What is the farthest you can run a t1 out of the smart jack on house c3 cable before a repeater of some sort is needed?

About as far as I've gone is around 300 feet. this link says 655 feet with 22 gauge.

Standard Line Build Outs for T-1s on the T-3 Mux's were:

1-133'
133-266'
266-399'
399-533'
533-655'

After 655' you needed repeaters. There were CO repeaters (130 Volt) that would push a T-1 for "A Country Mile" (about 6,000'). There were also "Short Haul" Repeaters designed for inside buildings that ran on 48V. Depending where you put them in the run (the middle or one of the ends) you could get 2,000-4,000' on a run.


Sam

I have read that the max 655 foot line build out is to the cross connect patch panel or frame within a building and then you are allowed an 85 foot cross connect. With that in mind the line build out from a smart jack to a cross connect patch panel (DSX-1) would be 655 feet, then the line build out from the the CSU/DSU to the cross connect patch would also be 655 feet. Add in the maximum of 85 foot of cross connect wire to connect the smart jack to the CSU/DSU and you get a max distance of 1395 feet.

I wouldn't use CAT3 for a PRI-- That really is made for DDS. You should use CAT5.

The 655 Foot rule is for the interconnection of line interface unit (LIU) DS1 modules when terminating at a DSX-1 cross-connect frame. Repeated T1 spans are two metallic wire twisted pairs with regeneration typically at 36dB insertion loss points based on 772KHz NYQUIST (one half the T1 line rate), and Roughly at 6000 Feet, or depending on AWG used.

Cat 3 or CAT 6 twisted pair can be used, as long as transmit & receive are run in seperate cables to reduce the incident of cross-talk errors when T1 signal losses are 15dB or greater between TX & RX.

All Verizon T1 NIU's, including all HDSL HTU-R remote units provide T1 signal regeneration on the input side of the unit.

The original Vz RFP requirement was that the NIU/HDSL Remote must recover a DS1 signal down -22.5 dBdsx. Most units will recover down to at least a -30.0 dBdsx value.

So extending a DS1 out to about 3 or 4KFT using 24AWG shold work. A Head to Head Line loss & BER test measurements will validate the T1 span design.

The T1 level output from the NIU/HTU-R however may vary based on region (Formet BA/GTE), and device type and will influence how long a T1 extension can be built before additional regeneration is required.

Vz's deployed Westell 6170 series T1 NIU's (Smartjacks) up until January 2007 were shipped from the factory with a 7.5 LBO option ON by default. The Pairgain (Now ADC) HDSL2 remote had the same 7.5 LBO output default option.

I can't recall if the Adtran HDSL-2/4 incorporated the same (I was not on that RFP when it took place in 2001/2002).

All Westell T1 NIU's deliverd since about January '07 now have the 7.5dB option turned OFF, but available if the CPE side installation requires it. The same 0 LBO output rule change applied to HDSL equipment.

The above default LBO change was my last engineering recomendation at Vz, and a reversal to a recomendion made during my NYNEX days, and implemented in '98 when BA rolled into town.

Field OPS

hey field ops-you're right. I have a cust. pushing 3000 ft from demarc smartjk to a carrier access device channel bank-I have a new clec that says their iax (cisco) has to be within 333 ft of(same) smartjk-Don't you think due to the standards you mentioned, all outputs would be to that spec, not an ethernet spec and if the smart jk can push it any complying equip would recieve it. I keep telling the clec engineers they need to order it right from the telco, which I don't know what that term would be and I worked for Vz for 30. In my day not too long ago there was dip switchs to adjust the output. Are they allowed to crank it up? Do you order it as +7.5. Full power Scotty....

To further my previous question, A T1 to the FCC is the same if its data, voice channels, PRI or combo, and all equipment for use on this spec is about whether it gets the ride in as DB's, and the smartjks have an automatic recovery back to 0 from -22 down and they can be optioned for 7.5 more? Damnit Jim, I am just a tech, not an engineer.Help

ndt4u

There are two basic DS1 circuit terminations that you remain cognizant of. The most typical DS1 termination is found at end user customer locations where the LEC (Service Provider) provides a T1 line termination at a Network Interface (NI).

Typically a tariffed offering for just about all hi-cap T1 services. Regardless of frame format (SF/ESF), or line code (AMI/B8ZS Option), and typically cited in PSC filings as being in compliance with ANSI T1.403.

The ANSI-T1.403 compliant interface is essentially the old fashion T1 span that at one time terminated in a LEC provided CSU. The CSU was the last line repeater in the design of legacy T1 span repeateed metallic 4W lines.

The CSU, just like it s counterpart in the Central Office , the Span Powering T1 Office Repeater (ORB), provided an interface between the outside metallic 4W plant, and terminal equipment.

Both the T1 ORB and T1 CSU provide signal regeneration in the receive direction (From the OSP cable), provide line build out to/from the OSP cable, and more critically, provide DS1 pre-equalization towards the DSX-1.

Equipment designed to operate at the DSX-1 typically will not regenerate a DS1 signal. The purpose of the DSX-1 is to engineer all DS1 equipment to reach the DSX at a specified signal value of 6 volts Peak-to-Peak (6v P-P) 3 volts base to peak signal, with an allowable deviation of =/- 1.2v from the 6v value.

Equipment designed for a standard DSX-1 interface is engineered to operate no more than 655 feet from the DSX-1 when using 22AWG shielded ABAM cable. As the cable gauge gets smaller, the distance between the DSX-1 and equipment gets shorter.

The problem that I have seen is where a channel bank equipped with a LIU module, as opposed to a CSU module, is connected to the t1.403 compliant T1 line, and where the T1 NIU (Smartjack), and CPE are not collocated. Since the LIU module is expected to be within 655 feet of the DSX-1, the signal coming in from the T1 NIU may be too low in signal amplitude to be detected by the channel bank's LIU. This assumes that 22AWG cable is used on the CPE I/W side.

Going the other way to the network is no problem, since the T1 NIU will regenerate a DS1 signal down to 30dBdsx.

Assuming that the you local LEC is using T1 NIU's that output a 0.0dBdsx, and if you are using 24AWG CAT5/6 cable, the maximum distance between the T1 NIU and a channel bank equipped with a LIU module will be about 1/3 less in distance than if you had used 22AWG cable. 24AWG, about 420 feet max. If the LIU can recover a DS1 signal lower than -3.5dB , then you may be able to go out farther on cable.

When IXc Carriers & CLEC's order a T1 access line from an ILEC, the ICX/CLEC use Network Channel (NC) Codes. The NC code defines the DS1 channel type. SF/ESF, AMI/B8ZS. Along with the NC code there is something known as a Network Channel Interface (NCI) code. This also specifies frame format & line code at each T1 interface.

End user customer locations that terminate in ANSI T1.403 compliant interfaces use a 04DU9 NCI codes (Requires a T1 CSU). DS1 interfaces that terminate in DSX-1 interfaces are assigned 04DS9 NCI codes (No T1 CSU required).

Equipment designed to connect to a 04DU9 interface (T1.403), does not mate very well with equipment designed to interface 04DS9 interfaces.

Sorry about the very long winded reply. But if you can remember the differences between equipment based on interface requirements, you will be able to quickly identify a design flaw that may surface under the right conditions.

Experience has shown that PBX trunk card line interfaces units are also not immune from the same design limitations as the channel bank you make mention of. Hence the requirement for an external T1 CSU, or CSU module, to regenerate the LEC DS1 receive signal towards the CSR terminal equipment.

Field Ops

PS: CLECS are notorious for using channel bank (For example: Adtran Total-Reach 750) equipment that does not regenerate the LEC receive signal when equipped with a plain old vanilla DSX-1 LIU module. as for DS1 (T1's) in general, payload is payload. Regardless of what type of information a T1 carries, voice or Data (To a DS1 line payload it all data) the rules for T1 design remain the same.

Filed Ops

Thanks.I had seen all the numbers before but did not see how they fit together. There was a time when all T1's had a CSU. I never got a good answer as to why they were needed because we would by-pass them on trouble when they were bad and the switch would come up.

When I worked on T1 cable spans they had the 130v on them and nothing to speak of on the cust side. Does that mean the early ones had 130v going into CSU or was it always stripped out before hand-off to the CSU?

The key is the code. Let me see if I get it.

Most orders these days are for 04ds9 NCI. That explain why you don't see CSU's much anymore. Assuming this is what my CLEC orders, he falls in the 655' category. If I tried to order it as 04du9, it would allow for a CSU but may not mate well with 04sds9 interfaces.

I'll go back to them to see what they order.

That may be why the extenders are on the market. They may comply in the latter.

While we are not on this subject, Do you know the difference between CO fail: TIA-547a and TR62411.
Does that have anything to do with CO equip/testers to access to testing, maybe past the CSU. I have been leaving this set to the former for years on Norstars.

To further explain this T1, they call it Dynamic.They have a Cisco IAX that acts as a channel bank to strip out the FAX channels/lines. The rest is voip voice and data, which I guess is all data to the T1. I am guessing a LIU is all it will have. Any suggestions for 3000' mixed 24/22. As you know, we can't get the telco to move past the demarc anymore. This is on a horse farm, and has a private 100pr the last half mile. I hate to add an additional point of failure like an extension kit.

I’m going to go backwards a lil’ bit on Field Ops explanation of NCI coding…

In the “protocol code filed” of the NCI code (position 3&4) there’s a possibility of common the “DS” and “DU.” “DS” literally is translated from Telcodia’s Carrier Coding Guide in to “Digital Hierarchy Interface” I always further translate that closer to English as a MUX, channel back etc. In other words, something that takes a signal up or down the digital hierarchal levels. Now, “DU” literally translates per Telcordia to “Digital Access Interface” which I always equate to something of an “access point” to the Telco’s digital network. A place for an end-user/customer to plug-in to the network, to over simplify a little bit for the sake of explanation.

The NCI I see the most at end-user location by far and away is: 04DU9.1SN
4-wire interface … DU= Digital interface (see above) … 9=100ohm IMPEDANCE … 1SN=1.544 Mb, ESF/B8ZS, "no line-power."

Unless something has changed they are a require interface. Some switches have a CSU built into their T-1 pack. I haven't put in a T-1 in a very long time so this requirement may have changed, but if so I haven't heard it did.

CnGRacin.

You are correct with regards to the DS NCI protocol of DS. It is generally found at the output of a MUX, since a MUX low speed card, like the channel bank or PBX LIU, is designed for a DSX-1 type interface with a fixed ZERO dB loss value.

DS NCI codes are typically found at an IC or CLEC POP location.

The DU protocol is used on about all traditional digital interfaces that represent a customer hand-off with no fixed loss values.

For example, 04DU5 is the base NCI code for DDS services. The DU NCI protocols signifies a repeated digital line interface that is engineered by the LEC not exceed a particular loss depending on service.

Some common examples;

DDS:

No more than 32dB of loss at the NI measured @ Nyquist (The DDS Data rate divided by two) from the LEC. ZERO dB from the DDS CSU.

DS1:

No more than 16.5dB towards the CPE from the LEC at the NI, and no more than 5dB of loss from the CPE towards the LEC at the NI.

Just Bill,

I agree, PBX trunk cards do include a CSU designation on card version. However I have also discovered that in some Lucent PBX interface modules the trunk card will not regenerate the T1 signal into the card, and requires the use of an external T1 CSU to accomplish the regen task.

The above PBX cards will typically respond to a T1 CSU loopback to perform testing, but that's all.

The current line of HDSL & T1 NIU's being deployed today by most LEC's will resolve any of the regen issues I have seen in the past. Unless the PBX is situated far away from the LEC NI equipment.

If the magic was too easy, then there would be no need for technicians to do any work!!!

Good Telephone Karma for all.

The Inter-Tel systems I use do not require an external CSU/DSU. It is built into the T1 card.

Come to think of it so do the Norstars(built in CSU).I had thought is was for testing, but it does have a line build out.

This Cisco IAX is built for Nuvox, the CLEC may not have any regen, although I hear the Call Manager switch has an internal CSU

NuVox uses the Cisco 2431 & 2432 series IADs. It's a typical channelbank, except the whole T1 is setup for data and voice calls are sent over IP.

You're looking at maximum of 655 feet from the ILEC DEMARC, just like virtually every other channelbank.

You're going to need a repeater or T1 extender if the install will be 3Kft from the ILEC DEMARC. Adtran makes a device to extend a T1, and requires one on each end of the extension. I know a technician who has a T1 extended close to a mile with this setup.

Thanks Kyle, something the cust won't be glad to hear because it isn't in their quote. The one in place is going to a Carrier Access Device (Adelphia/Telcove)channel bank, and it does not use one(extender) but it must have the ability to regenerate the signal. I guess from wht you say I won't have that same luxury with the Cisco.

NuVox should install their equipment a few days before the actual cutover. They call it a plant test date. Have a DEMARC extension in place for the technician to connect to. Give it a try and see what happens. The worst case scenario is you push out your service activation date a week or so while you order extenders.

That's what I was thinking. I know they hate that, but I told the salesman it was 3k, just in case it mattered.This is Windstream, former Verizon, former GTE where this is going in. If I get a tech from one of the formers, they will try any settings they can to help. The new local is more "Bell like".(this is what you ordered, this is what you get, if you don't like it, you need to re-order.) I am going to try to be there for their install date too.

Splitting Atoms.
As Filed ops provided, there are ways to order that could give me the boost, but yet be incompatible :shrug:

Update: on the Nuvox pre-test the system linked up on the 2700ft line and was measured by their equipment rec@ -8db. We were going to change it to long(the 655' setting-it is at default of the short 133ft) but that looked like it was going to add 36db so we left it as is. It's collecting error free but the real cut is not complete. So far so good.

Still error free and using it for 3 weeks-2700' extension

We just turned a voice T-1 up in Wilmore KY that is 1500 feet from smart jack to PBX.
Working like a charm for a week now.

Michael,
You didn't happen to make note on the manufacturer and model of the NID card (aka smart-jack) did you?

I didn't get the manufacturer of the 24-slot NID cage, but I will next time I'm down there.
But the customers dedicated OC-48 back to the local CO came into a Calix switch at the customer site.
Then it was 1500 feet from the NID cage to their Magix PBX.

If you get a chance to find out what the NID card is, that would be great. If not, no biggie. (I’m not too concerned about the housing.)

I’m not surprised at all that it DOES work, but that 1500 feet is way out of specs for most I’m familial with. I’d kinda like to know for future reference which card can reliably push a signal that far.

Using a CALIX to bring and Oc-48 to site! Whoa! Interesting.

Was a interesting place to work to say the least.
I will look at that NID card next time I'm down there.