The present invention relates to the general subject of electronic
power supply converters for energy wherein one example is a boost converter system
having a power factor correction (PFC) that is running in discontinuous mode using
an equation to calculate the point of time the zero-current state of the storage
inductor occurs without the requirement of a secondary winding or special voltage
comparators to detect zero current to achieve maximum power transfer and to avoid
a continuous mode operation.
The design of a boost converter system with a power factor converter
(PFC) requires small storage inductors together with high transferred power. These
two parameters are counteracting against each other. The transferred power reaches
its maximum if the storage inductor is recharged right after the inductor reaches
the zero current state.
Prior art power supplies are using either secondary windings, special
voltage comparators or analog current sensing circuits to detect the zero-current
state of the storage inductor.
US Patent 5,757,166 to Sodhi teaches a power correction factor boost
converter. A secondary winding is used for zero current detection of the storage
US Patent 5,861,734 to Fasullo et al describes a control system for
a boost converter using 2 interleaved boost circuits. A current sensing circuit
is provided that senses the current in each of the boost converters. 2 boost converter
switches have to be controlled.
US Patent 6,178,104 B1 to Nak-Choon-Choi describes a power factor
correction circuit using reverse saw tooth waves. The switch is coupled to a resistor
and a capacity that by forming a current detector, detect the current flowing through
the storage inductor of the boost converter. The PFC circuit is using reverse saw
tooth waves and is controlling the slope of the current.
Fine adjustment of the energy transfer overcoming the limitations
of discrete time in digital systems is a known problem. U.S. Patent 6,043,633 to
Lev et al. discloses a method and an apparatus for controlling a boost converter
that offers power factor correction by compensating for the parasitic capacitance
and parasitic oscillations. A zero current detector facilitates the compensation.
A dithering method to enhance the time resolution of clocked digital circuits is
Summary of the invention
A principal object of the present invention is to provide a highly
effective electronic power supply converter such as a boost converter having maximum
power, related to the size the of the storage inductor, transferred without the
requirement of a secondary winding or voltage comparators.
A further object of the present invention is to provide a highly effective
boost converter with a power factor corrector (PFC) having maximum power, related
to the size the of the storage inductor, transferred without the requirement of
a secondary winding or voltage comparators.
A further object of the present invention is to recharge the storage
inductor right after the point of time when zero current state occurs at the storage
inductor. This is key to a maximal transfer of energy.
A still further object of the present invention is to achieve a fine
adjustment of the energy transfer to minimise distortion and harmonics overcoming
the limitations of discrete time steps in clocked digital systems.
Another still further object of the present invention is to achieve
an optimal accuracy of the measurement of the voltages at the source and the load
side to achieve best accuracy to define the point of time of the zero current state
of the storage inductor and furthermore to achieve best accuracy required as input
for the fine-tuning of the energy transferred.
Another still further object of the present invention is to achieve
less manufacturing costs and to reduce the number of components required by avoiding
secondary windings or voltage comparators for the zero current detection.
In accordance with the objects of this invention a system used for
converting electronic power supply energy has been achieved. This system can be
used as a boost converter or a DC to DC converter. Maximal power is transferred
by recharging the storage inductor right after the point of time when zero current
occurs at the storage inductor, As an example of the usage of the invention a
boost converter with PFC (power-factor-corrector) having maximal power transferred
related to the size of the storage inductor by recharging the storage inductor
right after the point of time when zero current state occurs at the storage inductor
is achieved. Fig. 1 illustrates the main components of the system. The boost converter
is including a storage inductor coupled to an input voltage, a shunt switch controlling
a current flowing through said storage inductor and a rectifying diode for rectifying
the output voltage. Furthermore the system comprises of an analogue/digital converter
which is converting analogue values measured and reference voltages into digital
values required by a digital control unit to control frequency and pulse width
of the shunt switch using means to calculate the point of time when zero current
state occurs instead of detecting this point of time using secondary windings or
other analog circuits. Said digital control unit initiates the recharging of the
storage inductor right after the point of time of the zero current state is reached.
In accordance with an object of the invention a method of calculating
the point of time of the zero current state of the storage inductor is achieved.
Said point of time is calculated using an equation based on the ON time of said
shunt switch and the voltages measured at the source side (rectified mains supply)
and the load side. A safety margin to balance inaccuracies of the measurement is
added to this calculated point of time.
In accordance with another object of the invention to minimize distortion
and harmonics a fine adjustment of the energy transfer through fine tuning of the
pulse width of the shunt switch is introduced overcoming the limitations of discrete
time steps in clocked digital systems. This is achieved by either using patterns
or by a digital delta sigma modulator that is averaging the ON time values of the
shunt switch by toggling between neighboring ON time values (pulse-width) and
controlled by said digital control unit.
In accordance to the object of this invention the calculation of the
point of time of the zero current state of the storage inductor and the fine-tuning
of the energy transferred require a very high accuracy of the measurement of the
voltages at the load side, the inductor side and at the source side. This is achieved
by a calibration of the tolerances of the voltage dividers used for these measurements.
The voltage divider ratios can be measured at appropriate periods of time (see
Fig. 7) considering the small influence of the voltage of the forward bias of the
diode in the magnitude of 0.7volt hence. This calibration enables the usage of
tolerant voltage dividers. Said digital control unit is controlling the calibration
of said voltage dividers.
In accordance with the objects of the invention said three methods
of (1) calculating the point of time of the zero state of the storage inductor
and (2) of fine-tuning the energy transfer and (3) of calibrating the voltage dividers
to improve the accuracy of the voltages measured can all be used separately or
used in any combinations together.
Description of the drawings
In the accompanying drawings forming a material part of this description,
there is shown:
Description of the preferred embodiments
- Fig.1 illustrates the preferred principal embodiment of the PFC (Power-Factor-Corrector)
boost converter consisting of one storage inductor, a rectifier diode, a shunt
switch, voltage dividers and a control unit containing an analog-to-digital converter,
a delta-sigma modulator and a logical unit.
- Fig.2A illustrates the principal currents of the power supply.
- Fig. 2B illustrates the waveforms of the currents of Fig. 2A as
functions of the switch S.
- Fig. 3 illustrates the flow of the current IL through
the storage inductor dependent of the ON time of the switch and the transfer time
TTR of the energy.
- Fig. 4 illustrates the principal wave form of the current through the
storage inductor IL related to the ON time of the switch and
the period and the amount of energy transfer Euc related to the ON time.
- Fig. 5 illustrates one method how the point of time of zero current is
- Fig. 6 illustrates one method how to do the fine adjustment of the energy
- Fig. 7 illustrates one method how to improve the accuracy of voltage
measurement by calibrating the voltage dividers.
The preferred embodiments disclose a novel system used as converters
for energy in electronic power supply systems. Said system, running in discontinuous
mode, can be used in simple boost converters or in boost converters having a PFC
(power-factor-correction) or can be used as an DC-to-DC converter that is running
in a discontinuous mode. Some parts of the invention are applicable to other power
electronic systems as well. The design of such converters requires small inductors
together with maximal power transferred. These two parameters are counteracting
against each other. The power transferred reaches its maximum when the storage
inductor is recharged immediately after reaching the zero current state. For the
detection of the zero current generally a secondary winding at the storage inductor
is used. As an example of the invention a configuration of the principal components
of a boost converter with PFC is shown in Fig. 1. The main components are
the storage inductor L, the rectified main supply U1, the
shunt switch S, the rectifier diode D, the voltage dividers pairs
Z11 - Z12, Z21 -Z22
Z41 - Z42 are used for the measurement of the voltages
of the mains supply U1, the voltage of the inductor output
US, the voltage at the load side UC, the reference
voltage Uref and a oscillator driven digital control unit to control
the frequency and pulse width of the shunt switch S.
Fig. 2A shows the principal layout of a boost converter and
Fig. 2B shows the flow of the currents flowing through the storage inductor
L, the shunt switch S and the rectifying diode D. Said boost
converter is operating in a discontinuous mode. The current Is starts to flow after
the switch S is closed (or ON), the current IL is rising
as long the switch S is ON (short pulse width) and is decreasing during the
time period TTR. The time period TD describes
the time period when no current is flowing. In a discontinuous mode this time period
must be equal or greater than zero. The current ID at the load
side starts to flow when the switch S is opened (or in OFF state) and is
decreasing during the period of energy transfer TTR.
This invention proposes to calculate the point of time when the current
through the storage inductor IL reaches the zero state. This
happens always after the switch S is opened with the delay TTR. TTR
is the period of energy transfer. TTR
can be calculated using the
ON time Ton of the switch and the voltages U1
and Uc. The voltage across the storage inductor is
UL = L*dI / (dt)
The maximum current through the storage inductor (see Fig.
3 and Fig.2 A+B) is
ILMAX = U1 - US / (L)*TON
During the time period the switch is closed the voltage
Us equals zero. This means
|LMAX = U1 / (L)*TON
The transfer time TTR of the current ID is
TTR= LLMAX*L / (US-U1)
= U1*TON / (US-U1)
The forward voltage of the diode D is in the range of 0.7 Volts.
In the forward mode
The final equation to calculate the transfer time is
TTR = U1*Ton / (Uc + 0.7V
Using this equation the point of time T0 of the
zero current state of the current IL can be calculated.
A method how to calculate said T0 is illustrated
in Fig. 5. When the total system is inactive
U1 = US = UC + 0.7 volts.
In step 51 the system starts with a low value of
Ton to avoid saturation of L. In step 52 the input
voltage U1 is measured. In step 53 the shunt switch is
closed for time Ton. In step 54 the switch S is
opened after the pulse width Ton. In the next step
55 the voltages U1 at the source side and UC
at the load side are measured immediately after the switch is opened. Usually the
switching frequency of the converter is much higher than the mains frequency or
the variation of the voltage UC. If switching intervals are short
related to the frequency of the mains supply, which is the case under normal conditions
of operations, these measurements can be performed any time but preferably in step
52 already when the input voltage U1 is measured and step
55 can be skipped. The noise level of U1 and
Uc is much lower than at Us. In step
56 the digital control is calculating the transfer time period
TTR according to the equation above. In step 57 TTR
is added to the falling edge of the Ton pulse to get the point
of time T0 of the zero current state of the current
IL. In step 58 a safety margin of e.g. 1 clock cycle or
100ns is added before the next cycle starts with step 52 again. This safety
margin is required to avoid the risk that the converter goes into a continuous
mode with great power dissipation. Above-mentioned method is used to define the
frequency of the boost converter system.
The amount of energy transferred is mainly a function of the pulse
Fig. 4 illustrates how the pulse width
Ton correlates to the energy transferred EUC.
Power converters, that make low distortion and harmonics, need a fine adjustment
of the ON time of the switch. But in digital systems the ON-time is discrete. In
order to overcome the discreteness of the time steps an averaging of neighbouring
discrete ON time values (pulse width) is introduced.
The method used for the fine adjustment of energy is illustrated in
Fig. 6. In step 61 the voltage at the load side Uc
is monitored through the voltage divider Z41 and Z42
and by the A/D converter of Fig.1. In step 62 the digital control
unit compares Uc with the reference voltage Uref
(see Fig.1) digitised as well by said A/D converter. In step
63+64 the digital control unit is adjusting the pulse-width Ton
according to the comparison result of Uc and Uref
and is increasing or reducing the pulse-width Ton accordingly
to reach equivalence between Uc and Uref in
order to adjust the energy transferred. In the step 63 the gross adjustment
is done. In case of small differences between Uc and
Uref step 64 illustrates the averaging of neighboring
Ton values to fine-tune the energy transferred. The averaging
is done by either using defined patterns to vary the on time of the switch ("toggling")
in minimal steps possible or by using digital delta-sigma modulation with a multilevel
quantifier. One digital quantifier level corresponds to the period to load the
storage inductor. The delta-sigma principle as an averaging principle makes the
use of digital systems with moderate clock frequency possible. This fine adjustment
of the ON time using a digital delta sigma modulator is very useful for a SEPIC
(single ended primary inductor converter) when the load variation is large and
the ON times of the switch are short. For the time period a SEOIC is inactive,
Uc = 0. Increasing switching frequency and increasing
Ton increases Uc in all systems used.
For the said calculation of the point of time of the zero current
state in the storage inductor and for the control of the pulse-width
Ton by said comparison of Uc and
Uref an accurate measurement of the voltage at the source side
U1, the voltage on the storage inductor Us
and the voltage at the load side Uc is necessary (see Fig.1).
This invention proposes to increase the accuracy of the measurement with a calibration
of the voltage dividers Z11Z12, Z21Z22
and Z31Z32 (see Fig. 1) to overcome the tolerances
of the voltage dividers. This enables the usage of tolerant voltage dividers without
losing accuracy of the measurement.
Fig. 7 illustrates the method how and during which time periods
the calibration can be performed. Step 71+72 explain the period of zero
current state of the storage inductor is used to measure the ratios of the voltage
dividers Z11Z11 + Z12
and Z21Z21 + Z22.
Step 73+74 explain that during the energy transfer period TTR
the diode D is forward biased. The voltage across the diode D is
well known, it is about 0.7 Volts and much lower than the voltage to be measured.
During the forward bias of the diode D the voltage at the storage inductor
Us equals the voltage at the load side UC plus
0.7Volts. Hence the voltage divider ratios Z32Z32 + Z31
and Z22Z21 + Z22 can be measured. The digital control
unit uses the measured values of the voltage divider ratios to calculate more precisely
the point of time of the zero current state and to control more precisely the pulse
The methods of (1) calculating the point of time of the zero
current state of the storage inductor (Fig. 5) and of (2) of fine-adjusting
of the energy transferred (Fig. 6) and of (3) improving the accuracy
of voltage measurements by calibrating the voltage dividers (Fig. 7) can
be used either individually or together in any combination. The highest efficiency
of this kind of power supply combined with minimal distortion and harmonics will
be achieved by a combination all three said methods together. For more simple requirements
a solution could be achieved with just one or two of said methods.
While the invention has been particularly shown and described with
reference to the preferred embodiments thereof, it will be understood by those
skilled in the art that various changes in form and details may be made without
departing from the spirit and scope of the invention.