, He said "No problem". I don't trust them. Customer is going to have to take his chances. It's unfortunate customers reduced to taking what they can get from AT&T, but the company has lost most of it's trained professionals.
Thanks for al the help.

For anything over 655 ft., and for something that is not a gamble, I would install T-1 extender/repeater. I have a customer that have been using some with no problems on 4 circuits in a call center that were about 1500 ft from at&t's MPOE. We get ours from Data Comm for Business. www.dcbnet.com

A strong DS1 pulse transmitted from a service provider NIU toward the CSU/CPE will generally travel quiet a distance on healthy copper. Usually well over 1000ft. Most ILEC's by policy regen the DS1 pulse from the facility side of the NIU and transmit to the customer at 0 to 0.5 -dbsx/6vpp. The only LBO setting I have seen on any NIU/HRU in the last 10years is a transmit LBO toward the CPE/CSU from the NIU with generally 3 to 4 settings: 0 dbsx, -7.5dbsx, -15 dbsx, and -22.5 dbsx. Each LBO setting is referenced with a guesstimated cable footage with the highest usually being 655ft. Those cable footage estimates were based on old/ancient topology and really don't come anywhere close to how far DS1 will travel on CAT3 or CAT5 cabling.

Software settings & tolerances can vary between CPE/CSU manufacturers also. Typically a DS1 pulse can be recovered/regenerated on a T1 span in the OSP every 3,000 to 6,000ft between repeaters(up to -28.0 -dbsx), depending on cable gauge. I have seen integrated CSU's on PRI equipment that would not achieve DS1 sync with the ILEC's NIU with only -15dbsx atten on the receive pulse, either because that CSU was not designed to accept an overly attenuated DS1 pulse or it was not software set to regen the pulse before converting it to unipolar, or it had no such setting. The CSU can also be incorrectly set to pad(LBO) the DS1 pulse toward the providers NIU. I have seen many tech's on both sides of the fence incorrectly set LBO based on cable footage between NIU's & CSU's not realizing that what they were doing was causing the problem.

Not having the proper test equipment is a problem when it comes to DS1/T1. I would not be comfortable extending a DS1 across house cable,iw, etc. without having a good copper analyzer(Fluke,Sidekick,Dynatel) & a T-BERD or it's equivalent handy. The Fluke & Dynatel sets are usually optioned for wideband capabilities & you can do single ended testing with there respective far-end devices. Widebanding the copper pair at 772Khz(center freq. of baseband bipolar DS1) will give you pretty much the exact amount of attenuation you can expect on the DS1 pulse between devices.

Glad to see this post. Here is a scenario that I have for a customer who often looses his T1 lines at random intervals, although the restoration process sometime takes days. They have two T1 circuits, one is for a PRI and the other for their internet access. They have a number of POTS lines, just so their calls can be forwarded to them when the PRI T1 goes down. the T1's are from AT&T and transported by Verizon, albeit the old GTE territory before the acquisition by VZ sometime back

There is a 24/25 pair cable which feeds from the DMARC to the EU server Room
T1 CKTID: 23.HCGS.XXXXXX (Data) SMJK Shelf 1, Slot 6 is located on pairs 16 and 25 of the 25 pair cable
T1 CKTID: 23.HCGS.ZZZZZZ (PRI) SMJK Shelf 1, Slot 2 is located on pairs 13 and 23 of the 25 pair cable. The other working pairs are as follows:
POTS is working on pair 1
POTS is working on pair 2
POTS is working on pair 3
POTS is working on pair 4
Nothing is working on pair 5
DSL is working on pair 6
POTS is working on pair 7
POTS is working on pair 8
POTS is working on pair 9
POTS is working on pair 10
All other pairs in this cable are spare, non working
The approximate distance from the DMARC to the EU server room as measured by a TDR 1s 326' (Opens test)
The approximate distance from the DMARC to the EU server room as measured by a 965 Dynatel is 376' (again an opens test)
Random vacant pairs were selected to test for metallic faults with the two above mentioned test sets as well as with a Sidekick test set.
Pairs 11, 12, 14, 15, 17, 18, 19, 22 and 24 were tested and showed no apparent physical faults.
A Strap was placed on pair 17, which should have shown a SC (short circuit) Fault at approximately 336 - 376' +/-. The actual DTF (distance to fault) was measured as 642' with the Dynatel
A Fault locate test shows a Strap to fault reading of -333' and the DTF of 654' on pair 17 (Dynatel)
Placing straps on pairs 12, 14, 22 and 24 give a reading of 791' each with the TDR -s/b 326' STF
Pair 18 shows and DTF of 321', DTF of 46' and STF of 274 ' The faults are rather light but noticeable..

The apparent problem is that the 2 T1 circuits which are in the 25 pair cable are causing interference with each other, possibly the DSL service (or vice versa) and the vacant pairs due to the phenomenon known as signal coupling.

My recommendation to this customer is that the T1 circuits be placed in an Individually Shielded Twisted Pair (ISTP) 22AWG T1 cable. I plan to terminate the new wiring on RJ48C jacks at the DMARC end and RJ48X jacks at the EU end. I try to stay away from terminating IW in 8P8C pugs whenever I can avoid it - been on too many trouble calls where the IW cabling had pulled loose from the 8P8C contacts.

Silver Sam recommends testing the cabling with a T-BERD and I can do that, although I don't know if I can run a 24 hour test with the QRSS pattern for the two new cables. If I can, it would mean a return trip to the EU's premises Of course, I can just build that return trip cost into my proposal, can't I?........if they would accept that, that is.

Is there anything that you think I have overlooked or do you have any other recommendations for me?

Thanks to all of you on this forum.

Right off the bat I’m confused... I’m reading this as the failings are on the VZ side of the demarc, yet your resolution suggestion is to re-cable demarc to end-user server room? Am I misreading something?


Do you know or make note of what type of cards are in Slot 2 and Slot 6 for the two DS1 circuits? Depending on the technology used to deploy the circuits over copper to your customers premise, there is a possibility of issues there. I’m reluctant to say that will fix your issues but there are technolgies that do not play nice with each other.