AN4717
STEVAL-ILD005V1: Trailing edge phase control rotary wall
dimmer based on STF17N62K3 power MOSFET
Anselmo Liberti
Introduction
The STEVAL-ILD005V1 evaluation board implements a simple and cost-effective MOS-based reverse
phase cut dimming solution for analog wall dimmers using a pair of STF17N62K3 power MOSFETs
controlled by a triple 3-input NOR gate to effectively dim resistive or capacitive lighting loads available
for domestic or industrial use (i.e. halogen lamps, electronic low voltage transformers and various
dimmable CFL/LED lamps). The AC analog control dimming technique allows on/off power switching
and up/down dimming functions through a single linear rotary potentiometer equipped with a mechanical
switch which opens the circuit at the minimum setting. Control of the board is achieved with a triple 3input NOR gate (pin-to-pin compatible with ST's HCF4025BE 14-lead dual in-line CMOS gate logic
device in a plastic SO-14 micropackage), which acts as a gate driving voltage signal source and
performs zero crossing detection and timing through the use of a few other passive components and
diodes.
The 15.5 A /620 V SuperMESH3 MOSFET STF17N62K3 in the TO-220 full package used for the lamp
current commutations is the most suitable device for the requirements of this trailing edge phase cut
dimmer switch application operating at line frequency (50/60 Hz) because its high total input
capacitance and sufficiently low on-resistance at the operating currents provides the best trade-off
between efficiency and EMI in a particularly simplified dimmer design solution.
The main features of this dimmer are:
operation for 2-wire wall dimmer
trailing-edge control only (compatible with all lamps commonly found on the market)
operation on 110 Vrms or 230 Vrms line voltage (±10 %) and 50 Hz or 60 Hz line frequency
dimmable power range (without heatsinks mounted on power MOSFETs)
40 W to 300 W for 230 Vrms line
15 W to 90 W for 110 Vrms line
maximum operating ambient temperature: 50 °C
power efficiency @ Max power - 230 VAC > 95%
power factor @ Max power > 0.98
control and regulation interface with a switched type single linear rotary potentiometer
compliance with EMC standards:
EN55015 (for European market): compliant with 300 W - 230 VAC halogen lamp
IEC 61000-4-5: criteria A for 2 kV surge tests
RoHS compliant
Other features include:
flicker-free dimming performance
smooth dimming operation and silent dimming control with no buzzing or audible noise
acceptable levels of EMI noise without the need for LC filters to reduce conducted emission
disturbances
low inrush current generation
August 2015
DocID027977 Rev 1
1/48
www.st.com
Operation principle
AN4717
This application note describes the operating principle and working conditions of the STEVAL-ILD005V1
board. An electrical evaluation of the dimmer performance is also given in terms of temperatures, turnoff energies and EMC standard testing.
Figure 1: STEVAL-ILD005V1 board: top and side views
2/48
DocID027977 Rev 1
AN4717
Contents
Contents
1
Operation principle ........................................................................... 5
1.1
Functional block diagram .................................................................... 5
1.1.1
1.2
2
3
4
Digital control logic .............................................................................7
Steady-state stage-wise circuit analysis........................................... 11
Operating conditions and selected components ........................ 20
2.1
Basic features ................................................................................... 20
2.2
MOSFET ........................................................................................... 20
2.3
Triple 3-inputs CMOS NOR gate ...................................................... 20
2.4
Linear rotary potentiometer .............................................................. 21
Board performances ...................................................................... 22
3.1
Experimental test and results @ 230VAC ........................................ 22
3.2
Electromagnetic compatibility test .................................................... 26
3.2.1
EMC conducted noise compliance according to EN 55015 ..............26
3.2.2
Surge voltage immunity test .............................................................27
Getting started ................................................................................ 29
4.1
Board connection .............................................................................. 29
4.2
User interface ................................................................................... 29
4.3
Steps for board testing ..................................................................... 30
4.4
Safety instruction .............................................................................. 30
5
Dimmer schematic .......................................................................... 31
6
Board layout and silkscreen .......................................................... 32
7
Bill of material ................................................................................. 33
8
Temperatures and turn-off energies ............................................. 34
9
Conduction times calculation ....................................................... 36
10
Regulation stage setting ................................................................ 40
11
Experimental test and results @ 110Vac...................................... 43
12
Conclusion ...................................................................................... 46
13
Revision history .............................................................................. 47
DocID027977 Rev 1
3/48
List of figures
AN4717
List of figures
Figure 1: STEVAL-ILD005V1 board: top and side views .......................................................................... 2
Figure 2: STEVAL-ILD005V1 functional block diagram ............................................................................ 5
Figure 3: Digital control logic functional diagram ...................................................................................... 7
Figure 4: Single triple input NOR gate schematic ..................................................................................... 8
Figure 5: Digital signals processed by the digital control logic ................................................................ 10
Figure 6: Theoretical waveforms and command sequences ................................................................... 11
Figure 7: Conduction phase on the positive main voltage envelope ....................................................... 13
Figure 8: Turn-off phase on the positive main voltage envelope ............................................................ 15
Figure 9: Conduction phase on the negative main voltage envelope ..................................................... 17
Figure 10: Turn-off phase on the negative main voltage envelope ......................................................... 19
Figure 11: Steady state operation with ~2.0 ms turn-on time @ minimum input power .......................... 22
Figure 12: Steady state operation with ~4.5 ms turn-on time @ medium input power ............................ 23
Figure 13: Steady state operation with ~ 8.2 ms turn-on time @ maximum input power ........................ 23
Figure 14: Voltage signals VPIN3 and VPIN5 in steady state operation @ minimum input power ......... 24
Figure 15: Voltage signals VPIN3 and VPIN5 in steady state operation @ medium input power ........... 24
Figure 16: Voltage signals VPIN3 and VPIN5 in steady state operation @ maximum input power ........ 24
Figure 17: Voltage signals VPIN3, VPIN9 and VPIN13 in steady state operation @ minimum input power
................................................................................................................................................................ 25
Figure 18: Voltage signals VPIN13 and VGS in steady state operation @ minimum input power .......... 26
Figure 19: Conducted noise @ minimum input power ............................................................................ 27
Figure 20: Surge voltage pulse @ medium input power ......................................................................... 28
Figure 21: Connection diagram .............................................................................................................. 29
Figure 22: Dimmer schematic ................................................................................................................. 31
Figure 23: Component layout – top and bottom assembly...................................................................... 32
Figure 24: PCB layout – top and bottom view ......................................................................................... 32
Figure 25: Temperatures and turn-off energies @ 230 VAC .................................................................. 34
Figure 26: Turn-off energy @ medium input power ................................................................................ 35
Figure 27: Regulation stage .................................................................................................................... 40
Figure 28: RC series circuit .................................................................................................................... 41
Figure 29: Steady state operation @ 110 VAC and minimum input power ............................................. 43
Figure 30: Steady state operation @ 110 VAC and medium input power ............................................... 43
Figure 31: Steady state operation @ 110 VAC and maximum input power ............................................ 44
Figure 32: Temperatures and turn-off energies @ 110 VAC .................................................................. 44
Figure 33: Turn-off energy@ medium input power ................................................................................. 45
4/48
DocID027977 Rev 1
AN4717
Operation principle
1
Operation principle
1.1
Functional block diagram
The simplified functional diagram of the board is shown in the figure below. In this dimmer
design, a single source driving network is used to simultaneously activate and deactivate
both gates of the two STF17N62K3 MOSFETs (Q1 and Q2) having sufficiently low RDS(on)
at the operating currents (RDS(on) max = 0.34 Ω @ VGS = 10 V and ID = 7.5 A at TC = 25 °C)
and connected to each other in an anti-series configuration or common source electrical
topology.
In steady state working conditions, the digital control logic stage supplies the gate driving
network with a voltage signal VDR equal to 15 V and pulsed at twice the mains frequency in
order to properly drive the MOSFETs with a PWM gate to source voltage signal VGS ~ 14V.
The dimmer operates in trailing edge operation mode and each MOSFET is therefore
always turned on instantanly at the beginning of the half line cycle in both positive and
negative main envelopes and maintains the on-state condition up to a controllable time
instant set internally to the same half cycle.
Figure 2: STEVAL-ILD005V1 functional block diagram
From the block diagram above, the following distinct functional stages interact with each
other to perform the overall dimming functionality:
Zero crossing detection network: consists of two passive sub-networks (each
consisting of resistors, capacitors and diodes) connected separately to both phase
and neutral lines and symmetrically arranged with respect to the ground plan. It allows
detection of the crossing point of the input main voltage with the zero level in the
ascendant or descendant slope phases for the respective positive and negative
portions of the main line envelope in order to properly activate the conduction phases
of the two MOSFETs.
DocID027977 Rev 1
5/48
Operation principle
AN4717
Supply stage: allows the generation of the dc voltage signal VCC clamped to around
15 V by a Zener diode in order to provide the VDD supply voltage necessary for the
triple 3-input NOR gate logic activation.
Gate driving network: consists of resistances only with an appropriate setting to
guarantee the proper driving condition for the MOSFET gates and to achieve the best
trade-off between efficiency and EMI performance.
Timing control and regulation stage: consists of the potentiometer and other
passive components (resistances and capacitances) opportunely set to allow the
regulation and control of the conduction time of the MOSFETs in order to perform the
dimming function.
Digital control logic stage: implemented with a triple 3-Input NOR Gate logic to
detect the zero crossing point instant time and activate the switch on phases of the
MOSFETs with the application of the gate driving voltage signal for an interval set by
the Timing control and regulation stage and depending on the specific operating
condition for the dimming function.
No EMI filtering stage (capacitor + inductor) is required at the input of the dimmer because
the gate driving network is set to slow the current falling edge down at the turn-off switching
transients in order to reduce the conducted electromagnetic noise and meet the
requirements of the EN55015 standard.
The electrical design is also optimized to limit the MOSFET power losses during conduction
phases and maximize the overall efficiency for the best trade-off between power dissipation
and EMI. Due to the alternating trend of both line mains and lamp current, as the two
MOSFETs are connected to each other in an anti-series configuration with a single gate
driving network implemented to simultaneously control both devices, in every half cycle of
the main voltage, the current always flows through one MOSFET in forward conduction
mode (or in the first quadrant operation and crossed by a positive drain current) and
simultaneously through the other MOSFET in reverse conduction mode (or in the third
quadrant operation and crossed by a negative drain current). In this way, it is also possible
to exploit the ohmic resistance reverse conduction capability of the MOSFET channel when
conducting a reverse direction current (from source to drain) by maintaining the same
RDS(on) characteristic as the conventional first quadrant operation.
Obviously, the reverse drain current flows through the MOSFET channel resistance by
shunting the parasitic body diode, provided that the VDS(on) value is small enough and lower
than the forward drop voltage of the intrinsic reverse diode, due to the appropraite low onresistance value in that particular operating condition for the MOSFET current. External
diodes (D5 and D6) connected in antiparallel to each MOSFET are also inserted to allow
an additional path for the sharing of the lamp current with the MOSFET (channel or intrinsic
diode) during the negative drain current conduction phase and in particular during the startup phases when the average lamp current levels can be higher than the steady state
working condition as a consequence of the cold state of the board and lighting load.
At 230 VAC input main voltage conditions, the minimum and maximum duration of the
settable duty cycles are approximately equal to 20% and 82% respectively of the whole half
line cycle.
For a positive mains half cycle, the lamp current passes through:
Q1 working on the first-quadrant operation or forward conduction with positive current
Q2 working on the third-quadrant operation and reverse conduction current (through
channel resistance or intrinsic diode)
D6 working on forward biasing condition
For negative mains half cycle, the lamp current passes through:
6/48
Q2 working on the first-quadrant operation or forward conduction with positive current;
DocID027977 Rev 1
AN4717
Operation principle
Q1 working on the third-quadrant operation and reverse conduction current (through
channel resistance or intrinsic diode);
D5 working on forward biasing condition.
A dead time for the current across the MOSFETs also occurs when the gate driving voltage
signal is zero and remains in this state until the end of the half line cycle with no current
flowing in the lamp.
1.1.1
Digital control logic
The following digital control logic stage functional diagram shows the interconnection of the
three 3-input NOR gate PINS and the relative PIN descriptions.
Figure 3: Digital control logic functional diagram
VDD
(PIN 14)
IN
PIN 3
PIN 4
IN
PIN 5
PIN 6
PIN 1
F
PIN 2
F
PIN 8
OUT
PIN 9
PIN 10 ( PIN 8)
OUT
PIN 13
PIN 10
IN
PIN 12
PIN 11 F
F
(PIN 7)
VSS
F
GSPG07072015DI1610
Table 1: Pin description
Pin no.
Name and function
3, 5, 13
Data inputs
9, 10 (≡ 8)
Data outputs
1≡6
Intermediate connection
2, 4, 11, 12
Connected to GND
7
Negative supply voltage VSS to GND
14
Positive supply voltage to VDD
The digital control is implemented by a logic circuit with 3 inputs and 2 outputs and the
associated logic function (calculating the correlation between each output and the inputs)
can be expressed via the following mathematical expressions (where Ix and 0y indicate the
respective input and output PINs to the logic circuit):
DocID027977 Rev 1
7/48
Operation principle
Equation 1
AN4717
𝑶 = (𝑰3 + 𝑰5 ) ∙ 𝑰13
{ 9
𝑶10 = ̅̅̅̅
𝑰13
All inputs are protected by a CMOS protection network as shown in the figure below,
referring to the single triple input NOR gate where the numbers in parentheses are terminal
numbers for the other gates (refer to the CMOS NOR gate CD4025B datasheet).
Figure 4: Single triple input NOR gate schematic
Therefore, for all the logic inputs, the voltage range (with reference to V SS = 0 V pin
voltage) is from -0.5 V to VDD +0.5 V, where the positive supply voltage VDD= VCC – VD7
and the maximum dc voltage signal VCC (generated by the supply stage) is clamped to
+15 V by a Zener diode between VCC and the negative supply voltage pin (VSS) of the logic.
The following truth table describes the Boolean function with all the possible operating
conditions (or states) that the logic circuit can theoretically assume in the steady state
working condition and expressed as correlations between inputs and outputs with binary
symbols 0 and 1.
8/48
DocID027977 Rev 1
AN4717
Operation principle
Table 2: Truth Table of the logic states
Truth table
Inputs
Pin
Logic
values
Outputs
State
Working phase
Operating
condition
3
5
13
9
10
0
0
0
0
1
1
Steady state (positive and
negative main voltage
envelope)
MOSFET on state
0
0
1
0
0
2
Steady state (positive and
negative main voltage
envelope)
MOSFET off state
(first phase)
0
1
0
0
1
3
X
0
1
1
1
0
4
Steady state (negative
main voltage envelope)
1
0
0
0
1
5
X
1
0
1
1
0
6
Steady state (positive main
voltage envelope)
1
1
0
0
1
7
X
1
1
1
1
0
8
X
MOSFET off state
(second phase)
MOSFET off state
(second phase)
In the above table, the symbol X indicates that the input combinations referring respectively
to STATES 3, 5, 7 and 8 never occur during regular dimmer operation and therefore can be
ignored.
The real operational states are:
STATE 1: occurs during the steady on-state operation of the MOSFETs
STATES 2, 4 and 6: occur during the steady off-state operation of the MOSFETs
All the operating conditions for the above states can also occur during the dimmer startup
phases when a low biasing condition for the driving voltage VDR at the output of the digital
control logic circuit imposes an inadequate gate-to-source voltage VGS (lower than the
nominal value of ~ 14 V) because the supply voltage VDD still hasn’t reached the steady
state value of 15 V during its rise in the startup transient.
The supply stage voltage signal and all the input/output data from/to the Triple 3-input NOR
gate logic circuit are detailed below and distinguished in Figure 5: "Digital signals
processed by the digital control logic" as digital signals of the logic circuit functionality:
Supply stage: voltage signal VCC
Input data to the logic circuit:
signals from cross detection stage: voltage signals VPIN3 and VPIN5 at PIN3 and
PIN5 respectively
signal from timing control stage: voltage signal VPIN13 at PIN13
Output data from the logic circuit:
signal to gate driving stage: voltage signal VPIN10 at PIN10
signal to timing control stage: voltage signal VPIN9 at PIN9
DocID027977 Rev 1
9/48
Operation principle
AN4717
Figure 5: Digital signals processed by the digital control logic
The theoretical waveforms and command sequences from/to the logic circuit are depicted
in Figure 6: "Theoretical waveforms and command sequences" together with the input main
voltage variation under steady state working conditions (STATE 2 is not included as it is a
fast transitory in comparison to the others), with:
VLINE-min, VLINE-max = maximum and minimum values of the main voltage
VZ-D8 = D8 diode breakdown voltage (equal to 15 V with a tolerance of 20 %)
VCC-max = maximum value of the supply voltage VCC (with VCC = VDD+VD7~VDD)
VREG = Voltage drop on the whole resistive regulation stage with the resistive
potentiometer and the other series and parallel resistances (R21, R22, R23 and R24)
VIH-PIN3, VIH-PIN5, VIH-PIN13 = high voltage levels of the input data at PIN3, PIN5 and
PIN13 (with VIH-PIN3 = VIH-PIN5~ VCC max by neglecting the voltage drops on the
resistances R13/R14 and on the diodes D1/D2/D7; VIH-PIN13~ VCC max-VREG)
VT = the threshold value (set to around VCC max/2) of each NOR gate input
VOH-PIN9, VOH-PIN10 = high voltage levels of the output data at PIN9 and PIN10 (with
VOH-PIN9 = VOH-PIN10 ~ VCC max)
Moreover, for all the low voltage levels of the input (VPIN3, VPIN5 and VPIN13) and output (VPIN9
and VPIN10), the data is assumed to be:
Equation 2
VIL-PIN3 = VIL-PIN5 = VIL-PIN13 = VOL-PIN9 = VOL-PIN10 = VSS = 0 V
10/48
DocID027977 Rev 1
AN4717
Operation principle
Figure 6: Theoretical waveforms and command sequences
1.2
Steady-state stage-wise circuit analysis
In a trailing edge phase-cut dimmer, the MOSFETs are turned on when the line voltage
(and therefore the current through the devices) passes through zero during the ascendant
or descendent slopes of its respective positive or negative envelopes. The dimmer uses a
timing control and regulation circuit to turn the current supply to the fixture on and off at
DocID027977 Rev 1
11/48
Operation principle
regular intervals, typically 100 or 120 times per second (for 50 Hz or 60 Hz line
frequencies), too fast to be perceived by the human eye.
AN4717
The circuit implements two STF17N62K3 MOSFETs as switches: when the switches are
simultaneously turned on, a significant current flows through the devices and the lamp,
while when the switches are simultaneously turned off, negligible current flows through the
circuit. The amount of time in which the current is on determines the brightness or dimming
effect of the lamp. In fact, by turning the switch on and off twice every line cycle, the rms
voltage applied to the lamp is reduced from 230 VAC to a lower value, so the rms current
that flows through the lamp is also reduced and produces less light.
During the dimmer startup phases, the MOSFETs working in the on state may operate
under a low biasing condition for the gate driving voltage signal because the supply voltage
VCC, such as all the input/output logic circuit PINs, hasn't properly reached the nominal
voltage of ~15 V.
This section provides a comprehensive description of the trailing edge phase cut dimmer
circuit operation through a stage-by-stage circuit analysis under steady state conditions,
divided into sequential functional phases according to the variation over a period of the
main voltage starting from the time instant at zero voltage value on the ascendant slope of
the positive envelope. For simplification purposes, STATE 2 is not included (considered a
fast transitory compared to the others during the normal operation) and the leakage
currents flowing on the diodes working in reverse-bias operation are neglected.
12/48
Conduction phase on the positive main voltage envelope (STATE 1): the zero
crossing detection network recognizes the instant when the main line passes through
the zero value on the ascendant voltage slope because all the three input signals to
the digital control logic circuit VPIN3, VPIN5 and VPIN13 are at the same low voltage
detected as zero logic level conditions (respectively I3=0, I5=0 and I13=0). In this
case, the turn on phase of the two MOSFETs is instantaneously activated by the logic
circuit that applies a high logic level signal (O10=1, VDR=VOH-PIN10 ~VCC max) at the
output of the PIN10 for the resistive driving network to the gates, maintaining a zero
logic level output at PIN9 (09 = 0, VPIN9 = 0 V).
Due to the slow discharging transient of the supply stage (i.e., paralleled capacitances
C3 and C4) through the regulation stage (with the resistive potentiometer and the
other series and parallel resistances R21, R22, R23 and R24), the voltage VPIN13
increases during the whole conduction phase via the charging current IREG-ON on the
capacitive node corresponding to PIN13. Therefore, the conduction phase lasts for the
time required for the voltage on PIN13 (VPIN13) to reach the threshold value of around
VCC max /2 starting from the zero value condition.
DocID027977 Rev 1
AN4717
Operation principle
Figure 7: Conduction phase on the positive main voltage envelope
DocID027977 Rev 1
13/48
Operation principle
AN4717
Turn-off phase on the positive main voltage envelope (STATE 2 and STATE 6):
STATE 2: when the voltage VPIN13 of the input signal at PIN13 reaches the
threshold value VT~Vccmax/2, the following changes occur in sequence:
a. input signal VPIN13 changes its status detected by the digital control logic
from low logic level (I13=0) to high logic level (I13=1);
b. output signal VPIN10 changes its status from high logic level (O10=1,
VPIN10~Vccmax) to zero logic level (O10=0, VPIN10 = 0 V).
Under this operating condition, the voltage signal VDR applied to the resistive driving
network for the MOSFET gates is turned off instantly.
STATE 6: As a consequence of the disabled condition for driving the MOSFETs,
the charge of the capacitance C1 placed on the cross detection network with a
current IDET-3 becomes involved and the following changes occur in sequence:
a. input signal VPIN3 changes its status from low logic level (I3=0) to high logic
level (I3=1, VPIN3~ VCC max)
b. output signal VPIN9 changes its status from zero logic level to high logic level
(O9=1, VPIN9~VCC max)
c.
input signal VPIN13 instantly rises to VIH-PIN13 = VCC max – Vreg and a very low
discharge current IREG-OFF continues to flow from the supply stage (paralleled
capacitances C3 and C4) towards the node associated with PIN13 involving
full discharge and voltage polarity inversion phases sequentially on
paralleled capacitances C5 and C6.
When the main voltage reaches zero during its positive descendant slope, the following
changes occur in sequence:
1.
2.
3.
capacitance C1 is fully discharged and the input signal VPIN3 changes status from high
logic level (I3=1 associated with VPIN3~ VCC max) to zero logic level (I3=0)
output signal VPIN9 changes status to zero logic level (O9=0), as the input signal VPIN5
is at low logic level (I5=0)
input signal VPIN13 instantly passes from high to zero logic level condition.
This last operating (STATE 1) condition, with all the three VPIN3, VPIN5 and VPIN13 input
signals at the same zero logic level condition (I3=0, I5=0 and I13=0), sets the digital control
logic circuit for the new conduction phase in the next half main cycle.
14/48
DocID027977 Rev 1
AN4717
Operation principle
Figure 8: Turn-off phase on the positive main voltage envelope
Conduction phase on the negative main voltage envelope (STATE 1): the zero
crossing detection network recognizes the instant when the main line passes through
the zero value on the descendant voltage slope because all the three input signals to
the digital control logic circuit VPIN3, VPIN5 and VPIN13 are at the same low voltage
detected as zero logic level conditions (respectively I3=0, I5=0 and I13=0). In this
case, the turn-on phase of the two MOSFETs is instantaneously activated by the logic
circuit that applies a high logic level signal to the output of PIN10 (O10=1,
DocID027977 Rev 1
15/48
Operation principle
AN4717
VDR=VPIN10~VCC max) for the resistive driving network to the gates, maintaining the
output at PIN9 at zero (09=0,VPIN9 = 0 V).
Due to the slow discharging transient of the supply stage (paralleled capacitances C3
and C4) through the regulation stage (resistive potentiometer and the series and
parallel resistances R21, R22, R23 and R24), the voltage VPIN13 rises during the whole
conduction phase via the charging current IREG-ON on the capacitive node associated
with PIN13. Therefore, the conduction phase lasts for the period required for VPIN13 to
reach the threshold value of around Vccmax /2 starting from the zero value condition.
16/48
DocID027977 Rev 1
AN4717
Operation principle
Figure 9: Conduction phase on the negative main voltage envelope
Turn-off phase on the negative main voltage envelope (STATE 2 and STATE 4):
STATE 2: when the input signal voltage VPIN13 reaches the threshold value
VT~Vccmax/2, the following changes occur in sequence:
a. input signal VPIN13 changes its status detected by the digital control logic from low
logic level (I13=0) to high logic level (I13=1)
b. output signal VPIN10 changes its status from high logic level (O10=1, VPIN10~Vccmax)
to zero logic level (O10=0, VPIN10 = 0 V)
DocID027977 Rev 1
17/48
Operation principle
AN4717
Under this operating condition, the voltage signal VDR applied to the resistive driving
network for the MOSFET gates is instantly turned off.
STATE 4: as a consequence of the disabled condition for driving the MOSFETs, the
charge of capacitance C2 placed on the cross detection network with current IDET-5
becomes involved and the following changes occur in sequence:
a. input signal VPIN5 changes its status from the low logic level (I5=0) to the high
logic level condition (I5=1, VPIN5~ Vccmax)
b. output signal VPIN9 changes from zero to high logic level (O9=1, VPIN9~Vccmax)
c. input signal VPIN13 voltage instantly rises to VIH-PIN13 = Vccmax – Vreg and a very low
discharge current IREG-OFF continues to flow from the supply stage (paralleled
capacitances C3 and C4) towards the node relating to PIN13 involving the full
discharge and voltage polarity inversion phases sequentially on the paralleled
capacitances C5 and C6.
When the main voltage reaches zero during its negative ascendant slope, the following
changes occur in sequence:
1.
2.
3.
capacitance C2 is fully discharged and input signal VPIN5 changes from high logic level
(I5=1, VPIN5~ Vccmax) to zero logic level (I5=0)
output signal VPIN9 changes to zero logic level (O9=0), as the input signal VPIN3 is in
the low logic level condition (I3=0)
input signal VPIN13 instantly passes from the high to zero logic level.
This last operating (STATE 1) condition with all the three input signals VPIN3, VPIN5 and
VPIN13 at the same zero logic level (respectively I3=0, I5=0 and I13=0) sets the digital
control logic circuit for the new conduction phase in the next half main cycle.
18/48
DocID027977 Rev 1
AN4717
Operation principle
Figure 10: Turn-off phase on the negative main voltage envelope
DocID027977 Rev 1
19/48
Operating conditions and selected components
2
Operating conditions and selected components
2.1
Basic features
AN4717
This board is designed to dim resistive or capacitive lighting loads available for domestic or
industrial use. The main features of this dimmer are:
2.2
operation for 2-wire wall dimmer
trailing-edge control only (compatible with all lamps commonly found on the market)
operation on 110 Vrms or 230 Vrms line voltage (±10 %) and 50 Hz or 60 Hz line
frequency
dimmable power range (without heatsinks mounted on MOSFETs)
40 W to 300 W for 230 Vrms line
15 W to 90 W for 110 Vrms line
maximum operating ambient temperature: 50 °C
power efficiency @ Max power - 230 VAC > 95%
power factor @ Max power > 0.98
control and regulation interface with a switched type single linear rotary potentiometer
supported loads
dimmable CFL/LED lamps
incandescent bulbs with halogen technology
electronic low voltage transformers dimmable by trailing edge phase control
MOSFET
The STF17N62K3 15.5 A, 620 V SuperMESH3 MOSFET in a TO-220 full package is the
most suitable device for the requirements of this trailing edge phase cut dimmer switch
application operating at line frequency (50/60 Hz) thanks to its high total input capacitance
(Ciss=3100 pF @ VDS = 50 V) and sufficiently low on-resistance at operating currents.
The main features of STF17N62K3 are:
2.3
extremely low on-resistance: RDS(on) max = 0.34 Ω with VGS = 10 V and ID = 7.5 A at
TC = 25 °C
high avalanche capability with IAR = 15.5 A and EAS = 260 mJ
extremely high dv/dt capability: dv/dt = 9 V/ns with ISD ≤ 15.5 A, di/dt ≤ 400 A/µs
zener-protected.
Triple 3-inputs CMOS NOR gate
The CMOS NOR gate CD4025B (pin-to-pin compatible with ST's HCF4025BE 14-lead dual
in-line CMOS gate logic device in a plastic SO-14 micropackage) fits very well with the light
dimmer application requirements by performing the control of the board.
Its main features are:
20/48
propagation delay time: tPD = 60 ns (typ.) at VDD = 10 V, CL = 50 pF
buffered inputs and outputs
standardized symmetrical output characteristic
100 % tested for maximum quiescent current at 20 V
5 V, 10 V and 15 V parametric ratings
maximum input current of 1 μA at 18 V over full package temperature range; 100 nA at
18 V and 25 °C
DocID027977 Rev 1
AN4717
2.4
Operating conditions and selected components
noise margin (over full package temperature range): 1 V at VDD=5 V, 2 V at VDD=10 V
and 2.5 V at VDD=15 V.
Linear rotary potentiometer
A single linear rotary potentiometer equipped with a mechanical switch which opens the
circuit at minimum setting is used to implement lamp current dimming and power regulation
with on/off power switch and up/down dimming functions.
The main characteristics include:
resistance range (LAW): 1 kΩ to 1 MΩ
resistance tolerance: ±20 % (±10 % per selection)
rated dissipation at 40 °C: 0.4 W
limiting element voltage: 500 V DC or AC RMS
terminal resistance: 5 Ω maximum
noise (ENR): 2 % maximum
mechanical rotation: 300°
operating temperature range: -25 to +70 °C
DocID027977 Rev 1
21/48
Board performances
AN4717
3
Board performances
3.1
Experimental test and results @ 230VAC
The MOS-based dimmer solution includes two STF17N62K3 MOSFETs connected in antiseries configuration and a single linear potentiometer to set the turn-on time (or duty cycle)
of the devices as in the following three operating conditions representing the minimum,
medium and maximum levels of lamp current dimming and input power absorbed by the
board:
minimum power: ~2.0 ms (or ~20 % of duty cycle in the semiperiod)
medium power: ~4.5 ms (or ~45 % of duty cycle in the semiperiod)
maximum power: ~8.2 ms (or ~82 % of duty cycle in the semiperiod)
The following figures show the waveforms acquired on the steady state operation with
300 W halogen technology incandescent bulbs at 230 VAC input mains for the working
conditions described above for the MOSFETs.
The waveforms were acquired by scope with the signal color codes:
Q1 drain current: signal ID1 = yellow
Q2 drain current: signal ID2 = blue
Q1 drain-source voltage: signal VDS1 = red
Q2 drain-source voltage: signal VDS2 = green
Figure 11: Steady state operation with ~2.0 ms turn-on time @ minimum input power
22/48
DocID027977 Rev 1
AN4717
Board performances
Figure 12: Steady state operation with ~4.5 ms turn-on time @ medium input power
Figure 13: Steady state operation with ~ 8.2 ms turn-on time @ maximum input power
The following figures show the waveforms acquired for the input signals to the digital logic
circuit, respectively the voltages on the pins PIN3 (VPIN3) and PIN5 (VPIN5) together with the
230 VAC input mains, during the steady state operation with halogen technology
incandescent bulbs at the minimum, medium and maximum input power conditions.
Waveforms were acquired by scope with the following signal color codes:
input signal on the PIN3: voltage VIN3 = green
input signal on the PIN5: voltage VIN5 = yellow
input main voltage: signal VLine = red
DocID027977 Rev 1
23/48
Board performances
AN4717
Figure 14: Voltage signals VPIN3 and VPIN5 in steady state operation @ minimum input power
Figure 15: Voltage signals VPIN3 and VPIN5 in steady state operation @ medium input power
Figure 16: Voltage signals VPIN3 and VPIN5 in steady state operation @ maximum input
power
24/48
DocID027977 Rev 1
AN4717
Board performances
The figure below shows the waveforms acquired for the digital logic circuit input and output
signals, respectively the voltages on the pins PIN3 (VPIN3), PIN13 (VPIN13) and PIN9 (VPIN9)
together with the 230 VAC input mains during the steady state operation with halogen
technology incandescent bulbs at the minimum input power condition and 230 V AC input
mains.
Waveforms were acquired by scope with the following signal color codes:
input signal on the PIN3: voltage VIN3 = green
input signal on the PIN13: voltage VIN13 = blue
output signal on the PIN9: voltage VOUT9 = yellow
input main voltage: signal VLine = red
Figure 17: Voltage signals VPIN3, VPIN9 and VPIN13 in steady state operation @ minimum
input power
The following figure shows the waveforms acquired respectively for the input digital logic
circuit signal of the voltage on PIN13 (VPIN13), the gate-to-source voltage signal VGS and the
dimmer input voltage signal Vin during the steady state operation with halogen technology
incandescent bulbs at the minimum input power condition and 230 V AC input mains.
Waveforms were acquired by scope with the following signal color codes:
input signal on the PIN13: voltage VIN13 = yellow
gate-source voltage: signal VGS = blue
dimmer input voltage: signal VIN = red
DocID027977 Rev 1
25/48
Board performances
AN4717
Figure 18: Voltage signals VPIN13 and VGS in steady state operation @ minimum input power
As can be seen, the MOSFET gate-source voltage is set to approx. 14 V during the
conduction phases in order to keep the channel of the devices activated while crossed by
the lamp current.
3.2
Electromagnetic compatibility test
3.2.1
EMC conducted noise compliance according to EN 55015
Conducted emission noise was tested according to standard EN 55015 under the following
conditions:
230 VRMS - 50 Hz mains voltage
300 W/230 VAC incandescent bulbs with halogen technology
tests were carried out with the rotary potentiometer set to different rotation angles
corresponding to different conduction angles/times of the MOSFETs and input power
levels absorbed by the dimmer
The following figure gives the measurement of the maximum conducted noise level
reached for the minimum input power level condition (40 W).
26/48
DocID027977 Rev 1
AN4717
Board performances
Figure 19: Conducted noise @ minimum input power
The above figure demonstrates that the conducted emission noise measured at the
minimum input power condition (i.e., worst case for the generation of the EMI disturbances)
was compliant with the EN55015 electronic standard specifications, with the QP and AV
detectors always below the corresponding QP and AV limits in the frequency range from
9 kHz to 30 MHz. As can be seen, thanks to the appropriate setting of the resistances on
the gate driving network to the MOSFETs, no LC noise input filter is needed to meet the
European EMC standard regulation in force for the conducted disturbances.
3.2.2
Surge voltage immunity test
Surge immunity tests were carried out on the MOS-based dimmer in order to analyze the
board performance when sequences of high energy but relatively slow transient
overvoltages are applied on the power lines. The tests were performed according to the
standard IEC 61000-4-5 relating to the immunity requirements by applying voltage surges
of amplitudes equal to 2 kV in both positive and negative polarities with different phase
angles shifted in relation to the input main voltage.
A 300 W/230 VAC halogen technology incandescent bulbs is connected in series to the
board running under normal operating conditions at 230VAC and 50Hz frequency.
The figure below shows the STF17N62K3 MOSFET response during the tests performed
when a 2 kV surge pulse with positive polarity is applied to the AC power port in differential
mode (symmetrical configuration as line-to-line between the phase and neutral lines) with
0º and 90º shifted phase angles respectively under medium input power working conditions
for the dimmer. The waveforms acquired by scope for the MOSFET Q1 are the drainsource voltage (VDS1 signal in red), the drain-source current (IDS1 signal in yellow) and
the input main voltage (VLine signal in green).
DocID027977 Rev 1
27/48
Board performances
AN4717
Figure 20: Surge voltage pulse @ medium input power
The MOSFET STF17N62K3 was able to withstand the surge pulses without any failure
condition and the dimmer continued to operate normally without any temporary degradation
or loss of functionality and with automatic return to normal operation after the disturbance
ceased.
28/48
DocID027977 Rev 1
AN4717
Getting started
4
Getting started
4.1
Board connection
The figure below shows the connection diagram of the STEVAL-ILD005V1 to the lamp to
be dimmed.
Please unplug the circuit from the line before setting the different connections.
Figure 21: Connection diagram
1.
2.
header J1 is used to connect the two-wire dimmer to the mains in series with the lamp
terminals T1 or T2 of the Header J1 can either be connected to the lamp or directly to
the mains
Appropriate connection of the lamp to mains neutral or line depends on the
applicable electrical safety regulations, even if the lamp must usually be
connected to neutral.
4.2
User interface
The user interface consists of a rotative potentiometer with integrated ON/OFF switch:
1.
2.
3.
the ON/OFF switch is used to turn the dimmer board and lamp on and off
the clockwise rotation of the potentiometer up to the maximum rotational angle
(representing the maximum resistive value) allows continuous power absorption and
light emission modification by increasing lamp brightness up to the maximum intensity
level
the anticlockwise rotation of the potentiometer up to the minimum rotational angle
(representing the minimum resistive value) allows continuous power absorption and
light emission modification by decreasing lamp brightness down to the minimum
intensity level
DocID027977 Rev 1
29/48
Getting started
4.3
AN4717
Steps for board testing
Follow this procedure to use the STEVAL-ILD005V1 board:
1.
2.
3.
4.
5.
connect one lamp terminal to one J1 header terminal (T1 or T2) as shown in Figure
21: "Connection diagram"
connect the other lamp terminal to a plug of the 230 Vac AC power source switched
off
connect the other J1 terminal to the other plug of the 230Vac AC power source
switched off
switch the 230Vac AC power source on
if the potentiometer switch opens the circuit in the OFF position, close the switch by
rotating the potentiometer clockwise to turn the dimmer board and lighting load on
At this point, the STEVAL-ILD005V1 is working correctly if:
4.4
the lamp brightness increases by rotating the potentiometer clockwise
the lamp brightness decreases by rotating the potentiometer in anticlockwise
the lamp brightness remains stable when no rotation is performed
Safety instruction
Follow this procedure to use the STEVAL-ILD005V1 board:
Waring: The high voltage levels used to operate the MOS-based dimmer evaluation
board could present a serious electrical shock hazard. This evaluation board must
be used in a suitable laboratory by qualified personnel only, familiar with the
installation, use, and maintenance of electrical power systems.
The STEVAL-ILD005V1 evaluation board is designed for demonstration purposes only,
and shall not be used either for domestic installation or for industrial installation.
30/48
DocID027977 Rev 1
AN4717
5
Dimmer schematic
Dimmer schematic
Figure 22: Dimmer schematic
A 5 A/250 VAC fuse F1 is mounted to protect MOSFETs in case of lamp flash-overs or
accidental short-circuits.
DocID027977 Rev 1
31/48
Board layout and silkscreen
6
AN4717
Board layout and silkscreen
Figure 23: Component layout – top and bottom assembly
Figure 24: PCB layout – top and bottom view
32/48
DocID027977 Rev 1
AN4717
7
Bill of material
Bill of material
Table 3: Bill of material
Reference
Part / value
U1
CD4025BM
Q1, Q2
STF17N62K3
P1A/P1B
1 MΩ/0.4 W
R1, R2, R3, R4, R5, R6, R7, R8, R9 R10, R11, R12
12 kΩ / 2 W
R13, R14
2.2 kΩ / 0.25 W
R15, R16
100 kΩ / 0.125 W
R17, R18
10 kΩ / 0.25 W
R19, R20
1 kΩ / 0.25 W
R21
1 MΩ / 0.25 W
R22
560 kΩ / 0.25 W
R23
150 kΩ / 0.25 W
R24
120 kΩ / 0.25 W
VR1
Varistor / 420 V
R26
0 Ω / 0.25 W
R27, R28
22 kΩ / 0.25 W
D1, D2, D3, D4, D7
S1M
D5, D6
1N4007
D8
BZV55B15
C1, C2
2.2 nF / 0.25 W
C4
47 µF / 25V
C3
100 nF / 0.25 W
C5
22 nF / 100VCC
C6
N.M.
Rfuse
5 A / 250 VAC
FUSE HOLDER
6.3 A / 250 VAC
J1
2-pin PCB screw terminal block
DocID027977 Rev 1
33/48
Temperatures and turn-off energies
8
AN4717
Temperatures and turn-off energies
The figure below shows the different temperatures and turn-off switching energies
measured for the STF17N62K3 MOSFETs working under the three different operating
conditions for minimum, medium and maximum levels of lamp current dimming and input
power absorbtion by the board connected to 300 W/230 VAC halogen technology
incandescent bulbs.
Figure 25: Temperatures and turn-off energies @ 230 VAC
The maximum case temperatures (approx. 76 °C without heatsinks) are reached in free air
conditions (at 25 °C room temperature) for the medium input power level operation and can
be reduced by mounting appropriate heatsinks in contact with the MOSFETs. Switching
losses can be reduced by decreasing the original value of the two gate drive resistances
R27 and R28 (set to 22 kΩ) to speed up the turn-off switching transients. This modification
implies a new noise emission condition for the conducted disturbances introduced by the
dimmer and may require the use of an appropriate EMI input filter to guarantee compliance
with the EN55015 standard.
The following figure depicts the turn-off transient detail for MOSFET Q2 switching under the
medium input power operating condition.
34/48
DocID027977 Rev 1
AN4717
Temperatures and turn-off energies
Figure 26: Turn-off energy @ medium input power
DocID027977 Rev 1
35/48
Conduction times calculation
9
AN4717
Conduction times calculation
Assuming a sinusoidal envelope over the period T = 1/freq for the current lamp waveform
under normal steady state conditions, as in:
Equation 3
𝐼𝑙𝑎𝑚𝑝 (𝑡) = 𝐼𝑙𝑎𝑚𝑝_𝑀𝐴𝑋 ∙ sin(2𝜋 ∙ 𝑓𝑟𝑒𝑞 ∙ 𝑡)
for 0<t<TCOND and T/2<t<T/2+TCOND
by neglecting the voltage drops on the RDS(on) channel resistances and on the forward
biased antiparallel diodes during the MOSFET conduction phases, we can consider the
mains voltage entirely applied on the lamp and therefore express the power P IN for
𝑇
𝐷 = 𝐶𝑂𝑁𝐷
and conduction time TCOND thus:
𝑇
Equation 4
𝑃𝐼𝑁 =
𝐼𝑙𝑎𝑚𝑝_𝑀𝐴𝑋 ∙ 𝑉𝑚𝑎𝑖𝑛_𝑀𝐴𝑋
1
[𝜋 ∙ 𝐷 − ( ∙ 𝑠𝑖𝑛(2𝜋 ∙ 𝐷) ∙ 𝑐𝑜𝑠(2𝜋 ∙ 𝐷))]
𝜋
2
For a 90° MOSFET conduction angle, 𝐷 =
𝑇𝐶𝑂𝑁𝐷
𝑇
1
= 4, so:
Equation 5
𝑃𝐼𝑁_90𝑑𝑒𝑔 =
1
∙𝐼
∙𝑉
4 𝑙𝑎𝑚𝑝_𝑀𝐴𝑋 𝑚𝑎𝑖𝑛_𝑀𝐴𝑋
The max value Ilamp_MAX for the current lamp signal (equalt to the maximum dimmer input
current IIN_MAX ) can be expressed with the following formula where PIN_90deg is the input
power level at 90° MOSFET conduction angle and Vmain_MAX is the maximum mains voltage
value:
Equation 6
𝐼𝑙𝑎𝑚𝑝𝑀𝐴𝑋 =
4 ∙ 𝑃𝐼𝑁_90𝑑𝑒𝑔
𝑉𝑚𝑎𝑖𝑛_𝑀𝐴𝑋
By inserting:
𝑃𝐼𝑁_90𝑑𝑒𝑔 = 𝑃𝐼𝑁_𝑀𝐸𝐷 + ∆% ∙ 𝑃𝐼𝑁_𝑀𝐸𝐷 ≅ 158𝑊 with PIN_MED = 150 W and ∆ % = 5 %
and
𝑉𝑚𝑎𝑖𝑛_𝑀𝐴𝑋 = 230 ∙ √2𝑉
We obtain:
Equation 7
𝐼𝑙𝑎𝑚𝑝_𝑀𝐴𝑋 =
4 ∙ 158
230 ∙ √2
≅ 1.94𝐴
The rms value of the lamp current Ilamp_RMS (equal to the rms value of the dimmer input
current IIN_RMS) can be expressed with the following formula where PIN_MAX is the maximum
input power, Vmain_RMS is the rms value of the mains voltage and PFMAX is the power factor
at the maximum input power:
Equation 8
𝐼𝑙𝑎𝑚𝑝_𝑅𝑀𝑆 = 𝐼𝐼𝑁_𝑅𝑀𝑆 =
36/48
𝑃𝐼𝑁_𝑀𝐴𝑋
𝑉𝑚𝑎𝑖𝑛_𝑅𝑀𝑆 ∙ 𝑃𝐹𝑀𝐴𝑋
DocID027977 Rev 1
AN4717
Conduction times calculation
By imposing the following conditions for maximum input power operation:
PIN_MAX = 300 W
Vmain_RMS = 230 V
PFMAX = 0.98
We obtain:
Equation 9
𝐼𝑙𝑎𝑚𝑝_𝑅𝑀𝑆 =
300
= 1.33𝐴
230 ∙ 0.98
The following formula correlates Ilamp_MAX with Ilamp_RMS in the phase cut dimming principle
𝑇
with a lamp current Ilamp having sinusoidal profile where 𝐷 = 𝐶𝑂𝑁𝐷
for MOSFET conduction
𝑇
time TCOND:
Equation 10
𝐼𝑙𝑎𝑚𝑝_𝑀𝐴𝑋 =
𝐼𝑙𝑎𝑚𝑝_𝑅𝑀𝑆
√𝐷 − 𝑠𝑖𝑛(4𝜋𝐷)
4𝜋
Which can be reformulated thus:
Equation 11
√𝐷 −
𝐼𝑙𝑎𝑚𝑝_𝑅𝑀𝑆
𝑠𝑖𝑛(4𝜋𝐷)
=(
)
4𝜋
𝐼𝑙𝑎𝑚𝑝_𝑀𝐴𝑋
Where:
Equation 12
𝐼𝑙𝑎𝑚𝑝_𝑅𝑀𝑆 1.33
=
= 0.686
𝐼𝑙𝑎𝑚𝑝_𝑀𝐴𝑋 1.94
With an appropriate choice for TCOND = 8.2 ms:
Equation 13
𝐷=
and by inserting this value in √𝐷 −
𝑇𝐶𝑂𝑁𝐷 8.2 𝑚𝑠
=
= 0.41
𝑇
20 𝑚𝑠
𝑠𝑖𝑛(4𝜋𝐷)
4𝜋
for Equation 11, we obtain:
Equation 14
√𝐷 −
𝐼𝑙𝑎𝑚𝑝_𝑅𝑀𝑆
𝑠𝑖𝑛(4𝜋𝐷)
−(
) ≅ 0.694 − 0.686 = 8 ∙ 10−3
4𝜋
𝐼𝑙𝑎𝑚𝑝_𝑀𝐴𝑋
Therefore, from the result obtained in the above equation, the time TCOND = 8.2 ms is a
good approximation of the right value of the MOSFET conduction time during maximum
input power conditions.
Assuming a linear envelope in the period T=1/freq for the current and voltage lamp
waveforms during normal steady state operation at minimum input power, as in:
Equation 15
𝐼𝑙𝑎𝑚𝑝 (𝑡) =
𝐼𝑙𝑎𝑚𝑝_𝑀𝐴𝑋
𝑇𝐶𝑂𝑁𝐷
∙ 𝑡 for 0<t<TCOND and T/2<t<T/2+TCOND
DocID027977 Rev 1
37/48
Conduction times calculation
Equation 16
AN4717
𝑉𝑙𝑎𝑚𝑝 (𝑡) =
𝑉𝑙𝑎𝑚𝑝_𝑀𝐴𝑋
∙𝑡
𝑇𝐶𝑂𝑁𝐷
for 0<t<TCOND and T/2<t<T/2+TCOND
In the above equation, by neglecting the voltage drops on the RDS(on) channel resistances
and on the forward biased antiparallel diodes during the MOSFET conduction phases, we
can express Vlamp_MAX for the voltage lamp signal as a function of the mains voltage thus:
Equation 17
𝑉𝑙𝑎𝑚𝑝_𝑀𝐴𝑋 = 230 ∙ √2 ∙ sin(2𝜋 ∙ 𝑓𝑟𝑒𝑞 ∙ 𝑇𝐶𝑂𝑁𝐷 )
As D = TCOND/T, the lamp power for the minimum input power condition can be expressed
as:
Equation 18
𝑃𝐼𝑁_𝑀𝐼𝑁 =
2
∙𝐼
∙𝑉
∙𝐷
3 𝑙𝑎𝑚𝑝_𝑀𝐴𝑋 𝑙𝑎𝑚𝑝_𝑀𝐴𝑋
By imposing the following minimum input power working conditions:
PIN_MIN = 40 W
Vmain_RMS = 230 V
PFMIN = 0.31
we calculate the rms value of the lamp current Ilamp_RMS (the rms value of the dimmer input
current IIN_RMS) with Equation 8:
Equation 19
𝐼𝑙𝑎𝑚𝑝_𝑅𝑀𝑆 = 𝐼𝐼𝑁_𝑅𝑀𝑆 =
𝑃𝐼𝑁_𝑀𝐼𝑁
40
=
= 0.561
𝑉𝑚𝑎𝑖𝑛_𝑅𝑀𝑆 ∙ 𝑃𝐹𝑀𝐼𝑁 230 ∙ 0.31
The following formula correlates Ilamp_MAX with Ilamp_RMS in the phase cut dimming principle,
with lamp current Ilamp having a linear variation profile where D=TCOND/T for MOSFET
conduction time TCOND:
Equation 20
𝐼𝑙𝑎𝑚𝑝_𝑀𝐴𝑋 =
𝐼𝑙𝑎𝑚𝑝_𝑅𝑀𝑆
√2 ∙ 𝐷
3
Substituting Equation 17 and 20 into 18, we obtain:
Equation 21
𝑉𝑙𝑎𝑚𝑝_𝑀𝐴𝑋 ∙ 𝐷 =
3 𝑃𝐼𝑁_𝑀𝐼𝑁
𝑃𝐼𝑁_𝑀𝐼𝑁
√3
∙
⇒ √𝐷 ∙ 𝑠𝑖𝑛(2𝜋 ∙ 𝐷) =
∙
2 𝐼𝑙𝑎𝑚𝑝_𝑀𝐴𝑋
2 𝐼𝑙𝑎𝑚𝑝_𝑅𝑀𝑆 ∙ 𝑉𝑚𝑎𝑖𝑛_𝑅𝑀𝑆
where:
Equation 22
𝑃𝐼𝑁_𝑀𝐼𝑁
40
√3
√3
∙
=
∙
= 0.268
2 𝐼𝑙𝑎𝑚𝑝_𝑅𝑀𝑆 ∙ 𝑉𝑚𝑎𝑖𝑛_𝑅𝑀𝑆
2 0.561 ∙ 230
38/48
DocID027977 Rev 1
AN4717
Conduction times calculation
With an appropriate choice for TCOND = 2.0 ms:
Equation 23
𝐷=
𝑇𝐶𝑂𝑁𝐷
2 𝑚𝑠
=
= 0.1
𝑇
20 𝑚𝑠
and by substituting this value into √𝐷 ∙ sin(2𝜋 ∙ 𝐷) of Equation 21, we obtain:
Equation 24
√𝐷 ∙ 𝑠𝑖𝑛(2𝜋 ∙ 𝐷) −
𝑃𝐼𝑁_𝑀𝐼𝑁
√3
∙
≅ 0.185 − 0.268 = −8 ∙ 10−2
2 𝐼𝑙𝑎𝑚𝑝_𝑅𝑀𝑆 ∙ 𝑉𝑚𝑎𝑖𝑛_𝑅𝑀𝑆
Therefore, from the above equation, the time TCOND = 2.0 ms is a good approximation of the
right value of the MOSFET conduction time under minimum input power conditions.
DocID027977 Rev 1
39/48
Regulation stage setting
10
AN4717
Regulation stage setting
To calculate the potentiometer resistance variation range and magnitude of the other
passive components (resistances and capacitances) of the dimmer regulation stage, we
can analyze the conduction phases of the MOSFETS at minimum and maximum levels of
power absorbed in input.
Figure 27: Regulation stage
In both phases, the supply stage electrolytic capacitance C4 (set to 47μF) discharges on
the capacitive section of the regulation stage (C5 and C6 paralleled) through the resistive
section with the potentiometer and resistances R21, R22, R23 and R24. During this
discharge transient of capacitance C4, the two capacitances C5 and C6 are charging and
increase the voltage level up to the NOR gate input threshold voltage V T, set to around
VCCmax/2.
The condition of reaching the threshold VT corresponds with the duration time of the
conduction phase set by the potentiometer for that specific lamp dimming functionality
before the beginning of the MOSFET turn-off phase.The law governing the charging
transient of the regulation stage capacitive section during the MOSFET conduction phases
at both minimum and maximum input power conditions is given by the formula for the
charge of a capacitor in an RC series circuit supplied by a constant voltage generator,
generally expressed as:
Equation 25
𝑉𝐶 (𝑡) = (𝑉𝐶 (0) − 𝑉𝐶𝐶 ) ∙ 𝑒
−𝑡⁄
𝜏
+ 𝑉𝐶𝐶
where:
40/48
VC(t) is the voltage on the two paralleled capacitances C5 and C6 of the regulation
stage
VC(0) is the voltage at time t = 0 on the two paralleled capacitances C5 and C6 of the
regulation stage
VCC is the voltage on the supply stage electrolytic capacitance C4
τ=Rreg.Creg is the time constant of the circuit, where Rreg is the equivalent resistance of
the resistive section with the potentiometer and resistances R21, R22, R23, R24 and
Creg = C5||C6
DocID027977 Rev 1
AN4717
Regulation stage setting
Figure 28: RC series circuit
Due to the high value of the electrolytic capacitance C4 (set to 47μF) and the low current
levels flowing on the regulation stage during the C4 discharge phase, we can assume the
supply stage voltage VCC to be constant during the charging phase of the capacitive section
Creg = C5||C6.
So, by inserting Vc(0) = 0, Equation 25 is reduced to:
Equation 26
𝑉𝐶 (𝑡) = 𝑉𝐶𝐶 ∙ (1 − 𝑒
−𝑡⁄
𝜏)
We can reformulate in terms of the τ parameter thus:
Equation 27
𝜏=
𝑡
𝑉𝐶𝐶
𝑙𝑛 (
)
𝑉𝐶𝐶 − 𝑉𝐶 (𝑡)
Moreover, as the voltage level on C5 and C6 always rises during the MOSFET conduction
phase up to the NOR gate input threshold voltage VT (set to around VCCmax/2), we can
assume:
VC (TON ) = VCC/2
for the MOSFET conduction time t = TON so Equation 27 becomes:
Equation 28
𝜏 = 𝑅𝑟𝑒𝑔 ∙ 𝐶𝑟𝑒𝑔 =
𝑇𝑂𝑁
𝑙𝑛(2)
For the minimum input power condition, TON= 2 ms and so:
Equation 29
𝑅𝑟𝑒𝑔_𝑚𝑖𝑛 ∙ 𝐶𝑟𝑒𝑔 =
2 ∙ 10−3
= 2.88𝑚𝑠𝑒𝑐
𝑙𝑛(2)
By selecting Rreg_min = 120 kΩ for the equivalent resistance of the regulation stage resistive
section at minimum input power:
Equation 30
𝐶𝑟𝑒𝑔 =
2.88 ∙ 10−3
≅ 24 𝑛𝐹
120 ∙ 103
And therefore capacitance Creg = C5||C6 = 22 nF is chosen for the design at the minimum
input power condition.
DocID027977 Rev 1
41/48
Regulation stage setting
AN4717
At the maximum input power condition, due to the long time constant for the charging
transient, we consider a linear variation for the voltage on the two paralleled capacitances
C5 and C6 of the regulation stage; therefore, Equation 26 can be transformed into the
following linearized form:
Equation 31
𝑉𝐶 (𝑡) =
𝐼𝑅𝑒𝑔−𝑚𝑎𝑥
∙𝑡
𝐶
where for t = TON:
VC(TON) = VCC/2
Ireg_max can be chosen as the current at the t = 0 to equal Ireg_max = VCC/R.
By substituting these two conditions in Equation 31, for TON= 8.2 ms we obtain:
Equation 32
𝑅𝑟𝑒𝑔_𝑚𝑎𝑥 ∙ 𝐶𝑟𝑒𝑔 = 2 ∙ 𝑇𝑂𝑁 ⇒ 𝑅𝑟𝑒𝑔_𝑚𝑎𝑥 =
2 ∙ 𝑇𝑂𝑁 2 ∙ 8.2 ∙ 10−3
=
= 745 𝑘𝛺
𝐶𝑟𝑒𝑔
22 ∙ 10−9
The variability range of the equivalent resistance Rreg is therefore:
Equation 33
120 𝑘𝛺 ≤ 𝑅𝑟𝑒𝑔 ≤ 745 𝑘𝛺
The following values were therefore chosen for the potentiometer and resistances R21,
R22, R23, R24 of the regulation stage resistive section:
potentiometer = 1 MΩ
R21 = 1 MΩ, R22 = 560 kΩ, R23 = 150 kΩ, R24 = 120 kΩ
Using these resistances, the actual variation range for the equivalent resistance Rreg
becomes:
Equation 34
120 𝑘𝛺 ≤ 𝑅𝑟𝑒𝑔 ≤ 751 𝑘𝛺
42/48
DocID027977 Rev 1
AN4717
11
Experimental test and results @ 110Vac
Experimental test and results @ 110Vac
The following figures show the waveforms acquired under steady state operation with 90 W
halogen technology incandescent bulbs at 110 VAC input mains for the above mentioned
operating conditions for the MOSFETs.
Waveforms were acquired by scope with the following signal color codes:
Q1 drain current: signal ID1= yellow
Q2 drain current: signal ID2 = blue
Q1 drain-source voltage: signal VDS1 = red
Q2 drain-source voltage: signal VDS2 = green
Figure 29: Steady state operation @ 110 VAC and minimum input power
Figure 30: Steady state operation @ 110 VAC and medium input power
DocID027977 Rev 1
43/48
Experimental test and results @ 110Vac
AN4717
Figure 31: Steady state operation @ 110 VAC and maximum input power
The figure below shows the different temperatures and turn-off switching energies
measured for the STF17N62K3 MOSFETs working under minimum, medium and maximum
operating conditions for lamp current dimming and input powers absorbed by the board
connected to 90 W/110 VAC halogen technology incandescent bulbs.
Figure 32: Temperatures and turn-off energies @ 110 VAC
The maximum case temperatures (~48 °C without heatsinks) are reached in free air
conditions (25 °C room temperature) for the medium input power level operation with the
original setting of 22 kΩ.
44/48
DocID027977 Rev 1
AN4717
Experimental test and results @ 110Vac
The following figure details the turn-off transient for MOSFET Q2 switching during medium
input power operation.
Figure 33: Turn-off energy@ medium input power
DocID027977 Rev 1
45/48
Conclusion
12
AN4717
Conclusion
The STEVAL-ILD005V1 evaluation board implements a trailing edge operation mode for 2wire analog wall dimmers with flicker-free dimming and no audible noise performance for
dimmable R-C lighting loads in the power range from 40 W to 300 W at 230 VAC input
mains, by using a triple 3-input NOR gate to perform gate driving, zero crossing and timing
control functions and a switched type single linear rotary potentiometer as the control
interface. Thanks to the electrical characteristics of the STF17N62K3 MOSFET with high
Ciss values and low RDS(on) values at the operating currents, this MOS-based dimmer
solution is able to meet the European EMC standard for conducted emission noise
(EN55015), without requiring any bulky LC input filters, relying instead on appropriate
setting for the resistances of the gate driving network to the active devices. At the same
time, the MOSFETs show satisfactory thermo-electrical performance without any heatsink
components in contact with the case and the 620 V breakdown voltage ensures good
margins for passing surge immunity tests with up to 2 kV voltage pulses, in compliance with
the IEC61000-4-5 standards.
46/48
DocID027977 Rev 1
AN4717
13
Revision history
Revision history
Table 4: Document revision history
Date
Version
12-Aug-2015
1
DocID027977 Rev 1
Changes
Initial release.
47/48
AN4717
IMPORTANT NOTICE – PLEASE READ CAREFULLY
STMicroelectronics NV and its subsidiaries (“ST”) reserve the right to make changes, corrections, enhancements, modifications,
and improvements to ST products and/or to this document at any time without notice. Purchasers should obtain the latest relevant
information on ST products before placing orders. ST products are sold pursuant to ST’s terms and conditions of sale in place at
the time of order acknowledgement.
Purchasers are solely responsible for the choice, selection, and use of ST products and ST assumes no liability for application
assistance or the design of Purchasers’ products.
No license, express or implied, to any intellectual property right is granted by ST herein.
Resale of ST products with provisions different from the information set forth herein shall void any warranty granted by ST for such
product.
ST and the ST logo are trademarks of ST. All other product or service names are the property of their respective owners.
Information in this document supersedes and replaces information previously supplied in any prior versions of this document.
© 2015 STMicroelectronics – All rights reserved
48/48
DocID027977 Rev 1