 |
Description  |
|
|
BACKGROUND OF THE INVENTION
The development of the transformer isolated forward converter has resulted
in improvements in switched power supplies. This is because the forward
converter costs less than the other types of transformer isolated
switching power supplies of the prior art, and it is easier to design.
However, forward converters of the prior art are subject to certain
disadvantages in some instances in that considerable power must be
dissipated in each switching cycle, and also in that often only half of
the B/H hysteresis curve of the core is used, resulting in low core
efficiency and the requirement of larger cores as compared with other
types of converter circuits.
SUMMARY OF THE INVENTION
A transformer isolated switched power supply of the forward converter type
which includes a clamping circuit for switching unit of the power supply
which provides means for recovering the energy stored in the transformer
during each switching cycle without any need to dissipate the energy, and
which also serves to reverse the flux in the power transformer during each
switching cycle to restore the flux capacity of the power transformer to
the same level as in other types of converter circuits.
BRIEF DESCRIPTION OF THE DRAWINGS
FIGS. 1 and 2 are circuit diagrams showing portions of two types of the
prior art transformer isolated switched forward converter power supplies;
FIGS. 3A and 3B are curves useful in explaining the operation of the
switched forward converter power supply;
FIG. 4 is a circuit diagram representing one embodiment of the invention as
applied to a particular type of switched forward converter power supplies;
and
FIG. 5 is a more detailed schematic diagram of the circuit of FIG. 4.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT
The present invention is concerned with the single-ended pulse width
modulated, or "forward", converter. A typical prior art forward converter
is shown in FIG. 1. In the circuit of FIG. 1, T.sub.1 is an isolation
transformer with a center tapped primary winding. A direct current input
voltage V.sub.DC is applied between the center tap and one side of the
primary winding through an electronically operated switching circuit
designated SW1. The other side of the primary winding is connected back to
the negative side of the direct current input voltage V.sub.DC through a
diode CR1.
One side of the secondary winding of transformer T.sub.1 is connected
through a diode CR2 and a filter inductor L1 to one of the output
terminals of the circuit. The other side of the secondary winding is
directly connected to the other output terminal. A diode CR3 is connected
between the junction of diode CR2 and inductor L1 and the other side of
the secondary. A capacitor C1 is connected across the output terminals.
The circuit of FIG. 1 operates in the following known manner. When switch
SW1 is turned on, the direct current input voltage V.sub.DC is applied
across one-half of the primary winding of transformer T.sub.1. This causes
a secondary voltage to be induced across the secondary winding which is
phased to produce a positive voltage on the anode of diode CR2. This
positive voltage causes CR2 to conduct resulting in the secondary voltage
being applied to the filter inductor L1. After a period of time, switch
SW1 is turned off. The inductance of transformer T.sub.1 now causes the
voltage across switch SW1 to increase and the voltage across diode CR1 to
decrease, and this action continues until the voltage across diode CR1
starts to go negative. Diode CR1 then conducts limiting any further change
in the voltage on transformer T.sub.1, and this action effectively limits
the voltage across switch SW1 to twice the direct current input voltage.
At the same time, the secondary voltage also reverses which reverses the
bias on diode CR2 rendering it non-conductive. Diode CR3 then conducts to
provide a current path for the energy stored in the inductor L1. During
this period, the voltage applied to inductor L1 is effectively zero. The
circuit remains in this state for a time approximately equal to the "on"
time of switch SW1, at which time the energy stored in transformer T.sub.1
is discharged back to the direct current source. The voltage across the
primary winding of transformer T.sub.1 then falls to zero until switch SW1
is again turned on, causing the cycle to repeat.
The cyclic operation described above causes a train of pulses to be applied
to inductor L1, with the duty cycle of the pulses being equal to the duty
cycle of the "on" time of switch SW1. Since inductor L1 and capacitor C1
form an averaging filter, the output voltage which appears across
capacitor C1 is the average value of the pulse train. This average voltage
is determined by the duty ratio of the "on" to "off" period of switch SW1,
multiplied by the direct current input voltage, multiplied by the
transformer turns ratio. The following equation illustrates the average
output voltage as a function of these other parameters:
V.sub.av =V.sub.dc .times.N.times..eta. (1)
Where:
V.sub.av =average output voltage
V.sub.dc =DC input voltage
N=transformer secondary/primary turn ratio
.eta.=duty ratio=(T.sub.on)/T
T.sub.on ="on" time of switch SW1
T=the period of a complete switching cycle (T.sub.on +T.sub.off)
It should be noted that in order for the energy stored in transformer
T.sub.1 to be completely discharged back to the direct current source, the
"off" time of switch SW1 must be equal to or longer than the "on" time.
Therefore, the maximum value of .eta. is 0.5.
One of the factors influencing the cost and size of a switching converter
is the rating of the main power switching unit SW1. There are two factors
to this rating. One of the factors (V) is the voltage that the switch must
withstand in the "off" condition. The other factor (A) is the current the
switching unit must carry in the "on" condition. These two factors
multiplied together establish a volt-ampere (VA) VA rating for the power
switching unit which is a rough measure of the cost of this unit. In many
instances, the power switching unit consists of several components in
series and parallel. In such a case the VA product of the switching unit
is equal to the sum of the VA products of the individual components. This
means that the number of components which make up the switching unit, and
hence its cost is directly proportional to the overall VA product
requirement.
The current through switching unit SW1 in the "on" state is the average
output current multiplied by the transformer turns ratio N. Turns ratio N
is set by the desired output voltage and the minimum direct current input
voltage.
Manipulation of the Equation (1) gives:
##EQU1##
The maximum value of N and, therefore the maximum input current occurs at
the minimum value of V.sub.dc. Also, as mentioned above, the maximum
allowable duty ratio for the circuit is 0.5. Therefore:
##EQU2##
The current rating of the switch SW1 is then given by:
##EQU3##
Where: P.sub.av is the rated load power of the converter.
During the "off" state of switch SW1, the diode CR1 is conducting, and this
results in a voltage across the full primary of transformer T.sub.1 which
is twice V.sub.dc. The voltage rating of SW1 must therefore be at least
twice the maximum direct current input voltage. This gives a
voltage-ampere product rating for the switch SW1 as follows:
##EQU4##
The volt-ampere product given by equation (5) is consistent with the other
pulse width modulated converter circuits such as the bridge, half-bridge,
and push-pull converters. If the sum of the volt-ampere products of all
the switches of any one of the above configurations is taken, then the sum
will be identical to the value given in equation (5).
From the above, it would appear that equation (5) represents a fundamental
lower limit for the total volt-ampere (VA) product rating of the power
switches in a pulse width modulated converter. However, there is a
slightly modified version of the circuit of FIG. 1 which permits a lower
VA product. This circuit is shown in FIG. 2.
In the circuit of FIG. 2, the V.sub.dc input voltage is applied across the
entire primary of transformer T.sub.1 through the switching unit SW1, and
diode CR1 is connected between the junction of the switch and a direct
current reference voltage source V.sub.cl.
In the circuit of FIG. 2, the equation for output voltage is still equation
(1), and the switch current is still the product of the average output
current and the transformer turns ratio. The voltage V.sub.cl is chosen to
be equal to the sum of the minimum and maximum values of V.sub.dc in.
V.sub.cl =V.sub.dc (min) +V.sub.dc (max) (6)
FIGS. 3A and 3B illustrate the voltage waveform across switch SW1 of FIG. 2
for V.sub.in(min) and V.sub.in(max) respectively.
In FIG. 3A the volt/second area A.sub.1 applied across the primary winding
of transformer T.sub.1 in FIG. 2 must equal the reset volt/second area
A.sub.2. The same requirement exists for FIG. 3B, in which the area
A.sub.3 must equal the area A.sub.4. In FIGS. 3A and 3B, T is the time for
a full period of the waveform; and T must be equal to or greater than the
sum of T.sub.1 and T.sub.2 for FIG. 3A, and equal to or greater than the
sum of T.sub.3 and T.sub.4 for FIG. 3B.
The area A.sub.1 in FIG. 3A is given by the product V.sub.in(min)
.times.T.sub.1 ; and the area A.sub.2 equals the product (V.sub.cl
-V.sub.in (min)).times.T.sub.2. Since the voltage V.sub.cl was chosen to
equal the sum V.sub.in(min) +V.sub.in(max), the difference V.sub.L
-V.sub.in(min) =V.sub.in(max). Therefore, the area A.sub.2 equals the
product V.sub.in(max) .times.T.sub.2. Since A.sub.1 must equal A.sub.2,
the relationship of T.sub.1 and T.sub.2 is given by:
T.sub.1 V.sub.in (min) =T.sub.2 V.sub.in (max) (7)
From Equation (7), the sum of T.sub.1 and T.sub.2 becomes:
##EQU5##
The duty cycle of the circuit of FIG. 2 at minimum input voltage is given
by:
##EQU6##
Where: .eta.=duty cycle.
Since the minimum value that T may assume is T.sub.1 +T.sub.2 in this case,
the maximum duty ratio is given by:
##EQU7##
Substituting from equation (8):
##EQU8##
From equation 11 it can be shown that if V.sub.in(min) is less than
V.sub.in(max) then .eta..sub.(max) may be greater than 0.5 which is the
duty ratio limit of the circuit of FIG. 1.
Referring to equation (1) the secondary/primary turns ratio (N) of
transformer T.sub.1 may be reduced, which results in a reduced current in
the switch SW1 for the circuit shown in FIG. 2. Equation (4) is modified
as shown below:
##EQU9##
Referring to FIG. 3B, the area A.sub.3 represents the volt/second area
applied to the primary of transformer T.sub.1 during the "on" time of
switch SW1 in FIG. 2. For constant frequency operation, the value of T in
FIG. 3A must equal the value of T in FIG. 3B. Furthermore, because the
average voltage delivered by the converter is proportional to the "on"
time volt/second area divided by T, and because one of the objectives of
the converter is to deliver a constant average output regardless of the
input voltage, the value of A.sub.1 of FIG. 3A must equal the value of
A.sub.3 in FIG. 3B. Also, for the transformer T.sub.1 of FIG. 2 to reset
fully, A.sub.4 must equal A.sub.3. It follows that all the volt/second
areas A.sub.1, A.sub.2, A.sub.3 and A.sub.4 must be equal.
It was shown that T.sub.2 is equal to A.sub.2 divided by V.sub.in(max).
From FIG. 3B it can also be seen that T.sub.3 equals A.sub.3 divided by
V.sub.in(max). Since all of the areas A.sub.1, A.sub.2, A.sub.3 and
A.sub.4 are equal, it follows that T.sub.3 is equal to T.sub.2. Likewise,
it follows that T.sub.4 is equal to T.sub.1. This means that T.sub.3
+T.sub.4 =T.sub.1 +T.sub.2 which equals T. For input voltages between
V.sub.in(min) and V.sub.in(max) the sum of the "on" time for switch SW1,
and the reset time for transformer T.sub.1, for constant average output
voltage is less than the time T, which satisfies the requirement for
providing sufficient time to reset the transformer during the operating
cycle.
The maximum voltage that the switch SW1 in FIG. 2 must withstand is
V.sub.cl which has been defined as the sum of V.sub.in(min)
+V.sub.in(max). Using equation (12) and this value for maximum voltage on
SW1 for FIG. 2, the volt/ampere product rating for switch SW1 in FIG. 2
becomes:
##EQU10##
Equation 13 may be simplified to the following form:
##EQU11##
The form of equation (14) was chosen so that it may be readily compared
with equation (5) thereby illustrating the advantage of the circuit shown
in FIG. 2 over that of FIG. 1. This comparison is made by comparing the
quantity
##EQU12##
from equation (14) with the constant 4 from Equation (5).
In the extreme case where the input voltage remains constant, V.sub.in(min)
equals V.sub.in(max) ; then K=1 and the quantity
##EQU13##
which shows no advantage. However, in the practical case, where the
converter must regulate over a sizable variation of input voltage, a
typical value for K is about 2.5. This yields a value for
##EQU14##
of 1.96 which is substantially less than 4.
The above discussion shows that the circuit of FIG. 2 has a clear advantage
over the circuit of FIG. 1 in terms of power switch ratings when the
converter must regulate over wide input voltage variations. Furthermore,
it can be shown that the power switch rating for the circuit in FIG. 1, as
described by Equation (5), is equal to the total power switch ratings of
each of the popular push-pull, half bridge, full bridge, circuit
configurations, when such circuit configurations are used in a pulse-width
modulated regulator type switching converter, in which the output is
derived by averaging the rectified output to produce a DC voltage equal to
this average.
Therefore, the circuit of FIG. 2 not only has an advantage over the circuit
of FIG. 1, but it also has the same advantage over the other circuit
configurations listed above.
The foregoing advantages have been recognized by the industry for some
time. However, the implementation of the circuit of FIG. 2 has a serious
problem. Specifically, the from reference voltage source V.sub.cl must
absorb all of the energy stored in the inductance of transformer T.sub.1
when diode CR1 is conductive. This energy can represent considerable power
that must be dissipated in order to maintain the voltage from reference
voltage source V.sub.cl. In the circuit of FIG. 1 this energy does not
pose a problem because it is returned to the power source.
Also, the transformer of a usual prior art forward converter is larger than
the transformer of the other converter types because the reset schemes of
either FIG. 1 or FIG. 2 do not permit the current to fall below zero due
to the action of the clamp diode CR1. The result is that the flux in the
transformer core may only fall to zero, and it does not reverse as in the
other types of converters. This results in an effective flux capacity that
is half what it would be in the more symmetrical converters.
U.S. Pat. No. 4,672,517, which is assigned to the present Assignee,
describes one particular implementation by which a transformer isolated
switched converter may be constructed to utilize both sides of the
operating area of the B/H curve of the transformer core. The system of the
present invention represents another implementation.
The system of the present invention provides a means for recovering the
energy delivered to the clamp voltage V.sub.cl in FIG. 2, and also of
reversing the flux in the power transformer which restores the flux
capacity of the power transformer to the same level as in the other types
of converter circuits. One example of the system of the invention is shown
in FIG. 4.
The circuit of FIG. 4 is similar to the circuit of FIG. 2, except for the
addition of a capacitor C2, a switch SW2, and a comparator U1. Switch SW2
is an active switch similar to switch SW1, but of a much lower power
rating. Comparator U1 and switch SW2 are configured so that, when the
voltage across capacitor C2 exceeds the reference voltage from reference
voltage source V.sub.cl, switch SW2 is actuated to its "on" state.
Conversely, when the voltage across capacitor C2 is less than the
reference voltage from reference voltage source V.sub.cl, switch SW2 is
actuated to its "off" state.
When the switch SW1 turns "off", the voltage across the primary of
transformer T.sub.1 rises until diode CR1 conducts. The energy stored in
the inductance of transformer T.sub.1 is then delivered to C2, thereby
charging C2 above the voltage from reference voltage source V.sub.cl.
Because the voltage of C2 is greater than the voltage from reference
voltage source V.sub.cl, switch SW2 is turned on. When switch SW2 is
turned on, the voltage on the primary of T.sub.1 remains connected to C2
after the inductive current reverses. Capacitor C2 then discharges into
transformer T.sub.1 until the voltage on C2 is equal to the voltage from
reference voltage source V.sub.cl. Switch SW2 then switches to its "off"
state, releasing the primary of T.sub.1 from capacitor C2.
Because capacitor C2 is discharged back to its starting voltage, all of the
energy except for losses in diode CR1 and in switch SW2 is returned to the
inductance of the transformer T.sub.1. Since the inductive current has
reversed during this process, the transformer core has also had its flux
reversed. Therefore, the limitation of only being able to use half of the
flux capacity of the core in a forward converter has been overcome,
resulting in the feasibility of a smaller transformer for a given power
level, as compared with most prior art forward converters.
The value of capacitor C2 is chosen to be large enough so that, when
transformer T.sub.1 delivers its energy to the capacitor C2, the voltage
changes only a few percent, therefore the voltage across capacitor C2 may
be considered essentially constant and equal to the voltage from reference
voltage source V.sub.cl. The voltage from reference voltage source
V.sub.cl is simply used as a reference and the reference voltage source
does not have to dissipate any power.
A schematic diagram showing the present embodiment of the invention is
illustrated in FIG. 5. The polarities have been inverted in FIG. 5 with
respect to the circuits of FIGS. 1, 2 and 4. However, the operating
principle of the circuit in FIG. 5 is the same as the circuit of FIG. 4.
The various elements in FIG. 5 have the following values:
R108--180 kilo-ohms
R109--180 kilo-ohms
C110--0.001 microfarads
C111--0.001 microfarads
R120--1 kilo-ohm
R121--10 ohms
C119--0.01 microfarads
R119--6.2 kilo ohms
C118--0.1 microfarads
Zener diode CR114--1N961B
Comparator U101--LM311
R122--1.5 kilo-ohms
R123--1.5 kilo-ohms
Q107--2N2222A
Q108--2N2907A
C120--0.0015 microfarads
R124--10 ohms
FET109--BUZ311
CR116--IN4937
CR115--MUR8100
C112--0.15 microfarads
R110--330 ohms
CR107--1N4937
C113--0.0027 microfarads
CR109--1N4937
C114--1.0 microfarads
R112--100 kilo-ohms
Zener Diode CR111--IN5352
FET Q103, FET Q104, FET Q105, FET Q106--IRFP360
R114--10 ohms
R115--10 ohms
R116--10 ohms
R117--10 ohms
The components R118, R112, C113, C114, CR107 and CR111 form a snubber
network. A second snubber network 100 of the same circuitry is also
included in the circuit.
In the circuit of FIG. 5, the group of FET's Q103, Q104, Q105 and Q106 form
the switch SW1 of FIG. 4. The diode CR115 corresponds to diode CR1; FET
Q109 and diode CR116 form SW2; capacitor C112 corresponds to capacitor C2;
comparator U101 corresponds to comparator U1; and resistor R122, resistor
R123, capacitor C120, resistor R124, and transistors Q107 and Q108 form a
drive circuit to assure adequate drive to FET Q109 from the output of
comparator 101.
In the circuit of FIG. 5, the reference voltage is not equal to the clamp
voltage, that is the voltage across capacitor C112. Instead, the reference
voltage is set to 10 volts by Zener diode CR114 and resistor R120. The
voltage across capacitor C112 is divided down to 10 volts by a voltage
divider formed by resistors R108, R109, R121 and R119. Capacitors C110,
C111, C119 and C118 are for noise suppression so as to prevent the
comparator U101 from actuating the FET Q109 in response to spurious noise
pick-up.
The invention provides, therefore, an improved transformer isolated
switched power supply of the forward converter type which includes a
clamping circuit for the switching means of the power supply to improve
the operation of the power supply.
It will be appreciated that while a particular embodiment of the invention
has been shown and described, modifications may be made. It is intended in
the claims to cover all modifications which come within the true spirit
and scope of the invention.
* * * * *
|
|
 |
|
 |
Description |
|
