T1 question

[ITDale]

This has probably been answered many times before but not where I've been able to find.

I've read too many articles on T1 specs but none of them seem to mention loop vs ground start signaling. All that I have seen seems to indicate that only applies to analog lines.

I'm asking because during a recent risk assessment our Adtran Atlas was identified as a potential risk due to its age and availability. (if it were to fail the recovery time would not be acceptable) My question is why do I need it? It is connected between an AT&T T1 providing DNIS with 2 trunk groups, 1 in, 1 out and a Magix D100 card. I recently acquired responsibility for this system and the setup documentation is poor. 1 comment from a coworker that couldn't remember the details was that AT&T didn't provide the right signaling for the D100 and they thought it was related to loop vs ground start. Is that possible? Is there a white paper or article that explains this type of carrier connection?

Thanks
Newbie Dale

 

[SYQUEST]

The DS1 is just a pipe. The individual channels are circuits that can be a line/trunk or a data circuit depending on what service is provisioned. In the case of loop/ground start signalling, that would be determined by the equipment that is connected or in the case of a Channel Bank, the individual channel unit type and options set.

But what I think you are asking is, can I eliminate the Adtran Atlas and connect the DS1 to the Magix? If you are not doing any data channel drops in the Atlas and the Magix can be equipped with a 100D module and the channels configured to match and function accordingly, then it should work for you.

You need to get a complete picture of the channel configs and see if you can set up a config on the Magix to match. The DS1 on the Magix does support E/M wink, ground/loop-start on a per channel basis. If you run into to something odd on the DS1 channel side, you may be able to have that service changed to something that is compatable with the Magix.

Whatever you decide to do, you need to plan accordingly and check out all specifications and limitations for your situation.

Hope this helps!

....JIM....

 

[dagwoodsystems]

First, my take on hardware risks: Solid-state electronics is one of those things that seem to either suffer from a DOA, else they hum along contently for 15 or 20 years. The Adtran Atlas is a very good piece of engineering. Anyone who surmises that the Atlas is a risk must also conclude that the Magix itself is a risk. My feeling is that the telco tech working in the local B-Box poses a higher danger to your circuit stability than your hardware does.

In any case, Adtran still makes a version of the Atlas. If you've got the old 500 series and you're concerned about failure, perhaps you could invest in a newer model and purchase and advanced replacement plan.

Or perhaps such a mux isn't needed at all. But like SYQUEST said, we would need to know a little more detail about your setup. At first glance, it sounds like the Adtran is doing nothing more than a signal conversion, which is a pretty hefty box just for that purpose.

You are correct in that ground-start and loop-start apply to analog lines. T1 circuits--as you know--are digital. However, there is a digital equivalent to loop-start signalling. This equivalent tends to be used when the T1 is destined to terminate on a mux designed to convert the digital signal to individual analog POTS lines on the downlink. I'm sure that some PBX's out there will accept a T1 (with digital loop-start signalling) directly into their DSX port, but the Magix isn't one of them.

What happens if you put a loop-start emulated T1 with DNIS directly into the Magix T1 port? The Call Appearance on the telephones will blink, but it won't ring. Aside from that annoying fact, I believe there is a problem with disconnecting the call. The proper way to bypass the Adtran would be to have the carrier change the channel signalling to wink-start.

Tim Alberstein

http://www.dagwoodsystems.com

 

[ITDale]

Thanks Tim and Jim
Hope it's appropriate in this forum to respond to both in one reply.
I'm much more of an ethernet guy and hope any analogies I make are understood. I'm starting to gain a limited understanding that the 24 DS0s in a DS1 can be used to carry much more diverse traffic than I had originally thought; interesting that the termination equipment seems (to me still) to be lacking standardization. But then as you said "a pipe" can have many different connection standards and carry quite an array of materials. I will contact AT&T and see if I can get a detailed specification sheet to get a better handle on what is being delivered. The Magix (2.0) is using a 100D, that's mainly why I've been having difficulty understanding what role the Adtran is playing.


===

My concern for the Adtran is less failure related and more programming. I am actually somewhat concerned about failure modes for both the Adtran and the Magix chassis. One failure mode we need to allow for is building un-availability (Fire - flood etc.) Having a T1 circuit redirected to a tempory facility or setting a "ring-to" on my 800 service isn't nearly as bad as trying to replace and reconfigure both pieces of equipment.

This comment really has my attention and is what I was looking for: "I'm sure that some PBX's out there will accept a T1 (with digital loop-start signalling) directly into their DSX port, but the Magix isn't one of them." That may well be the situation I have and my contact with AT&T Monday should confirm.

DSX - meaning Digital Cross Connect? - Type of DS1 connection?
Where can I find out more about these types of communication signals on T1s?

===

Thanks for the answers from both. The support techs I've worked with have been patient and answered some questions well but I think they know that the more they teach the less I'll rely on them. Although they might do themselves a favor,, I'm getting to the point of knowing enough to be really dangerous. Could be a lengthy engagement in their future ...

Thanks again
Dale

 

[dagwoodsystems]

Don't get me wrong. The Magix does quite well with T1 circuits...you just have to have the signalling set for wink or ground-start for proper operation. The 100D board will interface with either a PRI or a whole or fractional "vanilla" T1. You do not have substandard equipment.

I guess all that Jim and I are concerned about is the power of that Adtran. That box can suck up a DS3 (for example) and bust it out into a T1 interface (configured one way), and second T1 interface (configured a wildly different way), individual analog POTS lines and an Ethernet connection (as a terminating point for a PPP connection). That's what that box does; it channelizes, bonds and otherwise redirects different services out different physical interfaces. So I think that all we would like to know is what exactly comes in and what is going out.

I will admit that reconfiguring the Atlas is a bit of a PITA. The commands are a bit cryptic, but manuals are readily available. You should make a serial cable for it, jump in with HyperTerminal or ProComm Plus and get the existing config. You'll sleep better at night. And by the way, Adtran technical support is totally free. They can help you out in regards to extracting the config and making sense of this while thing.

Tim Alberstein

http://www.dagwoodsystems.com

 

[ITDale]

G' morning Tim
I didn't take your reply to mean you were dissing the system, just that it had limitations, don't they all.

I was able to dig up an old cut sheet that I'm pretty sure still applies to this T.
D channel: N/A
B channels: 6242-2544 thru 3725

Configuration:
Channels
Frame: ESF
Line coding: B8ZS
Signaling: Robbed Bit
Supervision 4 wire E&M
Signaling IN & Out: DTMF/WINK

AT&T is sending DNIS

I know the Atlas is acting as a CSU because we have "repartitioned" the trunks twice and Atlas programming has had to be changed. AT&T has worked with a programmer both times to assign specific channels to the inbound and outbound trunk groups defined on the Atlas.

I will have to pull a system report off my laptop to see if I can decipher the port and trunk config as it exists on the Magix.

Thanks for all your input
Dale

 

[dagwoodsystems]

It sounds like what we call a "drop and insert" app. Basically, you've got a straight T1 where some channels are used for voice (and passed through to your Magix 100D card), and the other channels are used for something else.

That "something else" could be Internet access, or it could represent an endpoint of some point-to-point circuit.

I doubt that the voice trunks are separated by direction. That's really old school thinking. Most likely, whatever trunks are assigned as voice are bi-directional.

The only part about the report that bugs is the mention of B-Channels. A B-Channel (or Bearer Channel) is language that's pretty much used exclusively with ISDN circuits, such as a PRI. And every PRI--by definition--must have a D-Channel. Yet the report indicates that the D-Channel is "N/A".

On the other hand, the circuit apparently uses Robbed Bit signalling, which is clearly an indication of a straight T1 and not a PRI. Does the "6242-2544 thru 3725" mean anything to you? I'm wondering if these are the DNIS digits that the carrier is sending.

Tim Alberstein

http://www.dagwoodsystems.com