MIC3230 Datasheet, PDF (15/20 Pages) Micrel Semiconductor – Constant Current Boost Controller for Driving High Power LEDs

Micrel, Inc.
and for R_SLC. If I_{in _ actualLimit} is too high than use a higher
value for R_CS. The calculated value of R_CS for a 1.9A
current limit was 179mΩ. In this example, we have
chosen a lower value which results in a higher current
limit. If we use a higher standard value the current limit
will have a lower value. The designer does not have the
same choices for small valued resistors as with larger
valued resistors. The choices differ from resistor
manufacturers. If too large a current sense resistor is
selected, the maximum output power may not be able to
be achieved at low input line voltage levels. Make sure
the inductor will not saturate at the actual current limit
I _{in _ actualLimit} .
Perform a check at I_IN=2.34Apk.
V_{IS _ PIN} = 250μA × (0.78)× 511Ω + 2.34A × 150mΩ = 0.45V
Maximum Power dissipated in R_CS is;
Eq. (17)
P_RCS = I _{RCS _ RMS} ² × R_CS
Eq. (18)
I_{RCS _ RMS _max} = I_{FET _ RMS _ max} =
D_{⎜⎜⎛}I_IN
⎝
_
AVE
_
2
max
+
I_L
2
_ PP
12
⎟⎞
⎟
⎠
I = _{RCS _ RMS}
0.78_{⎜⎜⎝⎛}1.64 ²
+
0.26 ²
12
⎟⎞
⎟⎠
=
1.44A _ rms
P_RCS = 1.25² × .15 = 0.31watt
Use a 1/2 Watt resistor for R_CS.
Output Capacitor
In this LED driver application, the ILED ripple current is a
more important factor compared to that of the output
ripple voltage (although the two are directly related). To
find the C_OUT for a required ILED ripple use the following
calculation:
For an output ripple ILED_ripple = 20% of ILED_nom
ILED_ripple = 0.2 × 0.35 = 70mA
Eq. (19)
C_out
=
ILED_nom * D_nom * T
ILED_ripple * (R_adj + R_{LED _ total} )
Find the equivalent ac resistance R_{LED _ ac} from the
datasheet of the LED. This is the inverse slope of the
ILED vs. V_F curve i.e.:
Eq. (20)
R_{LED _ ac}
=
ΔV_F
ΔILED
In this example use R_{LED _ ac} = 0.1Ω for each LED.
If the LEDs are connected in series, multiply
R_{LED _ ac} = 0.1Ω by the total number of LEDs. In this
MIC3230/1/2
example of 6 LEDs, we obtain the following:
R_{LED _ total} = 6 × 0.1Ω = 0.6Ω
C_out
=
ILED_nom * D_nom * T
ILED_ripple * (R_adj + R_{LED _ total} )
= 4.1uF
Use the next highest standard value, which is 4.7uF.
There is a trade off between the output ripple and the
rising edge of the PWMD pulse. This is because
between PWM dimming pulses, the converter stops
pulsing and C_OUT will start to discharge. The amount that
C_OUT will discharge depends on the time between PWM
Dimming pluses. At the next PWMD pulse C_OUT has to
be charged up to the full output voltage V_OUT before the
desired LED current flows.
Input Capacitor
The input current is shown in Figure 5. For superior
performance, ceramic capacitors should be used
because of their low equivalent series resistance (ESR).
The input ripple current is equal to the ripple in the
inductor plus the ripple voltage across the input
capacitor, which is the ESR of C_IN times the inductor
ripple. The input capacitor will also bypass the EMI
generated by the converter as well as any voltage spikes
generated by the inductance of the input line. For a
required V_{IN_RIPPLE}:
Eq. (21)
C_IN
=
I_{IN _ PP}
8 ×V_{IN _ RIPPLE} × F_SW
=
(0.28A)
= 1.4μF
8 × 50mV × 500kHz
This is the minimum value that should be used. The
input capacitor should also be rated for the maximum
RMS input current. To protect the IC from inductive
spikes or any overshoot, a larger value of input
capacitance may be required and it is recommended that
ceramic capacitors be used. In this design example a
value of 4.7µF ceramic capacitor was selected.
MOSFET Selection
In this design example, the FET has to hold off an output
voltage maximum of 30V. It is recommended to use an
80% de-rating value on switching FETs, so a minimum
of a 38V FET should be selected. In this design
example, a 75V FET has been selected.
The switching FET power losses are the sum of the
conduction loss and the switching loss:
Eq. (22)
P_FET = P_{FET _ COND} + P_{FET _ SWITCH}
The conduction loss of the FET is when the FET is
turned on. The conduction power loss of the FET is
found by the following equation:
Eq. (23)
P_FET
_ COND
=
I_FET
2
_ RMS
× R_DSON
,
where
January 2009
15
M9999-011409-A