10

®

AFE2124

DISCUSSION OF

SPECIFICATIONS

UNCANCELED ECHO

A key measure of transceiver performance is uncancelled

echo. Uncancelled echo is the summation of all of the errors

in the transmit and receive paths of the AFE2124. It includes

effects of linearity, distortion and noise. Uncancelled echo is

tested in production by Burr-Brown with a circuit that is

similar to the one shown in Figure 7.

The measurement of uncancelled echo is made as follows:

The AFE is connected to an output circuit including a typical

1:2 line transformer. The line is simulated by a 135

Ω

resistor. Symbol sequences are generated by the tester and

applied both to the AFE and to the input of an adaptive filter.

The output of the adaptive filter is subtracted from the AFE

output to form the uncanceled echo signal. Once the filter

taps have converged, the RMS value of the uncancelled echo

is calculated. Since there is no far-end signal source or

additive line noise, the uncanceled echo contains only noise

and linearity errors generated in the transmit and receive

sections of the AFE2124.

The data sheet value for uncancelled echo is the ratio of

the rms uncanceled echo (referred to the receiver input

through the receiver gain) to the nominal transmitted signal

(13.5dBm into 135

Ω, or 1.74Vrms). This echo value is

measured under a variety of conditions: with loopback

enabled (line input disconnected); with loopback disabled

closed in Figure 7).

POWER DISSIPATION

Approximately 80% of the power dissipation in the AFE2124

is in the analog circuitry, and this component does not

FIGURE 7. Uncancelled Echo Test Diagram.

TYPICAL POWER DISSIPATION

BIT RATE

IN THE AFE2124 (per channel)

(symbols/sec)

(V)

(mW)

584 (E1)

+3.3

250

584 (E1)

+5

300

392 (T1)

+3.3

240

392 (T1)

+5

270

146 (E1/4)

+3.3

230

146 (E1/4)

+5

245

TABLE II. Typical Power Dissipation (per channel).

change with clock frequency. However, the power dissipa-

tion in the digital circuitry does decrease with lower clock

frequency. In addition, the power dissipation in the digital

section is decreased when operating from a smaller supply

remain in the range 4.75V to 5.25V).

The power dissipation listed in the Specifications Table

applies under these normal operating conditions: 5V analog

power supply; 3.3V digital power supply; standard 13.5dBm

delivered to the line; and a pseudo-random equiprobable

sequence of HDSL output pulses. The power dissipation

specifications includes all power dissipated in the AFE2124;

however, it does not include power dissipated in the exter-

nal load. The external power is 16.5dBm, 13.5dBm to the

line, and 13.5dBm to the impedance matching resistors. The

external load power of 16.5dBm is 45mW. The typical

power dissipation for each half of the AFE2124 under

various conditions is shown in Table II.

The T1 and E1 power measurements in the Specifications

are made with the output circuit shown in Figure 7. This

circuit uses a 1:2 transformer. The power measurements

shown in Table II use an equivalent resistive load instead of

the transformer to eliminate frequency dependent imped-

ances of the transformer.

rxHYB

N

1/2

AFE2124

rxLINE

P

rxLINE

N

rxD13 - rxD0

rxHYB

P

txLINE

N

txLINE

P

txDAT

P

100pF

1.5k

Ω

3k

Ω

13

Ω

13

Ω

5.6

Ω

1:2

5.6

Ω

1.5

Ω

100pF

750

Ω

Transmit

Data

Uncancelled

Echo

750

Ω

135

Ω

S

1