The SC480 is a combination switching regulator and
linear source/sink regulator intended for DDR1/2/3
memory systems. The purpose of the switching regulator
is to generate the supply voltage, VDDQ, for the memory
system. It is a pseudo-fi xed frequency constant on-
time controller designed for high effi ciency, superior DC
accuracy, and fast transient response. The purpose of the
linear source/sink regulator is to generate the memory
termination voltage, VTT, with the ability to source and
sink a 3A peak current.
For the VDDQ regulator, the switching frequency is constant
until a step in load or line voltage occurs at which time the
pulse density, i.e., frequency, will increase or decrease to
counter the transient change in output or input voltage.
After the transient, the frequency will return to steady-state
operation. At lighter loads, the selectable Power-Save
Mode enables the PWM converter to reduce its switching
frequency and improve effi ciency. The integrated gate
drivers feature adaptive shoot-through protection and
soft-switching. Additional features include cycle-by-cycle
current limiting, digital soft-start, over-voltage and undervoltage protection and a power good fl ag.
For the VTT regulator, the output voltage tracks REF,
which is ½ VDDQ to provide an accurate termination
voltage. The VTT output is generated from a 1.2V to VDDQ
input by a linear source/sink regulator which is designed
for high DC accuracy, fast transient response, and low
external component count. All three outputs (VDDQ, VTT
and REF) are actively discharged when VDDQ is disabled,
reducing external component count and cost. The SC480
is available in a 24-pin MLPQ (4x4 mm) package.
Typical Application Circuit
Features
Constant On-Time Controller for Fast Dynamic
Response on VDDQ
DDR1/DDR2/DDR3 Compatible
VDDQ = Fixed 1.8V or 2.5V, or Adjustable from
1.5V to 3.0V
1% Internal Reference (2% System Accuracy)
Resistor Programmable On-Time for VDDQ
VCCA/VDDP Range = 4.5V to 5.5V
VIN Range = 2.5V to 25V
VDDQ DC Current Sense Using Low-Side R
Sensing or External R
in Series with Low-Side
SENSE
DS(ON)
FET
Cycle-by-Cycle Current Limit for VDDQ
Digital Soft-Start for VDDQ
Analog Soft-Start for VTT/REF
Smart Over-Voltage VDDQ Protection Against Source-
Current Loads
Combined EN and PSAVE Pin for VDDQ
Over-Voltage/Under-Voltage Fault Protection
Power Good Output
Separate VCCA and VDDP Supplies
VTT/REF Range = 0.75V – 1.5V
VTT Source/Sink 3A Peak
Internal Resistor Divider for VTT/REF
VTT is High Impedance in S3
VDDQ, VTT, REF Are Actively Discharged in S4/S5
24 Lead MLPQ (4x4 mm) Lead-Free Package
Fully WEEE and RoHS Compliant
Applications
¡ Notebook Computers
¡ CPU I/O Supplies
¡ Handheld Terminals and PDAs
¡ LCD Monitors
¡ Network Power Supplies
5V
D1
20DL19
22DH21
23
1
2
3
4
5
6
PGND2
VTTS
VSSA
TON
REF
VCCA
0.1uF
24
LX
BST
VTT
VTTIN
U1
SC480
FB
EN/PSV
VDDQS
VTTEN
NC
7
9
8
10
11
C8
C1
1uF
VBAT
VTT
R1
1Meg
REF
VDDQ
VDDQVDDQ
C5
C4
10uF
10uF
VTTSNS
C7
1nF
R6
10R
C10
1uF
C9
1uF
November 3, 2006
PGND1
PGND1
ILIM
VDDP
VDDP
PGD
PAD
NC
12
R7 10R
VTT_EN
EN/PSV
C2
0.1uF
18
17
16
15
14
13
PAD
VBAT
C3
2x10uF
Q1
VDDQ
+
C6
PGOOD
5V
RILIM
L1
R4
C11
1uF
Q2
www.semtech.com1
SC480
NOT RECOMMENDED FOR NEW DESIGN
NOT RECOMMENDED
FOR NEW DESIGN
POWER MANAGEMENT
POWER MANAGEMENT
Absolute Maximum Ratings
Exceeding the specifi cations below may result in permanent damage to the device or device malfunction. Operation outside of the parameters specifi ed in the
Electrical Characteristics section is not implied. Exposure to absolute maximum rated conditions for extended periods of time may affect device reliability.
ParameterSymbolMaximumUnits
TON to VSSA-0.3 to +25.0V
DH, BST to PGND1-0.3 to +31.0V
BST, DH to LX-0.3 to +6.0V
LX to PGND1-2.0 to +25.0V
DL, ILIM, VDDP to PGND1-0.3 to +6.0V
VDDP to DL-0.3 to +6.0V
VTTIN to PGND2; VTT to PGND2; VTTIN to VTT-0.3 to +6.0V
EN/PSV, FB, PGD, REF, VCCA, VDDQS, VTTEN, VTTS to VSSA-0.3 to +6.0V
VCCA to EN/PSV, FB, REF, VDDQS, VTT, VTTEN, VTTIN, VTTS-0.3 to +6.0V
PGND1 to PGND2; PGND1 to VSSA; PGND2 to VSSA -0.3 to +0.3 V
Thermal Resistance Junction to Ambient
Operating Junction Temperature RangeT
Storage Temperature RangeT
Peak IR Refl ow Temperature, 10s - 40sT
ESD Protection Level
Notes:
1) Calculated from package in still air, mounted to 3” x 4.5”, 4 layer FR4 PCB with thermal vias under the exposed pad per JESD51 standards.
2) Tested according to JEDEC standard JESD22-A114-B.
(2)
(1)
θ
JA
J
STG
PKG
V
ESD
29°C/W
-40 to +150°C
-65 to +150°C
260°C
2kV
Electrical Characteristics
TEST CONDITIONS: VIN = 15V, VCCA = VDDP = VTTEN = EN/PSV = 5V, VDDQ = VTTIN = 1.8V, R
ParameterConditions
Input Supplies
S0 State (VTT on);
VCCA Operating Current
FB > Regulation Point, IVDDQ = 0A
VCCA Operating Current
FB > Regulation Point, IVDDQ = 0A
VCCA Operating Voltage54.55.5V
VDDP Operating CurrentFB > Regulation Point, IVDDQ = 0A70150μA
TEST CONDITIONS: VIN = 15V, VCCA = VDDP = VTTEN = EN/PSV = 5V, VDDQ = VTTIN = 1.8V, R
ParameterConditions
FB Input Thresholds
FB Logic Input LowVDDQ Set for 2.5V (DDR1)0.3V
FB Logic Input HighVDDQ Set for 1.8V (DDR2)
Gate Drives
Shoot-Thru Protection Delay
DL Pull-Down ResistanceDL Low0.8Ω
DL Sink CurrentVDL = 2.5V3.1A
DL Pull-Up ResistanceDL High2Ω
(4)(7)
DH or DL Rising30ns
= 1MΩ. T
TON
MinTypMaxMinMax
= -40 TO +85C.
AMB
25°C-40°C to 85°C
VCCA
- 0.7
Units
V
DL Source CurrentVDL = 2.5V1.3A
DH Pull-Down ResistanceDH Low, BST - LX = 5V2Ω
DH Pull-Up Resistance
DH Sink/Source CurrentVDH = 2.5V1.3A
VTT Pull-Up ResistanceVTTS < REF0.25Ω
VTT Pull-Down ResistanceVTTS > REF0.25Ω
VTT Peak Sink/Source Current
Notes:
1) The VDDQ DC regulation level is higher than the FB error comparator threshold by 50% of the ripple voltage.
2) Using a current sense resistor, this measurement relates to PGND1 minus the source of the low-side MOSFET.
3) clks = switching cycles, consisting of one high side and one low side gate pulse.
4) Guaranteed by design.
5) Thermal shutdown latches both outputs (VTT and VDDQ) off, requiring VCCA or EN/PSV cycling to reset.
6) VTT soft-start ramp rate is limited to 5.5mV/μs typical. If the VDDQ/2 ramp rate is slower than 5.5mV/μsec, the VTT soft-start ramp will follow the VDDQ/2
ramp.
7) See Shoot-Through Delay Timing Diagram on Page 6.
8) Semtech’s SmartDriver™ FET drive fi rst pulls DH high with a pull-up resistance of 10Ω (typ.) until LX = 1.5V (typ.). At this point, an additional pull-up device is
activated, reducing the resistance to 2Ω (typical). This creates a softer turn-on with minimal power loss, eliminating the need for an external gate or boost
resistor.
9) Provided operation below T
increases with temperature, and by available source voltage (typically VDDQ/2).
(8)
(9)
is maintained. VTT output current is also limited by internal MOSFET resistance which is typically 0.25Ω at 25°C and which
1) Only available in tape and reel packaging. A reel contains 3000 devices.
2) This product is fully WEEE and RoHS compliant.
1PGND2Power ground for VTT output. Connect to thermal pad and ground plane.
2VTTSSense pin for VTT. Connect to VTT at the load.
3VSSAGround reference for analog circuitry. Connect to thermal pad.
4TON
5REF
This pin is used to sense VBAT through a pull-up resistor, RTON, which sets the top MOSFET
on-time. Bypass this pin with a 1nF capacitor to VSSA.
Reference output. An internal resistor divider from VDDQS sets this voltage to 50% VDDQ (nominal). Bypass this pin with a series 10Ω/1μF to VSSA.
6VCCAAnalog supply voltage input. Use a 10Ω/1μF RC fi lter from +5V to VSSA.
7NCNo Connect.
8VDDQSSense input for VDDQ. Used to set the on-time for the top MOSFET and also to set REF/VTT.
9FBFeedback select input for VDDQ. See FB Confi guration Table.
10VTTEN
Enable pin for VTT. Pull this pin low to disable VTT (REF remains present as long as VDDQ is
present).
Enable/Power Save input pin. Tie to ground to disable VDDQ. Tie to +5V to enable VDDQ and
11EN/PSV
activate PSAVE mode. Float to enable VDDQ and activate continous conduction mode. If fl oated,
bypass to VSSA with a 10nF capacitor.
12NCNo Connect.
13PGD
Power good output for VDDQ. PGD is low if VDDQ is outside the power good thresholds. This
pin is an open drain NMOS output and requires an external pull-up resistor.
14,15VDDP+5V supply voltage input for the VDDQ gate drivers.
Current limit input pin. Connect to drain of low-side MOSFET for RDS(on) sensing or the source
for resistor sensing through a threshold sensing resistor.
17,18PGND1Power ground for VDDQ switching circuits. Connect to thermal pad and ground plane.
19DLGate drive output for the low side MOSFET switch.
20LXPhase node - the junction between the top and bottom FETs and the output inductor.
21DHGate drive output for the high side MOSFET switch.
22BSTBoost capacitor connection for the high side gate drive.
23VTTIN
24VTT
THERMAL
T
PAD
Input supply for the high side switch for VTT regulator. Decouple with a 1μF capacitor to
PGND2.
Output of the linear regulator. Decouple with two (minimum) 10μF ceramic capacitors to
PGND2, locating them directly across pins 24 and 1.
Pad for heatsinking purposes. Connect to ground plane using multiple vias. Not connected
internally.
The SC480 requires an external +5V bias supply in addition
to the battery. If stand-alone capability is required, the
+5V supply can be generated with an external linear
regulator. To minimize crosstalk, the controller has seven
supply pins: VDDP (2 pins), PGND1 (2 pins), PGND2,
VCCA and AGND.
The controller requires its own AGND plane which should
be tied by a single trace to the negative terminal of the
output capacitor. All external components referenced
to AGND in the schematic should then be connected to
the AGND plane. The supply decoupling capacitor should
be tied between VCCA and AGND. A single 10Ω resistor
should be used to decouple the VCCA supply from the
main VDDP supply. PGND can then be a separate plane
which is not used for routing analog traces. All PGND
connections should connect directly to this plane with
special attention given to avoiding indirect connections
between AGND and PGND which will create ground loops.
As mentioned above, the AGND plane must be connected
to the PGND plane at the negative terminal of the output
capacitor. The VDDP input provides power to the upper
and lower gate drivers. A decoupling capacitor for the
VDDP supply and PGND is recommended. No series
resistor between VDDP and the 5 volt bias is required.
Pseudo-Fixed Frequency Constant On-Time
PWM Controller
The PWM control method is a constant-on-time, pseudofi xed frequency PWM controller, see Figure 1. The ripple
voltage seen across the output capacitor’s ESR provides
the PWM ramp signal, eliminating the need for a current
sense resistor. The on-time is determined by a one-shot
whose period is proportional to output voltage, and
inversely proportional to input voltage. A separate oneshot sets the minimum off-time (typically 425ns).
On-Time One-Shot (TON)
The on-time one-shot comparator has two inputs. One
input looks at the output voltage, while the other input
samples the input voltage and converts it to a proportional
current. This current charges an internal on-time capacitor.
The TON time is the time required for this capacitor to
charge from zero volts to VOUT, thereby making the ontime of the high-side switch directly proportional to output
voltage and inversely proportional to input voltage. This
implementation results in a nearly constant switching
frequency without the need of a clock generator.
T
RTON is a resistor connected between the input supply and
the TON pin.
VDDQ/VTT Enable & Power-Save
The EN/PSV pin controls the VDDQ supply and the REF
output (1/2 of VDDQ). VTTEN enables the VTT supply. The
VTT and VDDQ supplies may be enabled independently.
When EN/PSV is tied to VCCA the VDDQ controller is
enabled in power-save mode. When the EN/PSV pin is
fl oated, an internal resistor divider activates the VDDQ
controller with power-save disabled. If PSAVE is enabled,
the SC480 PSAVE comparator looks for inductor current
to cross zero on eight consecutive cycles. Once observed,
the controller enters power-save and turns off the lowside MOSFET when the current crosses zero. To improve
the effi ciency and add hysteresis, the on-time is increased
by 20% in power-save. The effi ciency improvement at light
loads more than offsets the disadvantage of slightly higher
output ripple. If the inductor current does not cross zero
on any switching cycle, the controller immediately exits
power-save. Since the controller counts zero crossings,
the converter can sink current as long as the current does
not cross zero on eight consecutive cycles. This allows the
output voltage to recover quickly in response to negative
load steps even when power-save is enabled.
VDDQ Voltage Selection
VDDQ voltage is set using the FB pin. Grounding FB sets
VDDQ to fi xed 2.5V. Connecting FB to +5V sets VDDQ to fi xed 1.8V. VDDQ can also be adjusted from 1.5 to 3.0V
using external resistors, see Figure 2. The voltage at FB is
then compared to the internal 1.5V reference.
Referencing Figure 2, the equation for setting the output
voltage is:
R
2
1V)(UTOx
5.1
R3
Current Limit Circuit
Current limiting of the SC480 can be accomplished in two
ways. The on-state resistance of the low-side MOSFETs
can be used as the current sensing element, or a sense
resistor in the low-side source can be used if greater accuracy is desired. RDSON sensing is more effi cient and
less expensive. In both cases, the R
resistor between
ILIM
the ILIM pin and LX sets the over-current threshold. This
resistor R
is connected to a 10μA current source within
ILIM
the SC480 which is turned on when the low-side MOSFET
turns on. When the voltage drop across the sense resistor
or low-side MOSFET equals the voltage across the R
ILIM
resistor, current limit will activate. The high-side MOSFET will
not be allowed to turn on until the voltage drop across the
sense element (resistor or MOSFET) falls below the voltage
across the R
resistor.
ILIM
The current sensing circuit actually regulates the inductor
valley current, see Figure 3. This means that if the current
limit is set to 10A, the peak current through the inductor
would be 10A plus the peak ripple current, and the average
current through the inductor would be 10A plus 1/2 the
peak-to-peak ripple current.
I
PEAK
+5V+VIN
+
C1
Q1
L1
D2
Q2
+
C3
Vout
BST
DH
ILIM
VDDP
PGND
SC480
D1
C2
LX
DL
R1
Figure 4
The schematic of RDSON sensing circuit is shown in Figure
4 with R
= R1 and RDS
ILIM
of Q2.
ON
Similarly, for resistor sensing, the current through the lower
MOSFET and the source sense resistor develops a voltage
that opposes the voltage developed across R
voltage developed across the R
drop across R
an over-current exists and the high-side
ILIM,
resistor reaches voltage
SENSE
. When the
ILIM
MOSFET will not be allowed to turn on. The over-current
equation when using an external sense resistor is:
R
ILx
OC
Aƫ10Valley
ILIM
R
SENSE
Schematic of resistor sensing circuit is shown in Figure 5
with R
The VDDQ controller has a power good (PGD) output.
Power good is an open-drain output and requires a pullup resistor. When the output voltage is +16%/-10% from
its nominal voltage, PGD gets pulled low. It is held low
until the output voltage returns to within +16%/-10% of
nominal. PGD is also held low during start-up and will not
be allowed to transition high until soft-start is over and the
output reaches 90% of its set voltage. There is a 5μs delay
built into the PGD circuit to prevent false transitions.
Output Over-Voltage Protection
When the VDDQ output exceeds 16% of its set voltage,
the low-side MOSFET is latched on. It stays latched and
the SMPS stays off until the EN/PSV input is toggled or
VCCA is recycled. There is a 5μs delay built into the OV
protection circuit to prevent false transitions. During a
VDDQ OV shutdown, VTT is alive until VDDQ falls to typically
0.4V, at which point VTT is tri-stated.
When VTT exceeds 12% above its set voltage, the VTT
regulator will tristate. There is a 50μs delay to prevent false
OV trips due to transients or noise. The VDDQ regulator
continues to operate after VTT OV shutdown. The VTT OV
condition is removed by toggling VTTEN or EN/PSV, or by
recycling VCCA.
Smart Over-Voltage Protection
In some applications, the active loads on VDDQ can
actually leak current into VDDQ. If PSAVE mode is enabled
at very light loading, this leak can cause VDDQ to slowly
rise and reach the OV threshold, causing a hard shutdown.
To prevent this, the SC480 uses Smart OVP to prevent
this. When VDDQ exceeds 8% above nominal, DL drives
high to turn on the low-side MOSFET, which starts to draw
current from VDDQ via the inductor. When VDDQ drops
to the FB trip point, a normal TON switching cycle begins.
This prevents a hard OV shutdown.
Output Under-Voltage Protection
When VDDQ falls 30% below its set point for eight clock
cycles, the VDDQ output is shut off; the DL/DH drives are
pulled low to tristate the MOSFETS, and the SMPS stays
off until the Enable input is toggled or VCCA is recycled.
When VTT is 12% below its set voltage the VTT output is
tristated. There is a 50μs delay for VTT built into the UV
protection circuits to prevent false transitions.
POR, UVLO and Soft-Start
An internal power-on reset (POR) occurs when VCCA exceeds 3V, resetting the fault latch and soft-start counter,
and preparing the PWM for switching. VCCA under-voltage
lockout (UVLO), circuitry inhibits switching and tristates the
drivers until VCCA rises above 4.2V. At this time the circuit
will come out of UVLO and begin switching and the softstart circuit will progressively limit the output current over
a pre-determined time period. The ramp occurs in four
steps: 25%, 50%, 75% and 100%, thereby limiting the slew
rate of the output voltage. There is 100mV of hysteresis
built into the UVLO circuit and when VCCA falls to 4.1V the
output drivers are shutdown and tristated.
MOSFET Gate Drivers
The DH and DL drivers are optimized for moderate, highside, and larger low-side power MOSFETs. An adaptive
dead-time circuit monitors the DL output and prevents the
high-side MOSFET from turning on until DL is fully off, and
conversely, monitors the DH output and prevents the lowside MOSFET from turning on until DH is fully off. (Note:
be sure there is low resistance and low inductance between the
DH and DL outputs to the gate of each MOSFET.)
Design Procedure
Prior to designing a switch mode supply for a notebook computer, the input voltage, load current, switching frequency
and inductor ripple current must be specifi ed.
Input Voltage Range
The maximum input voltage (VIN
highest AC adaptor voltage. The minimum input voltage
(VIN
) is determined by the lowest battery voltage after
MIN
accounting for voltage drops due to connectors, fuses and
battery selector switches.
Maximum Load Current
There are two values of load current to consider: continuous load current and peak load current. Continuous load
current has more to do with thermal stresses and therefore drives the selection of input capacitors, MOSFETs
and commutation diodes. Peak load current determines
instantaneous component stresses and fi ltering require-
ments such as, inductor saturation, output capacitors and
design of the current limit circuit.
Switching frequency determines the trade-off between
size and effi ciency. Higher frequency increases switch-
ing losses in the MOSFETs, since losses are a function of
F*VIN2. Knowing the maximum input voltage and budget
for MOSFET switches usually dictates the fi nal design.
Inductor Ripple Current
Low inductor values result in smaller size, but create higher ripple current and are less effi cient because of the high
AC current fl owing in the inductor. Higher inductor values
do reduce the ripple current and are more effi cient, but
are larger and more costly. The selection of the ripple current is based on the maximum output current and tends
to be between 20% to 50% of the maximum load current.
Again, cost, size and effi ciency all play a part in the selec-
tion process.
Stability Considerations
Unstable operation shows up in two related but distinctly
different ways: double pulsing and fast-feedback loop instability. Double-pulsing occurs due to noise on the output
or because the ESR is too low, causing insuffi cient voltage
ramp in the output signal. This causes the error amplifi er to
trigger prematurely after the 400ns minimum off-time has
expired. Double-pulsing will result in higher ripple voltage at
the output, but in most cases is harmless. In some cases,
however, double-pulsing can indicate the presence of loop
instability, which is caused by insuffi cient ESR. One simple
way to solve this problem is to add some trace resistance
in the high current output path. A side effect of doing this
is output voltage droop with load. Another way to eliminate
doubling-pulsing is to add a 10pF capacitor across the
upper feedback resistor divider network. This is shown in
Figure 6, by capacitor C4 in the schematic. This capacitance
should be left out until confi rmation that double-pulsing ex-
ists. Adding this capacitance will add a zero in the transfer
function and should eliminate the problem. It is best to
leave a spot on the PCB in case it is needed.
Loop instability can cause oscillations at the output as a
response to line or load transients. These oscillations can
trip the over-voltage protection latch or cause the output
voltage to fall below the tolerance limit.
The best way for checking stability is to apply a zero to
full load transient and observe the output voltage ripple
envelope for overshoot and ringing. Over one cycle of
ringing after the initial step is a sign that the ESR should
be increased.
BST
DH
ILIM
VDDP
DL
PGND
SC480
+5V
14
13
12
LX
11
10
9
8
+VIN
+
D1
C2
R1
C1
Q1
L1
D2
Q2
FBK
+
C3
R2
R3
C4
10pF
0.5V - 5.5V
Figure 6
SC480 ESR Requirements
The constant on-time control used in the SC480 regulates
the ripple voltage at the output capacitor. This signal
consists of a term generated by the output ESR of the
capacitor and a term based on the increase in voltage
across the capacitor due to charging and discharging
during the switching cycle. The minimum ESR is set to
generate the required ripple voltage for regulation. For most
applications the minimum ESR ripple voltage is dominated
by PCB layout and the properties of SP or POSCAP type
output capacitors. For applications using ceramic output
capacitors, the absolute minimum ESR must be considered.
If the ESR is low enough the ripple voltage is dominated
by the charging of the output capacitor. This ripple voltage
lags the on-time due to the LC poles and can cause double
pulsing if the phase delay exceeds the off-time of the
converter. Referring to Figure 5 on Page 11, the equation
for the minimum ESR as a function of output capacitance
and switching frequency and duty cycle is:
The output voltage adjust range for continuous-conduction
operation is limited by the fi xed 400nS (typical) Minimum
Off-time One-shot. For best dropout performance, use the
slowest on-time setting of 200KHz. When working with
low input voltages, the duty-factor limit must be calculated
using worst-case values for on and off times. The IC duty-
factor limitation is given by:
DUTY
Be sure to include inductor resistance and MOSFET onstate voltage drops when performing worst-case dropout
duty-factor calculations.
SC480 System DC Accuracy (VDDQ Controller)
Three IC parameters affect VDDQ accuracy: the internal
1.5V reference, the error comparator offset voltage, and
the switching frequency variation with line and load.
The internal 1%, 1.5V reference contains two error
components, a 0.5% DC error and a 0.5% supply and
temperature error. The error comparator offset is trimmed
so that it trips when the feedback pin is nominally 1.5
volts +/-1% at room temperature. The comparator offset
trim compensates for any DC error in the reference. Thus,
the percentage error is the sum of the reference variation
over supply and temperature and the offset in the error
comparator, or 1.5% total.
The on-time pulse in the SC480 is calculated to give a
pseudo-fi xed frequency. Nevertheless, some frequency
variation with line and load can be expected. This variation
changes the output ripple voltage. Because constant ontime converters regulate to the valley of the output ripple,
½ of the output ripple appears as a DC regulation error.
For example, If the output ripple is 50mV with VIN = 6
volts, then the measured DC output will be 25mV above
the comparator trip point. If the ripple increases to 80mV
with VIN = 25 volts, then the measured DC output will
be 40mV above the comparator trip. The best way to
minimize this effect is to minimize the output ripple.
To compensate for valley regulation it is often desirable
to use passive droop. Take the feedback directly from the
output side of the inductor, incorporating a small amount
(MIN)TON
(MAX)TOFF(MIN)TON
of trace resistance between the inductor and output
capacitor. This trace resistance should be optimized
so that at full load the output droops to near the lower
regulation limit. Passive droop minimizes the required
output capacitance because the voltage excursions due
to load steps are reduced.
Board components and layout also infl uence DC accuracy.
The use of 1% feedback resistors contributes up to 1%
error. If tighter DC accuracy is required use 0.1% feedback
resistors.
The output inductor value may change with current. This
will change the output ripple and thus the DC output
voltage (it will not change the frequency).
Switching frequency variation with load can be minimized
by choosing lower RDS
will cause the switching frequency to increase as the load
current increases. This will reduce the ripple and thus
the DC output voltage. This inherent droop should be
considered when deciding if passive droop is required, or
if passive droop is desired in order to further reduce the
output capacitance.
Output DC Accuracy (VTT Output)
The VTT accuracy compared to VDDQ is determined by
two parameters: the REF output accuracy, and the VTT
output accuracy with respect to REF. The REF output
is generated internally from the VDDQS (sense input),
and tracks VDDQS with 2% accuracy. This REF output
becomes the reference for the VTT regulator. The VTT
regulator then tracks REF within +/-40mV (typically zero).
The total VTT/VDDQ tracking accuracy is then:
DDR Reference Buffer
The reference buffer is capable of sourcing 10mA. The
reference buffer has a class A output stage and therefore
will not sink signifi cant current; there is an internal 50 kΩ
(typical) pulldown to ground. If higher current sinking is
required, an external pulldown resistor should be added.
Make sure that the ground side of this pulldown is tied to
the VTT ground plane near the PGND2 pin.
For stability, place a 10Ω/1μF series combination from
REF to VSSA. If REF load capacitance exceeds 1μF, place
at least 10Ω in series with the load capacitance to prevent
instability. It is possible to use only one 10Ω resistor, by
connecting the load capacitors in parallel with the 1μF,
and connecting the load REF to the capacitor side of the
10Ω resistor. (See the Typical Application Circuit on Page
1.) Note that this resistor creates an error term when REF
has a DC load. In most applications this is not a concern
since the DC load on REF is negligible.
Design Procedure
Prior to designing a switching output and making component selections, it is necessary to determine the input
voltage range and output voltage specifi cations. To dem-
onstrate the procedure, the output for the schematic in
Figure 7 on page 18 will be designed.
The maximum input v
oltage (V
) is determined by the
BAT(MAX)
highest AC adaptor voltage. The minimum input voltage
(V
) is determined by the lowest battery voltage af-
BAT(MIN)
ter accounting for voltage drops due to connectors, fuses
and battery selector switches. For the purposes of this
design example we will use a VBAT range of 8V to 20V to
design VDDQ.
ª
t
«
N)ON_VBAT(MI
«
¬
12
10 3.3
x
R
tON
3
1037
xxx
V
V
OUT
º
»
»
)MIN(BAT
¼
9
s
1050
x
and,
ª
t
«
X)ON_VBAT(MA
«
¬
12
10 3.3
x
R
tON
3
1037
xxx
V
From these values of tON we can calculate the nominal
switching frequency as follows:
tON is generated by a one-shot comparator that samples
V
via R
BAT
, converting this to a current. This current is
tON
used to charge an internal 3.3pF capacitor to V
equations above refl ect this along with any internal com-
ponents or delays that infl uence tON. For our example we
select R
= 1MΩ:
tON
OUT
x
t
Hz
·
¸
X)ON_VBAT(MA
¹
. The
OUT
Four parameters are needed for the design:
Nominal output voltage, V
1.
. We will use 1.8V with
OUT
internal feedback resistors (FB pin tied to VCCA).
Static (or DC) tolerance, TOLST (we will use +/-2%).
2.
Transient tolerance, TOLTR and size of transient (we
3.
will use +/-8% for a 10A to 5A load release for this
demonstration).
Maximum output current, I
4.
(we will design for 10A).
OUT
Switching frequency determines the trade-off between
size and effi ciency. Increased frequency increases
the switching losses in the MOSFETs, and losses are a
function of VBAT2. Knowing the maximum input voltage
and budget for MOSFET switches usually dictates where
the design ends up. The default R
values of 1MΩ and
tON
715k Ω are suggested only as a starting point.
The fi rst thing to do is to calculate the on-time, tON, at
V
BAT(MIN)
and V
, since this depends only upon V
BAT(MAX)
BAT
, V
OUT
and RtON.
t
ON_VBAT(MIN)
f
SW_VBAT(MIN)
= 820ns and, t
= 274kHz and f
ON_VBAT(MAX)
SW_VBAT(MAX)
= 358ns
= 251kHz
Now that we know tON we can calculate suitable values for
the inductor. To do this we select an acceptable inductor
ripple current. The calculations below assume 50% of I
which will give us a starting place.
We will select an inductor value of 1.5μH to reduce the
ripple current, which can be calculated as follows:
and,
For our example:
I
RIPPLE_VBAT(MIN)
= 3.39A
P-P
and I
RIPPLE_VBAT(MAX)
= 4.34A
P-P
From this we can calculate the minimum inductor current
rating for normal operation:
For our example:
I
INDUCTOR(MIN)
= 12.2A
(MIN)
Next we will calculate the maximum output capacitor
equivalent series resistance (ESR). This is determined by
calculating the remaining static and transient tolerance
allowances. Then the maximum ESR is the smaller of the
calculated static ESR (R
(R
ESR_TR(MAX)
Where ERR
):
is the static output tolerance and ERR
ST
ESR_ST(MAX)
) and transient ESR
is
DC
the DC error. The DC error will be 1% plus the tolerance
of the internal feedback. (Use 2% for external feedback,
which is 1% plus another 1% for the external resistors.)
For our example:
ERR
= 36mV and, ERR
ST
R
ESR_ST(MAX)
= 8.3mΩ
= 18mV, therefore,
DC
where ERRTR is the transient output tolerance. For this
case, I
is the load transient of 5A (10A - 5A).
TRANS
For our example:
ERRTR = 144mV and ERR
R
ESR_TR(MAX)
= 17.6mΩ for a full 5A load transient.
= 18mV, therefore,
DC
We will select a value of 6mΩ maximum for our design,
which would be achieved by using two 12mΩ output capacitors in parallel. Now that we know the output ESR we
can calculate the output ripple voltage:
ESR
I
x
PPV)MIN(TRIPPLE_VBA
V
R
)MIN(TRIPPLE_VBA
and,
ESR
I
x
PPV)MAX(TRIPPLE_VBA
V
R
)MAX(TRIPPLE_VBA
For our example:
V
RIPPLE_VBAT(MAX)
= 20mV
and V
P-P
RIPPLE_VBAT(MIN)
= 26mV
P-P
Note that in order for the device to regulate in a controlled
manner, the ripple content at the feedback pin, VFB, should
be approximately 15mV
no smaller than 10mV
at minimum V
P-P
. Note that the voltage ripple at
P-P
, and worst case
BAT
FB is smaller than the voltage ripple at the output capacitor, due to the resistor divider. Also, when using internal
feedback (FB pin tied to 5V or GND), the FB resistor divider is actually inside the IC. If V
the FB point is less than 15mV
- whether internal or ex-
P-P
RIPPLE_VBAT(MIN)
as seen at
ternal FB is used - the above component values should be
revisited in order to improve this. For our example, since
the internal divider reduces the ripple signal by a factor of
(1.5V/1.8V), the internal FB ripple values are then 17mV
and 22mV, which is above the 15mV minimum.
When using external feedback, and with VDDQ greater
than 1.5V, a small capacitor, C
with the top feedback resistor, R
ripple at VFB is large enough. C
than 100pF. The value of C
lows, where R
calculating the value of Z
Since our example uses internal feedback ,this method
cannot be used, however the voltage seen at the internal
FB point is already greater than 15mV.
Next we need to calculate the minimum output capacitance required to ensure that the output voltage does not
exceed the transient maximum limit, POSLIMTR, starting
from the actual static maximum, V
OUT_ST_POS
, when a load
release occurs:
V
OUT_ST_POS
V
OUT
ERR
DC
V
For our example:
V
OUT_ST_POS
= 1.818V,
POSLIMx
TR
V
OUT
TOL
TR
V
Where TOLTR is the transient tolerance. For our example:
POSLIMTR = 1.944V,
We will select 440μF, using two 220μF, 12mΩ capacitors
in parallel.
Next we calculate the RMS input ripple current, which is
largest at the minimum battery voltage:
I
VVVIxx
OUT)MIN(BATOUT)RMS(IN
OUT
V
A
RMS
MIN_BAT
For our example:
I
IN(RMS)
= 4.17A
RMS
Input capacitors should be selected with suffi cient ripple
current rating for this RMS current, for example a 10μF,
1210 size, 25V ceramic capacitor can handle approximately 3A
. Refer to manufacturer’s data sheets and
RMS
derate appropriately.
Finally, we calculate the current limit resistor value. As described in the current limit section, the current limit looks
at the “valley current”, which is the average output current minus half the ripple current.
I
VALLEY
I
OUT
I
2
)MIN(TRIPPLE_VBA
A
The minimum output capacitance is calculated as follows:
I
init
I
I
)MAX(OUT
2
)MAX(TRIPPLE_VBA
A
The ripple at low battery voltage is used because we want
to make sure that current limit does not occur under normal operating conditions.
and,
Ifinal
2
F
2
For our example:
I
VALLEY
2
C
OUT(MIN)
L
x
Iinit
POSLIM
TR
2
V
OUT_ST_POS
This calculation assumes the condition of a full-load to noload step transient occurring when the inductor current is
The junction temperature of the device may be calculated
as follows:
T
where TJ is the junction temperature, T
temperature, PD is the total SC480 device dissipation,
The SC480 device dissipation can be determined using:
PD = VCCA • ICCA + VDDP • IDDP + VTT •|ITT|
The fi rst two terms are losses for the analog and gate drive
circuits and generally do not present a thermal problem.
Typical ICCA (VCCA operating current) is roughly 1.5mA,
which creates 7.5mW loss from the 5V VCCA supply. The
VDDP supply current is used to drive the MOSFETs and
can be much higher, on the order of 30mA, which can
create up to 150mW of dissipation.
The last term, VTT * |ITT|, is the most signifi cant term
from a thermal standpoint. The VTT regulator is a linear
device and will dissipate power proportional to the VTT
current and the voltage drop across the regulator. If VTT
= VDDQ/2, then the voltage drop across the regulator
is always VDDQ2, regardless of whether the regulator is
sinking or sourcing current. In either case the power lost
in the VTT regulator is VTT * |ITT|. The average or longterm value for ITT should be used.
The thermal resistance of the MLPQ package is affected
by PCB layout and the available ground planes and vias
which conduct heat away. A typical value is 29°C/watt.
One (or more) ground planes are recommended to
minimize the effect of switching noise and copper losses,
and maximize heat dissipation. The IC ground reference,
VSSA, should be connected to PGND1 and PGND2 as a
star connection at the thermal pad, which in connects
using 4 vias to the ground plane. All components that are
:
referenced to VSSA should connect to it directly on the
chip side, and not through the ground plane.
VDDQ: The feedback trace must be kept far away from
noise sources such as switching nodes, inductors and
gate drives. Route the feedback trace in a quiet layer if
possible, from the output capacitor back to the chip. Chip
supply decoupling capacitors (VCCA, VDDP) should be
located next to the pins (VCCA/VSSA, VDDP/PGND1) and
connected directly to them on the same side.
VTT: Because of the high bandwidth of the VTT regulator,
proper component placement and routing is essential to
prevent unwanted high-frequency oscillations which can
be caused by parasitic inductance and noise. The input
capacitors should be located at the VTT input pins (VTTIN
and PGND2), as close as possible to the chip to minimize
parasitics. Output capacitors should be directly located at
the VTT output pins (VTT and PGND2). The routing of the
feedback signal VTTS is critical. The trace from VTTS (pin
2) should be connected directly to the output capacitor
that is farthest from VTT (pin24); route this signal away
from noise sources such as the VDDQ power train or highspeed digital signals.
The switcher power section should connect directly to the
ground plane(s) using multiple vias as required for current
handling (including the chip power ground connections).
Power components should be placed to minimize loops
and reduce losses. Make all the connections on one side
of the PCB using wide copper fi lled areas if possible. Do
not use “minimum” land patterns for power components.
Minimize trace lengths between the gate drivers and the
gates of the MOSFETs to reduce parasitic impedances
(and MOSFET switching losses); the low-side MOSFET is
most critical. Maintain a length to width ratio of <20:1 for
gate drive signals. Use multiple vias as required by current
handling requirement (and to reduce parasitics) if routed
on more than one layer. Current sense connections must
always be made using Kelvin connections to ensure an
accurate signal. The layout can be generally considered
in three parts; the control section referenced to VSSA, the
VTT output, and the switcher power section.
Looking at the control section fi rst, locate all components
referenced to VSSA on the schematic and place these
components at the chip. Connect VSSA using a wide
(>0.020”) trace. Very little current fl ows in the chip ground
therefore large areas of copper are not needed. Connect
the VSSA pin directly to the thermal pad under the device
as the only connection from PGND1 and PGND2 from
VSSA.
Decoupling capacitors for VCCA/VSSA and VDDP/PGND1
should be placed is as close as possible to the chip. The
feedback components connected to FB, along with the
VDDQ sense components, should also be located at the
chip. The feedback trace from the VDDQ output should
route from the top of the output capacitors, in a quiet
layer back to the FB components.
Next, looking at the switcher power section, there are a
few key guidelines to follow:
Finally, connecting the control and switcher power sections
should be accomplished as follows:
Route VDDQ feedback trace in a “quiet” layer, away
1.
from noise sources.
Route DL, DH and LX (low side FET gate drive, high
2.
side FET gate drive and phase node) to the chip using
wide traces with multiple vias if using more than one
layer. These connections are to be as short as possible
for loop minimization, with a length to width ratio less
than 20:1 to minimize impedance. DL is the most
critical gate drive, with power ground as its return
path. LX is the noisiest node in the circuit, switching
between VBAT and ground at high frequencies, thus
should be kept as short as practical. DH has LX as its
return path.
BST is also a noisy node and should be kept as short
3.
as possible.
Connect PGND1 pins on the chip directly to the VDDP
4.
decoupling capacitor and then drop vias directly to
theground plane. Locate the current limit resistor at
the chip with a kelvin connection to the phase node.
There should be a very small input loop, well
1.
decoupled.
The phase node should be a large copper pour, but
2.
still compact since this is the noisiest node.
Input power ground and output power ground should
3.
not connect directly, but through the ground planes
instead.