This page is a work in progress, and is currently a mess, but I'm getting there. Don't blame me yet if it's meandering, hard to follow, doesn't flow well, incomplete, inaccurate, or has any other problem.
A while ago, I went down the rabbit hole of trying to untangle T1 line buildouts and equalization. Information on anything related to T1 circuits seems to be mired in unclear and sometimes contradictory jargon and terminology making it difficult to get to the bottom of things at times. This page is an effort to get to the bottom of T1 line buildouts and do so in a well defined and consistent manner while referencing a variety of sources for the content on the page.
Along the way, we'll have to discuss the general arrangement of a traditional, repeatered T1 span as well as some other aspects of T1 to build the correct context for the subject.
T1 line buildout and equalization generally refers to the setting of the amplitude and pulse shape of the pulses transmitted by a T1 transmitter. In general, there are two rough classes of pulse, which I will refer to as short-haul and long-haul. The literature sometimes describes these classes of pulse as 'DSX-1' and 'DS1', but I will stick to the terms short-haul and long-haul. We will also see later that these two classes of pulse are, in fact, closely related.
Some of the information here must be framed in the context of the arrangement of T1 circuits in the real world, and as such some additional terminology to describe that becomes relevant.
A T1 office repeater is the device used in a central office to receive a T1 signal, weak from transmission through the outside cable plant, and regenerate it for use with equipment in the central office or retransmission back into more outside cable plant. There are repeaters used outside of the central office as well, the primary difference simply being the location in which they are used.
Terminal equipment is equipment which terminates a T1 span. A common example would be a channel bank, but these days all manner of switching equipment, muxes, data units, and more could be considered to 'terminate' a T1.
A Channel Service Unit (CSU) is customer premises equipment (CPE) that sits between the T1 network interface (NI) and the customer terminal equipment.
A T1 Network Interface (NI) is the demarcation point where the telephone company network ends and the customer premise wiring begins.
A DS-1 connector is a device that sits just on the phone company side of the NI to provide a few basic functions such as completing span power looping, isolating the customer side from span power, and providing a loopback towards the network when a loopback code is sent. This device was sometimes called a smartjack.
First, we'll discuss T1 receivers. These are the functions of a T1 device which receive and interpret T1 signal pulses.
The lowest signal level that a T1 receiver can properly receive and interpret is known as the sensitivity of the receiver. A receiver with a lower sensitivity can successfully process a lower received signal level.
Generally, there are two classes of T1 receiver: short-haul and long-haul. You may recognize these terms as I've used them before to describe pulse shape/amplitudes. In fact, the relationship is simple: short-haul T1 receivers are designed to receive only short-haul pulses, while long-haul receivers are designed to receive long-haul pulses. Some, but not all, receivers are designed to receive both kinds of pulses.
As I will describe in more detail later, short-haul pulses are actually a higher amplitude than long-haul pulses. As such, long-haul receivers need to have a higher sensitivity than short-haul receivers, which can make them more complex (and expensive).
The only reason some long-haul receivers cannot successfully process short-haul pulses is that receivers also have a maximum signal level they can process. For a given receiver, that maximum may be less than the signal level of short-haul pulses, and as such a long-haul receiver may not be equipped to deal with the higher amplitude of short-haul pulses.
The last point to discuss is where short-haul and long-haul receivers are used. The general idea is that short-haul receivers get used when the distance between equipment is expected to be short (a few hundred feet, generally) such as within a central office or customer premise. Long-haul receivers get used where the distances between equipment is expected to be long such as outside plant between buildings.
In more modern times, economies of scale and improvements in technology used to make T1 receives has generally resulted in most off the shelf T1 receivers being capable of processing the lowest strength long-haul pulses all the way up to the strongest short-haul pulses, and as such the differentiation is often blurred. However, there is still a great deal of older equipment in existence which may not have the same capabilities.
Next, we'll discuss the structure of an office-to-office T1 line. That is, a T1 line which terminates at each end in two different central offices.
235-200-100 (T1 digital line general description) covers this case, and I'll be referring to figure 23 as an example (reproduced here for your convenience).
There's a lot going on in this diagram, but there are a couple relevant points to note: First, note that this diagram is actually showing two T1 lines: two pairs in one direction, and two pairs in the other direction (a T1 line being made up of two pairs, one in each direction). Next, in an intermediate office, each received T1 signal pair is repeated in an office repeater and the transmit pair of that repeater is then cross connected to an outbound pair and on to the next office. Lastly, note that in a terminal office where a T1 line terminates, the received pair of a T1 line lands at an office repeater before being cross connected to terminal equipment, but a send pair from terminal equipmemnt does not go through a repeater before leaving the office. Instead, the send pair from the terminal equipment passes through a passive equalization network before it leaves the office. The document explains that historically, this equalization network was in the office repeater bay, but was eventually moved into the terminal equipment.
It should also be noted that while this diagram shows only an intermediate office and a terminal office, for any given T1 line, there can be theoretically any number of intermediate offices along the way. The diagram also does not cover T1 lines that terminate in customer premises as the document is focussed only on T1 lines between two offices.
For an example of a T1 line terminating on customer premises, I'll refer to 314-645-100 (DS1 digital service description). I'll be referring to figure 6 (reproduced here for your convenience) which shows a line between two customer premises, but it's worth noting that a T1 line could exist between central office and customer equipment as well.
This diagram shows some components we haven't seen before, while also leaving out some of the central office detail from the previous diagram. Firstly, we see some fiber optic multiplexers in the middle, but these aren't really relevant to our discussion. Focusing on the customer ends, we see some new components: a PBX (the customer terminal equipment), the Channel Service Unit (CSU), a Network Interface (NI), and the DS-1 connector.
Although the diagram doesn't show this level of detail, there is an equalization network (or equivalent functionality) between the PBX and the CSU, and a line buildout network (or equivalent functionality) between the CSU and the network interface. Generally the equalization function is provided internal to the terminal equipment and the line buildout function internal to the CSU.
We've now looked at the structure of a single T1 line, but the last piece is to consider how multiple T1 lines from a CO to multiple different customer premises will look. The simplest and most important point to consider is that the length of a T1 line depends primarily on one thing: the relative distance (in terms of the cable routes) between the customer premise and the CO. As each customer premise location is unique, the length of a T1 line to each customer premise can also be unique.
To see why line build-out is needed, we need to understand crosstalk. Crosstalk is the tendency of one signal in a cable pair to be influenced by the signals of nearby cable pairs. It is important to note that crosstalk is a reciprocal phenomenon: any signal that crosstalks to another signal also receives crosstalk back from that other signal. Thus, it is important to think not only of how other signals crosstalk to one signal, but also how that one signal crosstalks to the others.
There are different ways to measure crosstalk, but in general crosstalk becomes a problem if there is enough crosstalk that it begins to degrade the signal enough to cause errors. Crosstalk itself is primarily a factor of cable construction. The degradation that results from crosstalk depends on the amount of crosstalk, the strength of the signal being received, and the strength of the adjacent signals not being received (which 'crosstalk' with the signal we do wish to receive). That is, a stronger signal is degraded less, more crosstalk in a cable allows more degradation, and stronger adjacent signals allow more degradation as well.
ANSI T1.403 goes over line build-out and crosstalk in Annex H. Figure H.1 (reproduced here for convenience) shows an example of a T1 cable end section with a non-repeated route junction.
This scenario shows one T1 circuit, circuit A, which terminates in some customer equipment (top). Between the customer equipment and the repeater location #1, there is some loss which is the sum of L1 and L2.
Similarly, between the repeater location #1 and the next device in circuit B, there is loss given by LR.
Suppose that the signals towards the central office (leftward) on circuit A and circuit B at at the same amplitude where they begin. Now suppose that loss LR is much greater than the loss of L1 + L2. Where circuit A rejoins circuit B, it will have a much greater amplitude than circuit B which has undergone more loss. As a result, we run the risk of crosstalk from circuit A being enough to cause degradation in circuit B.
How do we prevent this issue? We could use better cable that has less crosstalk, but better cable is more costly, so that option is less desirable. We could increase the signal strength of circuit B, but suppose it is already being transmitted at maximum power to overcome the loss LR. The last option is to decrease the signal strength of circuit A, thus reducing the amount of degradation in circuit B.
However, we must be careful: if we decrease the signal strength of circuit A too much, we may discover that we have caused the opposite problem, and now circuit B is much stronger than circuit A causing a crosstalk problem the other way.
The solution, ultimately, is to choose the signal strengths to meet a simple criteria: ensure that all adjacent signals have similar signal strength so that they are simultaneously not strong enough to cause degradation of the other signals, and strong enough to not be degraded by the other signals. ANSI T1.403 states that the signal strength difference between two adjacent signals should be no more than 7.5dB.
The means of decreasing the transmitted signal strength is the line build-out.
The loss of the line build-out is modeled to simulate the loss of cable to ensure that the shape of the signal after all loss (cable loss plus line build-out) is as expected. Line build-out values are required to consist of 0, 7.5dB, and 15dB under FCC part 68 rules.
So then, where does line equalization come in?
Referring back to figure 23 of 235-200-100, we remember that terminal equipment in a central office sends a T1 signal through an equalization network before the signal makes it out of the office. In a central office, the length of cabling between the terminal equipment and the office repeater bay can be highly variable. By the time the signal from the terminal equipment makes it to the repeater bay, it could already have significant signal loss.
To compensate, the terminal equipment transmits with a higher signal strength than normally present in the network. An equalization network is used to adjust the shape and amplitude of the signal so that it is correct for transmission into the network by the time it arrives at the office repeater bay. The equalization network must be chosen to achieve this in conjunction with the length of cabling between the terminal equipment and the office repeater bay.
Similarly, at the customer premises, the customer side of the CSU expects to see a specific amplitude level from the terminal equipment. The line equalization ensures that the proper signal level and pulse shape arrives at the customer side of the CSU from the customer terminal equipment.
We now have all the pieces to put together a complete picture.
We'll use an example of a T1 line from a central office to a customer. We'll start with the direction of transmission from the central office to customer equipment and then work our way back.
At the terminal central office, the terminal equipment transmits T1 pulses of an amplitude greater than those used in the network normally. These are short-haul pulses.
These pulses traverse through a line-equalization network and a length of cable pair which together result in a signal that is of the correct (reduced) amplitude and pulse shape to traverse the network. At this point, they traverse out of the central office to the first line repeater.
After having traveled through the outside cable plant, the first receiver is receiving weakened long-haul pulses. As such, repeaters must be equipped with long-haul receivers.
The repeater regenerates the pulses and transmits full strength long-haul pulses for transmission further down the line. The pattern repeats this way with weakened long-haul pulses arriving at the next repeater to be regenerated and sent further down the line. There can be effectively any number of repeaters along the way.
Finally, the signal is regenerated in the last repeater before the T1 line heads to the customer premises. The signal is again weakened by the length of cable between the customer and the last repeater.
At the customer premises, the signal passes through the DS-1 connector, through the network interface, along the customer premise cabling, and finally arrives at the CSU. The CSU, which is receiving weak long-haul pulses on its network-side interface, must have a long-haul receiver here. The CSU regenerates and transmits the signal using short-haul pulses towards the customer terminal equipment. It will also generally have a line equalization setting to control the pulse shape and amplitude towards the terminal equipment.
Finally, the terminal equipment will receive the short-haul pulses with a short-haul receiver.
In the other direction, the terminal equipment starts by transmitting short-haul pulses. Generally the terminal equipment has a line equalization setting to control the pulse shape and amplitude towards the CSU.
The CSU receives the short-haul pulses using a short-haul receiver. It processes and regenerates the signal, ensuring that it meets all requirements of a proper T1 signal before transmission into the network. The CSU transmits the regenerated long-haul pulses towards the network, first passing through its line build-out network to attenuate the signal as needed.
The weakened pulses transmitted by the CSU rejoin other T1 signals in the telephone company cabling. If line build-out has been configured correctly, the signal strength is within 7.5dB of the other nearby signals.
The T1 signals make their way to the next repeater on the cable. The same as in the other direction, the T1 repeaters regenerate and transmit new long-haul pulses towards the central office.
Finally, the signal from the last repeater before the CO enters the CO and is regenerated one last time in the office repeater. The office repeater regenerates and transmits new full-strength long-haul pulses to the terminal equipment. As there is not much loss between the office repeater and the terminal equipment, the terminal equipment can still use a short-haul receiver.
The following is a list of references with links that are used in this page as supporting information.