An Improved Single Phase Full Bridge Inverter Using Sequential

Triggering for Achieving Zero Voltage without

Inductor


Prof. Dr. M.G. Osman, S. S. Eskander, Assistant professor, Eng. Ahmed Eid Mosa,
Electrical Power and Machines dept,
Faculty of Engineering,
Mansoura University, Egypt.
.

.
.

.
Abstract:
This paper introduces new technique for triggering full bridge inverter based on power transistor as
switches. This method based on the design of a sequential timer as triggering circuits. The designed
timer is designed to trigger the power switches of the two branches of the inverter and provides enough
delay time between the triggering signals of the inverter branches. The interval of the delay time must
be properly chosen to ensure that the current of branch reaches to its zero before applying the triggering
to the second branch of the inverter. Consequently, the load current crosses the zero value and goes
down safety to the negative direction. In this paper the full bridge inverter based on power transistor,
control strategy for triggering, interface stages between the timer and the transistors bases are designed.
The system undergoes is tested by using the lead acid battery and the solar cells array as a power
supply. The system offers efficient operation with the two supplies.
would be still on after removing the triggering
signal by a time equal to the transistor off time
Introduction
[1]. This phenomenon is not desired during the
Single phase full bridge inverter (SPFBI) is
operation. Most triggering circuit such as
used in wide range for transferring the DC-to
astable multivibrator circuits provides pulses
AC. It used also for motor drive. The mean
with no phase shift or time delay. This problem
problem for the inverter is how to trigger the
requires an application of soft-switching circuit
inverter taking into account the time duration
[2], or using a zero-voltage power switch, which
which the power switches still on after
may increase the electromagnetic interface
removing the triggering signal. This time is
[EMI] and requires a complicated snubber
equal to the power transistor storage time plus
circuit [3]. Therefore the cost of the inverter is
its falling time. This problem is very important
increased. The efficiency of it may be reduced
also and in some cases requires a large LCfilter circuit.
since the power switch is still on. So the
transistors of each branch of the bridge inverter
Analysis of Operation

The full bridge of a single phase inverter,

which uses four switching devices (power
transistor, GTO, IGBT,tc), is the most common
type of inverters. Since the source voltage is
switched at regular intervals to produce an
alternating output voltage. Varying the pulse
width of each cycle of the triggering pulse
controls the output of the inverter. This
technique is accomplished by changing the
control strategy of the transistor triggering.
Figure 1 shows a single phase inverter using the
power transistors as switches.
When switches Q1 and Q3 are on the current
flow from left to right, and when switches Q2
and Q4 are turned on (in the same time Q1 and
Q3 are off), the current flows from right to left
producing the negative half cycle of the
waveform. Hence, an alternating current is
produced. The pulse sequence on the bases of
the power switches and the inverter output
voltage are illustrated in table 1. When the pulse
signal is applied to the base of Q1 and Q3, the
output voltage will be equal to +E (where E is
the DC voltage level to be converted). On the
other hand, the application of the triggering
pulse on Q2 and Q4 bases results in an output
voltage level of E. The results are a variable
voltage output alternates between +E and E

levels. The main problem is that when the

triggering pulse is provided to Q2 and Q4 bases
at the falling edge of Q1 and Q3 pulses, Q1 and
Q3 transistors are still on ( since the transistor is

on during its off time ). Consequently, the pulse

applied to Q1 and Q3 must be terminated earlier
before applying the pulses to Q2 and Q4 bases.
Otherwise, the failure due to secondary
breakdown is carried out [4].
By proper designing of a proposed trigger
circuit involves sequential timer (ST), time
delay circuit and interface stages between the
inverter switches and the sequential timer based
on proposed control strategy. Proper design of
the sequential timer circuit results in a time
delay between the applicable pulses on the
inverter power switches. Figure 2 shows the
topological circuit of the designed sequential
timer. It has four stages consists of an astable
multivibrator and three cascaded monostable
multivibrator circuits. The proposed trigger
circuit is designed for producing two pulses
with a selected time delay between them. This
time must be equal or slightly greater than the
power switch off time. The frequency of each
pulse is the same and equals to the inverter
output.
Table 1 the power switch status and the
corresponding output voltage.
Q1
Q2
Q3
Q4
Vo
on

off

on

off

+E

off

on

Off

on

-E

Figure1. the single phase inverter bridge based on the power transistor switches

Figure 2. The sequential timer for producing the pulses with a time delay

Design of Triggering Circuit

The main circuit designed for triggering the
inverter power switches is composed of two
main parts. The first part of the circuit is a
sequential timer circuits. The second part is the
interface stages connecting between law rate
trigger circuits (sequential timer, ST) and the
high rate power switches. An accurate design of
each part is introduced in this paper.
Design of Sequential Timer Circuits
The sequential timer under design contains
four circuits. The first one is the astable
multivibrator (ASM) for obtaining a train of
pulses with frequency of 50 Hz. The second
stage of the timer is designed as a monostable
multivibrator (MONO1) which produces a pulse
with 9.99995 ms pulse width for driving power
switches Q1 and Q3 simultaneously. The third
stage of the timer, MONO2, is designed for
producing a pulse with pulse width slightly

greater than the transistor off time. The output

of MONO2 is triggered the fourth stage of the
timer which is designed as a MONO3. The
output of MONO3 generates the pulse required
for driving switches the power transistors Q2
and Q4.
Design of Astable Multivibrator
The astable multivibrator under design must
produce a pulse with frequency of 50 Hz and
duty cycle of 98%. Design of the astable
multivibrator circuit depends on IC555 time
data sheet specifications. The charging current
of capacitor CA, Ic, must be selected such that
[5];
Ic >>Ith , and

(1)

Ic >>Itri,

(2)

Where Ith is the IC555 threshold current and

Itri is the IC555 triggering current. The value of
.Ith and Itri are obtained from the timer data sheet

(a)

which are equal to 0.25

and 0.5
respectively. Consequently, the values of Ic is
selected as 1mA.
The design values of RA, RB and CA are
given from the following equations:
Vcc
3 Ic (min)

(3)

CA =

t1
0.693(RA+ RB|)

(4)

RB =

t2
0 .693 C1

(5)

RA + RB =

The typical values of the designed elements

are equals to 3198, 81.6 and 7.07
respectively. Figure 3a represents the shape of
output pulse of the astable multivibrator as the
circuit simulated by multisim program. On the
other hand, Figure 3 b introduces the
experimental output pulse from the designed
multivibrator circuit recorded by the
oscilloscope.
.

Design of Monostable Multivibrator

The monostable 1 multivibrator (MONO1) is
triggered by the output signal from the previous
astable multivibrator. At the trailing edge of the
astable multivibrator output initiates the
MONO1, the output pulse width (PW1) of
MONO1 must be designed such that

PW =

T
Toff
2

Where T is the periodic time of the inverter

output waveform and Toff is the power transistor
off time.
The schematic diagram of MONO1 is
illustrated in Figure4. The elements of MONO1
circuit are designed from the following
equation:

V CC
3 I CC
PW
=
1 .1R m1

R m1 =
C m1

(b)

(b)
Figure 3. astable output by (a) as simulated by
multisim program, (b) by the experimental
result.
The typical value of Rm1 and Cm1 are 160 k_
and 56.8 nF respectively. Figure 4 represents
the output pulses of the astable (Figure 4a) and
MONO1 multivibrator (Figure 4b) as given by
the simulation program as well as the
experimental pulse output from MONO1
(Figure 4c).

(6)

(a)

(b)

(7)
(8)

(a)

(c)
Figure 4. The MONO1 output with the astable
as a trigger for MONO1: (a) Astable output, (b)
MONO1 output by simulation, (c) experimental
results of mono1.

. Design of Monostable 2 Multivibrator

T off
1 .1 R m 2

Figure 5. the output of the MONO2: (a) by the

simulation program, (b) by the experimental
results.

Design of Last Stage of Sequential Timer

The MONO2 is illustrated in Figure 2 as a

third stage of the sequential timer. This
multivibrator is used for triggering the MONO3
after MONO1 by a time interval equal to the
power transistor turn off time. Hence its output
pulse must be accurately equal to Toff. For the
simplicity of the design procedure Rm2 is
selected equal to Rm1. The capacitance Cm2 is
obtained as follows,

Cm2 =

(b)

(9)

It this has a typical value of 2.84 pF. Figure 5

shows the experimental output pulse of
MONO2,

The last stage sequential timer under design

as illustrated in Figure 2 is a MONO3. The
MONO3 is designed such that its pulse output
has the same specifications of the output pulse
of MONO1. This means that, MONO1 and
MONO3 output pulses have the same pulse
width and frequency. Consequently, the design
of MONO3 electrical elements Rm3 and Cm3 are
similar to that of MONO1 and they have the
same value of Rm1 and Cm1. The output pulse of
this multivibrator is used for triggering the
power switches Q2 and Q4. Figure 6 represents
the output pulse from MONO1 (Figure 6a),
which triggers the power switches Q1 and Q3,
MONO2 (Figure 6b), which triggers Q2 and Q4
along with the inverter output wave form
(Figure 6c).

(a)

(b)
Figure 7. the complete interface circuit
so,
C1 =

3I * Tr
V cc

(12)

The typical values used for Rc and Cc are 20 k_

and 15 pF respectively.

(c)
Figure 6. the wave form output of: (a)
MONO1, (b) MONO3, and (c) the transformer.

Design of the Coupling Elements

The coupling elements Rc and Cc also shown
in Figure 2 are used for connecting the
sequential timer stages between each other. The
behavior of Rc resistor alike with that used with
normally off transistor circuit. It is used to
assure the off condition of the circuit. A typical
value of this resistance is in the range of
1k c20K_ [4]. The design of coupling
capacitor Cc depends on the rise time of IC555
timer used Tr. The capacitance of Cc is given by
I *T
C1 =
(10)
v
The current I represents the capacitor
charging current which is given by
V cc 0
(11)
I =
Rc
V
The change of Cc voltage is taken as cc
3

The Interface Stages

The first step of the inverter design is the
selection of suitable power switches capable of
carrying the load current. The designed inverter
has power transistors as power switches. The
base current required for driving the previous
switch is very larger than the sequential timer
output current. Hence interface stages must be
coupled between the power switches and the
sequential timer circuit. The number of the
interface stages depends upon the difference
between the driving current of power switches
and the current output from sequential timer.
The stages contain different types of transistors
with different ratings. Some of these stages are
designed as a normally on transistor circuits
while the remaining are the normally off
circuits. The design procedure of each stage is
similar to the design of a normally on or off
transistor circuit. Figure 7 illustrates the
designed interface stages coupled between the
low rate sequential timer triggering circuit and
the power switches Q1, Q3 and Q2, Q4.

.
this Figure shows the complete circuit
including the trigger sequential circuit, the
interface circuit and the inverter bridge circuit.

Experimental Performance of the Inverter

Figure 8 measures the output voltage with
insolation as an evaluation of the inverter
performance feeds directly from solar cell array.
This Figure shows the relation between the
output load voltage at different levels of
insolation over the day light to evaluate the
inverter performance with the solar insolation.

Figure 9. the circuit feed directly from the

battery.
So we must use a battery as a storage system
to feed the system with constant voltage and
provide continuous operation during the night.
The relation between output voltage and time
when the inverter is fed from the battery as
shown in Figure 9 indicates that the output
voltage is more stable and has low variations in
its value with the time. So the battery is the
essential component in the standalone
photovoltaic system to reduce the time-variation
of the voltage, this leads to nearly constant
voltage during the operation. Another parameter
may affects the output voltage is the pulse width
of the triggering circuit. Since the control of the
output voltage can be achieved by changing the
trigger circuit pulse width or by changing the
triggering pulse frequency.

Experimental Results
Figure 8. measured output voltage of the
experimental inverter with the different level of
insolation.
This Figure shows that the output voltage
may vary with the variation of the insolation,
since the insolation is not constant over the day
time and absent at the night.

Figure 10 shows the experimental results for

the waveform of the sequential circuit and the
inverter. These waveforms illustrate turn-on and
turn-off switches characteristics of the main
device under considering the time-delay in
order to verify the zero voltage transition
without any continuous operation of the power
switch in the inverter bridge. The interface
circuit directly coupled with the sequential
timer on off characteristics is represented by
this figure as an example of the whole stages of
the interface circuit.

(a) MONO 1 output

(b) MONO 2 output

(c) Interface 1 circuit output

(d) transformer output

Figure 10 experimental output waveforms from different parts of the inverter circuits.

Conclusions
In this paper, an improved method for the
single phase inverter was designed to
achieve zero-voltage transition, without
complicated snubber circuit or soft-switch
circuit. The application of these circuits
increases the cost and the complexity of
the inverter circuits. A comprehensive
analysis of the proposed circuit indicates
the zero-voltage transition without using
inductor in the circuit. Hence, this method
is characterized by the simplicity and low
cost. The simulation results were found to
be in a good agreement with experimental
results indicating that zero-voltage
operation during transition was achieved
by designing the proposed circuits..

REFERENCES:
[1] Ashfaq Ahmed Power Electronic for
Technology ,book , pp. 48-50, 1998.
[2] Brian T. Irving and Milan M. Jovanovi_
Analysis, Design, and Performance
Evaluation of Flying-Capacitor Passive
Lossless Snubber Applied to PFC Boost
Converter IEEE 2002.
[3] Byeon-Mun Song, Seong-Ryong Lee,
and Jih-Sheng Lai An Improved ThreePhase Auxiliary Resonant Snubber Inverter
for AC Motor Drive Applications
Proceeding of IEEE Power Electronics
Specialists Conference, Record, may1998,
pp. 423-428.
[4] S. B. Dewan and A. Straughen Power
Semiconductor Circuit, volume, 1997, pp.
357-359.
[5] F.F.Mazda," Power Electronics Handbook",
British Library Publication, 1990.