Texas Instruments | Active output impedance for ADSL line drivers | Application notes | Texas Instruments Active output impedance for ADSL line drivers Application notes

Texas Instruments Active output impedance for ADSL line drivers Application notes
Amplifiers: Op Amps
Texas Instruments Incorporated
Active output impedance for
ADSL line drivers
By Randy Stephens (Email: r-stephens@ti.com)
Systems Specialist, Member Group Technical Staff
Introduction
The exceptional bidirectional data transmission rates over traditional telephone
lines are a major factor for the widespread
industry growth of ADSL. The ability to
transmit data at over 8 MBps over an existing infrastructure of copper telephone lines
with limited costs is exciting. There are
several key components within the ADSL
system, but this article deals solely with the
line driver amplifiers.
Because ADSL is considered to be a fullduplex system, able to transmit and receive
at the same time, a receiver must be incorporated into the design. The most common
way of accomplishing this is to use a hybrid
network. The hybrid’s function is to cancel
out the transmit signal while still being
capable of receiving the signals from the
customer-premise equipment (CPE) end
(also known as the remote-terminal [RT]
end). To accomplish this task, seriesmatching resistors, RS, are needed and
should be equal to one-half the total
reflected transmission line impedance to
properly match the line impedances (see
Figure 1).
RS =
RLine
2n2
,
(1)
Figure 1. Line driver voltage and current levels to meet
ANSI T1.413 requirements
R
+VCC
2R
Vpeak = 8.85 V
Ipeak = 354 mA
–
6062a
+
+VCC
TX
VIN +
+
6032a
–
RS
12.5 Ω
R
RF
RF
Vpeak =
8.85 V
Line =
100 Ω
1:n
(Typical value
is 1:2)
+VCC
TX
VIN –
–
6032b
+
–VCC
Vpeak = 17.7 V
Ipeak = 177 mA
–VCC
2R G
RX
VOUT +
RS
12.5 Ω
–VCC
R
R
+VCC
–
6062b
+
2R
RX
VOUT –
Vpeak = 8.85 V
where n is the transformer ratio indicated
Ipeak = 354 mA
as 1:n.
–VCC
The problem with using the seriesmatching resistor is the associated voltage
drop across this resistance. The voltage
appearing at the transformer primary side is only one-half
from 5.3 to as high as 7, depending on the manufacturer
the voltage developed at the line driver amplifier output.
and the system goals involved.
This is one of the key issues when the power dissipation of
This large voltage requirement is a key reason for using a
an ADSL line driver is considered.
transformer and two amplifiers configured differentially to
drive the line. Differential circuits have several advantages
Traditional line driver requirements
over single-ended configurations. This includes minimizing
ANSI T1.413 specifies that the central office (CO) can nomcommon-mode signals and interference, improving powerinally transmit at –40 dBm/Hz on a 100-Ω telephone line
supply rejection, and the obvious advantage of doubling
from approximately 25 kHz to 1.104 MHz. This corresponds
the voltage swing that appears at the transformer leads.
to roughly 3.16 VRMS (or +20 dBm) being transmitted on
Another advantage of the differential configuration is that
the line. The problem is that ANSI T1.413 also dictates
even-order harmonics are reduced by as much as 10 to
that there shall be a bit-error rate (BER) of 1 × 10–7. In
20 dB, resulting in a very low distortion system.
order to accomplish this feat the ADSL signal must have
Because RS forces the amplifier to swing twice the
a peak-to-rms ratio, also known as crest factor (CF), of
transformer voltage requirement, the power supplies
about 5.6 (15 dB). Taking the crest factor into account,
(±VCC) must be increased accordingly. This increase in
the line voltage must now have a peak voltage of about
power-supply voltage leads to the primary issue with
17.7 Vpeak (34.4 VPP). Note that the crest factor can vary
ADSL line drivers—power dissipation.
24
Analog and Mixed-Signal Products
www.ti.com/sc/analogapps
4Q 2002
Analog Applications Journal
Amplifiers: Op Amps
Texas Instruments Incorporated
Line driver power dissipation
New VOUTRMS = 5 V ÷ 5.6 = 0.893 V.
Power dissipation in the line driver amplifier is a dominant
factor in CO applications. Let’s take an approximation at
the power dissipation levels required for the traditional
line driver circuit. Let’s assume that the amplifier requires
at least 2 V of power-supply voltage headroom (i.e.,
VOUT(max) = VCC – 2 V) and there is about a 10% tolerance on the power supply. Since power dissipation of
amplifiers is calculated based on the average current flowing into the amplifiers and the dc voltage, the following
line driver amplifier power dissipation approximation can
be made:
PDiss = 2( VCC − VOUTRMS )(IOUTRMS × 0.8*) + PQuiescent . (2)
New IOUTRMS = old IOUTRMS = 63.2 mA.
PQuiescent ≈ 4 × VCC × ICC × 0.7. * *
(3)
To solve for power dissipation, let
VCC(min) ≈ VOUT(max) + VHeadroom + VCCTolerance
= 8.85 V + 2 V + 1..1 = 11.94 V
(choose standard voltage 12 VDC).
VOUTRMS = 8.85 Vpeak ÷ 5.6 = 1.58 VRMS .
IOUTRMS = 354 mA peak ÷ 5.6 = 63.2 mA RMS .
Let ICC = 12 mA DC .
∴ PDiss ≈ 1.05 W + 0.40 W ≈ 1.45 W.
As you can see from the calculation, 1.45 W is a lot of
power for a single device to dissipate. To compound the
problem, there are as many as 72 ADSL lines on a single
PCB. This is an enormous amount of heat to try to dissipate
while trying to maintain proper silicon die temperatures.
Minimizing power dissipation
Power reduction is easily accomplished by reducing the
series-matching resistors (RS) to a much smaller value. The
voltage drop across these resistors is then minimized. The
amplifier output voltage is reduced by the same amount
that allows the power-supply voltages to be reduced.
Because the voltage difference between the power-supply
voltage and the rms output voltage is reduced, power dissipation is also reduced. The quiescent power is reduced
as well, due to the dropping power-supply voltages. Using
the previous example, we can see the amount of power
that will be saved by simply utilizing a smaller resistor. Let
new RS equal 13% of the original RS value.
∴ New PDiss ≈ 0.72 W + 0.22 W ≈ 0.94 W.
This is a savings of 0.51 W, or 35%, per ADSL channel.
When there are several channels on a single PCB, this can
add up to substantial heat savings. The die temperature is
also reduced, allowing for better performance and longer
life of the amplifier.
However, this configuration fails to allow for proper line
impedance matching. To get the best of both worlds, utilizing small series resistors and matching the line impedance,
we need to use an “old” circuit configuration—the active
termination circuit (also known as synthesized impedance).
Active termination
Active termination has been around for several years.1,2
The idea is to use a small ohmic value resistor for RS. The
circuit then utilizes positive feedback to make the impedance of this resistor appear much larger from the line side.
This accomplishes two things: (1) a very small resistance
when the line driver amplifier transmits signals to the line,
and (2) proper matching impedance between the line and
the amplifier. Most of the original designs, however, were
single-ended applications instead of the differential configuration used in ADSL systems.
Taking the general idea a step further, we can utilize the
fact that the signals from each amplifier are 180° out of
phase from each other in the differential system. We use
these signals and connect them into the traditional inverting node on the amplifier (minus input) instead of the
non-inverting node (plus input) used in the single-ended
application. The advantages of this are: (1) The effective
impedance of the noninverting inputs is not dictated by
the positive feedback resistance and voltage gain; and (2)
the active impedance achieves cross-coupling of the signals.
Cross-coupling helps minimize differences between the
two amplifier output signals, helping to keep the signals
fully differential. Figure 2 shows the basic circuit for differential positive feedback.
Figure 2. Basic active impedance circuit
TX VIN+
+
VO +
RS
–
New VOUT(max) = 1.13 × old VOUT(max) ÷ 2 = 5 V.
VOUT +
RF
New VCC = (5 V + 2 V ) × 1.1 = 7.7 V
(choose standard voltage 8 VDC).
RP
2R G
* The ADSL signal is considered to have a Gaussian distribution in the time
domain. Because of this, multiplying the amplifier’s rms output current by
approximately 0.8 yields the average current drawn from the power supply
due to the output signal current.
**This multiplication factor accounts for the fact that part of the quiescent
current in a Class-AB amplifier gets diverted to the load when there is a
signal appearing at the output of the amplifier driving a load. The number
chosen is only an approximation and is shown only as a reference. Typical
numbers range from 0.4 to 0.9 and are based on numerous circuit parameters internal to the amplifier.
1:n
100 Ω
RP
RF
RS
–
TX VIN – +
VOUT –
VO –
25
Analog Applications Journal
4Q 2002
www.ti.com/sc/analogapps
Analog and Mixed-Signal Products
Amplifiers: Op Amps
Texas Instruments Incorporated
The first question to answer is: How does this circuit
configuration increase the effective resistance of RS when
looking from the line? If we assume that the TX inputs are
grounded and apply a voltage at VOUT –, this creates a voltage at VO+ equal to VOUT – × –RF/RP. If we also realize
that the voltage at VOUT+ is equal to –VOUT –, then
VO+ = VOUT+ × RF/RP. This makes RS appear to be a
larger impedance, Z, by the following formula:
Z(Ω) =
RS
.
R
1− F
RP
(4)
The important thing to consider is that regardless of the
forward gain from VIN to VO, the active impedance value
remains constant. The drawback to this arrangement is
that the impedance will change at frequencies near the
amplifier’s bandwidth limit. We must ensure that the
amplifier used has a bandwidth high enough not to alter
the impedance at the ADSL frequencies from 25 kHz to
1.1 MHz. As a general rule of thumb, the amplifier must
have a minimum bandwidth of 10 times the maximum
operating frequency, or at least 11 MHz with the amplifier’s
intended gain.
RL
1
=
.
RL + RS 1 + X
(8)
We will also assume that we want the active impedance, Z,
equal to the terminating resistance, RL. Equation 4 is
manipulated to achieve
RP = RF
 1 


1− X
=
RF
.
1− X
(9)
Equation 9 shows that to properly match the active
termination impedance, we need only select an arbitrary
value of RF. Substituting Equations 7 through 9 in
Equation 5 leads us to the simplified forward voltage
gain of
AV =
RG[(1 + X)(2 − X)] + RF(1 + X)
.
2RGX
(10)
If we know the forward gain we want in the system, we can
rearrange Equation 10 to solve for the gain resistance, RG:
RG =
RF(1 + X)
.
2A V X − [(1 + X)(2 − X)]
(11)
Active impedance forward gain
Once the return impedance is corrected, we need to turn
our attention to the rest of the design parameters. The most
fundamental is the forward voltage gain from input to output. For simplicity, we will assume that the amplifier is
well within its linear range and ignore bandwidth effects.
Equation 5 shows the simplified forward gain from VIN
to VO.
AV
 RF 
1+ 
 RG ||RP 
V ±
= O =
VIN ±
 R   RL 
1−  F  
 RP   RL + RS 
where RL =
RLine
2n2
if RL << RP ,
.
(5)
(6)
In the original circuit (the classic design shown in
Figure 1), RS equaled one-half the total reflected line
impedance, which also equaled RL. We must now choose
RS as a percentage of RL in the active termination circuit.
If we define the variable X as this percentage, where
0 < X ≤ 1, then we can start simplifying the preceding
equations. Some references use the term “synthesis factor”
(SF) to describe the percentage. Synthesis factor is simply
1/X, but the remainder of this article uses the variable X. If
we realize that the term
RL
RL + RS
Because active impedance utilizes positive feedback, it is
possible to create negative impedance instead of positive
impedance. Negative impedance makes the series resistance appear to decrease rather than to increase as desired;
so we must ensure that there is always positive impedance.
We come to our first design constraint of the active termination circuit: There must be a minimum forward gain for
the system to work properly.
Because we want to match the line properly, we must
first arbitrarily choose RF. Using Equation 9 dictates a
specific fixed value for RP. This leads to RG solely dictating the forward voltage gain for any given value of X. The
minimum forward voltage gain allowed is when RG is not
even in the system, resulting in
A V(min) =
2 + X − X2 (1 + X)(2 − X)
=
.
2X
2X
(12)
Luckily, for most ADSL systems, the gain of the amplifiers
is typically greater than 10 V/V. Meeting the minimum gain
requirement is usually not an obstacle as long as the value
of X is greater than about 10%. As long as the minimum
forward gain is met, the low-power active termination
system will work properly.
Line impedance changes
is held constant, we can make several simplifications. The
first sets of assumptions are
RS = RLX and
Minimum active impedance forward gain
design constraint
(7)
Up until now, we have assumed that the line was a fixed
value (usually 100 Ω). But in reality, we know that the line
impedance is highly complex. Typically the line impedance
can range from as low as 50 Ω up to as high as 300 Ω over
the ADSL frequency spectrum. Since the positive feedback
26
Analog and Mixed-Signal Products
www.ti.com/sc/analogapps
4Q 2002
Analog Applications Journal
Amplifiers: Op Amps
Texas Instruments Incorporated
is obtained between RS and the reflected line impedance
(RL), it stands to reason that the forward voltage gain will
be affected.
To quantify the exact change in forward voltage gain,
the variable Y is introduced. Let the variable Y equal the
percentage change in the reflected line impedance (RL).
This leads to the new forward voltage gain:
RG[(2 − X)(1 + X + Y)] + RF(1 + X + Y)
.
RGX(2 + Y)
AV =
(13)
Figure 3 illustrates the percentage change in forward
gain with varying values of X. The forward gain with a
100-Ω line impedance will be used as the base line for
comparison.
It is interesting to note that the change in percentage
gain is independent of the transformer ratio, n; feedback
resistance, RF; gain resistance, RG; and initial amplifier
gain, AV.
The minimum forward gain will also vary with the line
impedance. The minimum forward gain becomes
(2 − X)(1 + X + Y)
A V(min) =
.
X(2 + Y)
(14)
Figure 4 illustrates the minimum forward gain with varying line impedance.
When an active termination system is designed, it does
not matter what initial design line impedance is used. As
Figure 3. Forward gain change with varying
line impedance
long as the minimum gain criterion is met, the system
should not create negative impedances.
Line impedance changes and the amplifier
output voltage
In a real system it is quite common for forward voltage
gain to change ±20%, which must be accounted for. If not,
the input signal can be amplified too high and clipping
could easily occur. Excess distortion, data transfer rate,
line reach, and even power dissipation could become
worse if the line impedance is not handled properly within
the active impedance circuit design.
Examining the circuit of Figure 2 and using Equation 7
will help us calculate how the line impedance changes the
amplifier’s output voltage. We will assume that RS is
designed for a 100-Ω system and is held constant. We will
also assume that the power on the line was done with a
100-Ω line impedance and is +20 dBm. This corresponds
to a line voltage of 3.162 VRMS. The formula used to find
the corresponding amplifier voltages is
VORMS =
(RLine + 2n2RS )
2nRLine
.
(15)
Figure 4. Minimum forward gain change with
varying line impedance
16
X = 20%
40
Min. Forward Voltage Gain (V/V)
Change in Forward Gain, AV (%)
RMS
The important number is the peak output voltage of the
amplifier (Vpeak = VRMS × CF) because a given supply
voltage determines how much voltage swing can occur.
Failure to plan for varying line impedances can cause
60
20
X = 40%
0
X = 30%
–20
–40
–60
VLine
X = 10%
0
50
150
100
200
Line Impedance, R Line (Ω)
250
300
14
12
X = 10%
10
X = 15%
8
X = 20%
6
X = 30%
4
2
X = 40%
0
50
100
150
200
250
300
Line Impedance, RLine (Ω)
27
Analog Applications Journal
4Q 2002
www.ti.com/sc/analogapps
Analog and Mixed-Signal Products
Amplifiers: Op Amps
Texas Instruments Incorporated
Lab tests
Setup
The first test examined how the resistor values affect the
system. Because the THS6032, like most ADSL line drivers,
is a current feedback (CFB) amplifier, the feedback resistance (RF) dictates the bandwidth and the stability of the
amplifier. Keeping a high bandwidth increases the amplifier’s
excess open-loop gain in the ADSL frequency band and
reduces distortion. At the same time, however, the amplifier bandwidth may be high enough to interact with the
transformer’s resonance frequency, which can cause
possible instabilities in the overall system. This is especially
true when active impedance circuits are used, as RS can
become very small, resulting in very little isolation between
the amplifier and the transformer. When you consider
Equations 13 to 15 along with the transformer’s impedance
at resonance, it is apparent that the system can potentially
become unstable. Using a simple RC snubber across the
transformer can be a simple solution for instability concerns.
To circumvent this potential issue, two new amplifiers
from Texas Instruments, the THS6132 and the THS6182,
incorporate special internal circuitry. These new amplifiers
yield extremely low distortion at the ADSL frequencies yet
have a bandwidth of only 10 to 20 MHz—depending on the
system design. For all other line drivers, the trade-off of
bandwidth and stability needs to be managed. As a side
benefit of reducing the feedback resistor, the overall output
noise of the line driver system can be significantly reduced.
Figure 6. Amplifier effective output impedance
viewed from transformer primary winding
1000
Effective Output Impedance, Z (Ω)
Transformer = 1:1.2
X = 100%
100
X = 40%
Figure 5. Amplifier peak output voltage
with X = 20%
15
Amplifier Peak Output Voltage (V)
some serious problems. Figure 5 illustrates this issue with
X = 20% (SF = 5) and a crest factor of 5.3.
Obviously, as the crest factor increases, the peak output
voltage will also increase. Additionally, when RS increases,
the amplifier output voltage will also increase. The obvious
question is: Why not use the smallest resistance possible?
There are several reasons for this that the remainder of
this article explains in detail.
PLine = +20 dBm @ 100 Ω
CF = 5.3
X = 20%
13
11
RLine = 300 Ω
9
RLine = 100 Ω
7
5
RLine = 50 Ω
3
1
1.2
1.4
1.6
1.8
Transformer Ratio (1:n)
2
For the THS6032 testing, a feedback resistor value of
1150 Ω was chosen. The rest of the system component
values were then easily calculated with the previous
equations. The only other variable was that the gain of each
amplifier was set to approximately +12 V/V. This allowed
testing of the X = 10% system where the appropriate
minimum gain requirement was about 10.5. As RS was
increased, the gain also had to be increased to account for
the additional voltage drop from the added series resistance.
The active impedance test
Figure 6 shows the impedance looking back into RS from
the transformer primary. It clearly shows the amplifier’s
closed-loop bandwidth effects. Eventually the amplifier’s
own output impedance takes over regardless of the termination system used. At this point the impedance is out of
the designer’s control. Since the ADSL spectrum is well
controlled, the system will meet its designated functionality
as a low-power line driver.
One area of concern with using active impedance is that
lightning surge tests could overwhelm the amplifiers’
internal circuitry and cause failures due to a decreased
real resistance between the amplifier and the transformer.
The larger the resistance, the better the chance that no
damage will occur within the amplifier. If the active
impedance configuration is utilized, then RS should be a
“respectable” value and not something trivial (for example, 1 Ω). Most systems should strive for a value of 20 to
30% of RL (SF = 3 to 5). This allows for respectable power
savings and reasonable isolation from surges on the line.
X = 14%
10
0.01
0.1
1
10
Frequency (MHz)
100
1000
28
Analog and Mixed-Signal Products
www.ti.com/sc/analogapps
4Q 2002
Analog Applications Journal
Amplifiers: Op Amps
Texas Instruments Incorporated
Power dissipation and distortion
The line impedance used in the testing was a 100-Ω resistor. Variable line impedance issues are not of concern but
should constrain the final system design. As a result, the
power dissipation numbers shown should be considered
optimal for a particular test setup. When a varying line
impedance is thrown into the mix, the power-supply voltages
will need to be adjusted accordingly and the power dissipation will increase.
The other factor hampering the power dissipation is
that the THS6032 requires 4-V headroom from the power
supplies. This is due to the Class-G architecture requiring
multiple series transistors in the output stage. If a very
low headroom amplifier were used (such as the THS6132
or THS6182), the power-supply voltage could be reduced
by at least ±2 V, decreasing power even more. As we are
concerned with power savings in general, these results can
be used to draw some general conclusions about the use
of active termination in an ADSL application.
Keep in mind that when you compare power numbers
from amplifier to amplifier, the entire system configuration
needs to be divulged. This includes things such as crest
factor; accounting for varying line impedances; accounting
for power-supply tolerances; and, of course, the synthesis
factor. Because of the numerous options available, doing a
true apples-to-apples comparison is often very difficult
when you just look at manufacturers’ data sheets.
As a reference for the active termination testing, a
THS6032 was tested with the traditional configuration
Figure 7. Traditional circuit design power
dissipation results
shown in Figure 1. To really see the effects of the Class-G
circuitry in action, refer to Figure 7, which shows how
changing the VCC–L supply voltages alters the power dissipation. For reference, it also shows the power consumed in
each set of supplies. In Class-AB mode, power dissipation
is about 1.8 W; but in Class-G mode, the best power
achieved is approximately 1.35 W with VCC–L at ±6 V. The
multitone power ratios (MTPRs) were –70 dBc for Class-AB
operation and –68 dBc for Class-G operation.
Figure 8 shows how the crest factor affects power dissipation with a 1:1.2 transformer and X = 20% (RS = 6.94 Ω).
The power-supply voltage was chosen to give an additional
±0.5-V headroom for a design margin. In the lab, we could
set the supplies ±1 V lower before clipping started to occur;
but this is not considered good practice, as power-supply
tolerances and component tolerances could come into
play. The power dissipation numbers shown are thus considered to be realistic and within the safe operating area
of the system.
When compared to the traditional circuit design, the
active termination circuit saved a huge 47% in power
dissipation. This was true for both Class-AB operation
and Class-G operation. For the active termination data,
the use of Class-G operation saved an additional 20 to
25% power dissipation compared to the Class-AB operation. As expected, when the crest factor increased, the
power dissipation also increased by as much as 25%. This
was mainly due to the increase in power-supply voltage
required to handle the larger peak voltages.
Figure 8. Power dissipation with 1:1.2
transformer and different crest factors
1.3
2.0
1.8
Power Dissipation (W)
Line Driver Power Dissipation (W)
Total Power
1.6
1.4
1.2
1.0
VCC –H Power
VCC –H = ±15 V
CF = 5.3
Transformer = 1:2
0.8
0.6
0.4
VCC –L Power
0.2
0.0
Transformer = 1:1.2
X = 20%
PLine = +20.0 dBm
R Line = 100 Ω
1.2
CF = 5.3; VCC–H = ±13 V
CF = 5.6; VCC–H = ±13.5 V
CF = 6; VCC–H = ±14 V
1.1
1.0
0.9
Class-AB Mode
VCC–L = ±0 V
Class-G Mode
VCC–L = ±5 V
0.8
Class-G Mode
VCC–L = ±6 V
0.7
0.6
0
2
4
VCC –L (±V)
6
8
5.3
5.4
5.5
5.6
5.7
5.8
Crest Factor (Vpeak /VRMS)
5.9
6
29
Analog Applications Journal
4Q 2002
www.ti.com/sc/analogapps
Analog and Mixed-Signal Products
Amplifiers: Op Amps
Texas Instruments Incorporated
Figure 9. Power dissipation with 1:1.2
transformer and varying RS
1.1
Line Driver Power Dissipation (W)
Figure 9 shows how changing RS affects the power dissipation. A common crest factor of 5.3 was used to illustrate
the change in the system.
If the power-supply voltages had been held constant and
no clipping had occurred, the power dissipation would
have decreased with an increase in RS; but the testing was
done to show the best possible performance with a given
set of constraints. The power-supply voltages thus were
increased as RS was increased to compensate for the
increase in output voltage required from the amplifier. The
power-supply voltages ranged from ±12.5 V (X = 14%) to
±14 V (X = 40%).
The last thing to check was the effect of MTPR distortion on the system.
Figure 10 shows us that as RS increases, the MTPR
distortion decreases. The designer has to choose between
lower distortion and lower power dissipation. As stated
earlier, a series resistance of 20 to 30% of RL should give
good results for both requirements.
Power dissipation and MTPR with multiple
transformer ratios
Class-G Mode
VCC–L = ±5 V
–75
–80
–85
Class-AB Mode
VCC–L = ±0 V
10
15
20
25
30
35
Series Resistance, RS (% of RL)
0.8
0.7
Class-G Mode
VCC–L = ±6 V
0.6
15
Transformer = 1:1.2
PLine = +20.0 dBm
R Line = 100 Ω
CF = 5.3
VCC–H = Optimum
20
25
30
35
Series Resistance, RS (% of RL)
40
Regardless of the power-supply voltages and the mode of
operation, as RS increases, the power dissipation increases.
This is generally dominated by the amplifier’s overhead
Line Driver Power Dissipation (W)
Line Driver MTPR (dB)
Class-G Mode
VCC–L = ±6 V
–70
Class-G Mode
VCC–L = ±5 V
Figure 11. Power dissipation with varying
transformer ratios
Transformer = 1:1.2
PLine = +20.0 dBm
R Line = 100 Ω
CF = 5.3
VCC–H = Optimum
–65
0.9
10
Figure 10. MTPR with 1:1.2 transformer and
varying RS
–60
1.0
0.5
The purpose of the next series of tests was to find out if
there is a general relationship between the transformer
ratio and the power dissipation. For each transformer ratio
tested, the corresponding resistor values and power-supply
voltages were accordingly changed. Figure 11 shows how
changing RS affects power dissipation with varying transformer ratios.
–55
Class-AB Mode
VCC–L = ±0 V
1.2
PLine = +20.0 dBm
R Line = 100 Ω
RS = 40%
CF = 5.3
1.1 V
CC–H = Optimum
1.0
RS = 30%
Class-AB Data
RS = 20%
0.9
Class-G (5 V ) Data
RS = 40%
RS = 30%
0.8
0.7
RS = 20%
1
1.2
1.4
1.6
1.8
Transformer Ratio (1:n)
2
40
30
Analog and Mixed-Signal Products
www.ti.com/sc/analogapps
4Q 2002
Analog Applications Journal
Amplifiers: Op Amps
Texas Instruments Incorporated
Conclusion
Reduced power dissipation is the main goal for using active
termination in ADSL systems. Using a 1:1.2 transformer
saved 47% of power regardless of the mode in which the
THS6032 was used. This translates to a savings of up to
0.85 W with Class-AB operation and 0.63 W with optimal
Class-G operation. In light of the distortion and power
savings, choosing a value for X of 0.2 to 0.3 (SF = 3 to 5)
shows about the best overall performance.
Using TI’s newest amplifiers, THS6132 (Class-G) or
THS6182 (Class-AB), can save substantially even more
power. Initial testing with the THS6132 in Class-G operation shows a total power consumption of as low as 0.53 W,
which is a power dissipation of roughly 0.43 W over the
THS6032. However, keep in mind the design constraints of
the active termination system. The line impedance variations, the minimum power-supply voltages, and the system
crest factor all contribute to the power consumption of the
line driver.
With any electrical circuit, there are trade-offs to using
one configuration over another. The active impedance
circuit is no exception. The trade-off to achieving lower
line driver power dissipation is that the receiver circuitry
will require more voltage gain to overcome the voltage
reduction appearing across RS. This can play a significant
role in the noise performance of the system. One way to
help alleviate this problem is to use a smaller transformer
ratio; but the power-supply voltages will have to be
increased, which can increase power dissipation. The
added benefits of an increased series resistance can help
in many other areas of the system, including distortion and
Figure 12. MTPR distortion with varying
transformer ratios
–60
Line Driver MTPR (dBc)
voltage requirements and quiescent current. We now come
to the final test—determining the effects of varying transformer ratios on MTPR distortion. Figure 12 shows the
effects of RS on MTPR distortion with a changing transformer ratio and the same setup that was used before.
The data tells us that increasing the physical value of RS
lowers MTPR distortion. This is because distortion in
operational amplifiers generally gets better with an
increase in load resistance. In the case of the ADSL
configuration, increasing RS also helps isolate the complex
loading that the transformer places on the amplifier.
Comparing the 1:2 transformer data with the traditional
circuit design shows that MTPR performance degrades by
4 to 5 dB as the transformer ratio increases.
RS = 40%*
RS = 30%*
–65
RS = 20%*
–70
RS = 30%
Class-AB Mode
–75
RS = 40%
Class-AB Mode
–80
–85
RS = 20%
Class-AB Mode
1
1.2
1.4
PLine = –20.0 dBm
R Line = 100 Ω
CF = 5.3
VCC–H = Optimum
1.6
1.8
2
Transformer Ratio (1:n)
*Class-G Mode (5 V)
surge isolation. Ultimately, the goal of saving power can
still be met while satisfying all requirements of the ADSL
line driver system.
Additional information will be available in an application
note to be released by January 2003, at
www-s.ti.com/sc/techlit/sloa100
References
1. Jerry Steele, “Ideas For Design - Positive Feedback
Terminates Cables,” Electronic Design (March 6,
1995), pp. 91-92.
2. Donald Whitney Jr., “Design Ideas - Circuit Adapts
Differential Input to Drive Coax,” Electronic Design
News (May 8, 1997), pp. 132-34.
Related Web sites
analog.ti.com
www-s.ti.com/sc/techlit/sloa100
www.ti.com/sc/device/partnumber
Replace partnumber with THS6032, THS6132 or THS6182
31
Analog Applications Journal
4Q 2002
www.ti.com/sc/analogapps
Analog and Mixed-Signal Products
IMPORTANT NOTICE
Texas Instruments Incorporated and its subsidiaries (TI) reserve
the right to make corrections, modifications, enhancements,
improvements, and other changes to its products and services at
any time and to discontinue any product or service without notice.
Customers should obtain the latest relevant information before
placing orders and should verify that such information is current
and complete. All products are sold subject to TI's terms and
conditions of sale supplied at the time of order acknowledgment.
TI warrants performance of its hardware products to the
specifications applicable at the time of sale in accordance with TI's
standard warranty. Testing and other quality control techniques are
used to the extent TI deems necessary to support this warranty.
Except where mandated by government requirements, testing of
all parameters of each product is not necessarily performed.
TI assumes no liability for applications assistance or customer
product design. Customers are responsible for their products and
applications using TI components. To minimize the risks
associated with customer products and applications, customers
should provide adequate design and operating safeguards.
TI does not warrant or represent that any license, either express or
implied, is granted under any TI patent right, copyright, mask work
right, or other TI intellectual property right relating to any
combination, machine, or process in which TI products or services
are used. Information published by TI regarding third-party
products or services does not constitute a license from TI to use
such products or services or a warranty or endorsement thereof.
Use of such information may require a license from a third party
under the patents or other intellectual property of the third party, or a
license from TI under the patents or other intellectual property of TI.
Reproduction of information in TI data books or data sheets is
permissible only if reproduction is without alteration and is
accompanied by all associated warranties, conditions, limitations,
and notices. Reproduction of this information with alteration is an
unfair and deceptive business practice. TI is not responsible or
liable for such altered documentation.
Resale of TI products or services with statements different from or
beyond the parameters stated by TI for that product or service
voids all express and any implied warranties for the associated TI
product or service and is an unfair and deceptive business
practice. TI is not responsible or liable for any such statements.
Following are URLs where you can obtain information on other
Texas Instruments products and application solutions:
Products
Amplifiers
Data Converters
DSP
Interface
Logic
Power Mgmt
Microcontrollers
amplifier.ti.com
dataconverter.ti.com
dsp.ti.com
interface.ti.com
logic.ti.com
power.ti.com
microcontroller.ti.com
Applications
Audio
Automotive
Broadband
Digital control
Military
Optical Networking
Security
Telephony
Video & Imaging
Wireless
www.ti.com/audio
www.ti.com/automotive
www.ti.com/broadband
www.ti.com/digitalcontrol
www.ti.com/military
www.ti.com/opticalnetwork
www.ti.com/security
www.ti.com/telephony
www.ti.com/video
www.ti.com/wireless
TI Worldwide Technical Support
Internet
TI Semiconductor Product Information Center Home Page
support.ti.com
TI Semiconductor KnowledgeBase Home Page
support.ti.com/sc/knowledgebase
Product Information Centers
Americas
Phone
Internet/Email
+1(972) 644-5580
Fax
support.ti.com/sc/pic/americas.htm
+1(972) 927-6377
Europe, Middle East, and Africa
Phone
Belgium (English) +32 (0) 27 45 54 32
Netherlands (English) +31 (0) 546 87 95 45
Finland (English) +358 (0) 9 25173948
Russia
+7 (0) 95 7850415
France
+33 (0) 1 30 70 11 64
Spain
+34 902 35 40 28
Germany
+49 (0) 8161 80 33 11
Sweden (English)
+46 (0) 8587 555 22
Israel (English)
1800 949 0107
United Kingdom
+44 (0) 1604 66 33 99
Italy
800 79 11 37
Fax
+(49) (0) 8161 80 2045
Internet
support.ti.com/sc/pic/euro.htm
Japan
Fax
International
Internet/Email
International
Domestic
Asia
Phone
International
Domestic
Australia
China
Hong Kong
Indonesia
Korea
Malaysia
Fax
Internet
+81-3-3344-5317
Domestic
0120-81-0036
support.ti.com/sc/pic/japan.htm
www.tij.co.jp/pic
+886-2-23786800
Toll-Free Number
1-800-999-084
800-820-8682
800-96-5941
001-803-8861-1006
080-551-2804
1-800-80-3973
886-2-2378-6808
support.ti.com/sc/pic/asia.htm
New Zealand
Philippines
Singapore
Taiwan
Thailand
Email
Toll-Free Number
0800-446-934
1-800-765-7404
800-886-1028
0800-006800
001-800-886-0010
tiasia@ti.com
ti-china@ti.com
C011905
Safe Harbor Statement: This publication may contain forwardlooking statements that involve a number of risks and
uncertainties. These “forward-looking statements” are intended
to qualify for the safe harbor from liability established by the
Private Securities Litigation Reform Act of 1995. These forwardlooking statements generally can be identified by phrases such
as TI or its management “believes,” “expects,” “anticipates,”
“foresees,” “forecasts,” “estimates” or other words or phrases
of similar import. Similarly, such statements herein that describe
the company's products, business strategy, outlook, objectives,
plans, intentions or goals also are forward-looking statements.
All such forward-looking statements are subject to certain risks
and uncertainties that could cause actual results to differ
materially from those in forward-looking statements. Please
refer to TI's most recent Form 10-K for more information on the
risks and uncertainties that could materially affect future results
of operations. We disclaim any intention or obligation to update
any forward-looking statements as a result of developments
occurring after the date of this publication.
Trademarks: All trademarks are the property of their
respective owners.
Mailing Address:
Texas Instruments
Post Office Box 655303
Dallas, Texas 75265
© 2005 Texas Instruments Incorporated
SLYT108
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertising