Switched Mode Power Conversion

Prof. Ramanarayanan. V
Department of Electrical Engineering
Indian Institute of Science, Bangalore

Module - 6
Lecture - 14
Isolated Converters - I

(Refer Slide Time: 00:29)

A good day to you all. In today’s lecture we will continue the ideas that we had developed
in the past into a new family of power converters. And these family of power converters
will be called the isolated power converters. What we had seen in the earlier lectures,
related to how this basic converter cell consisting of a single pole double throw switch and
an inductor L was used in different circuit topologies to get a power converter, which will
take in power at certain voltage Vg and provide power at a certain other voltage V0 .

All these converters had one polarity of the voltage common. The negative bus of the
source and the negative bus of the load are at the same potential. So, these converters are
called non isolated converters. The electrical potentials of the source side, and the electrical
potentials on the load side are all floating with respective to a common 0. Such converters
are non isolated converters. They have limited applications.
(Refer Slide Time: 01:53)

In general, a more general converter would try to provide power to a load connected at a
potential, which is different from the source completely, so that there is a total electrical
isolation between the source side and the load side. There is no common electrical potential
on the load side and on the source side. They are completely isolated by a barrier. On one
side is the source and on the other side is the load. Such converters are called isolated
power converters. They are also called galvanically isolated power converters because
there is no direct electrical connection between the source circuit and the load circuit. They
are coupled through appropriate; in this particular case, electromagnetic coupling, a
transformer which transfers the source side energy to a magnetic energy in the core which
in turn gets coupled to the secondary side as a different voltage and which is converted to
the required output voltage V0 which is at a different potential.

The isolation provided is electromagnetic isolation. These converters are called isolated
converters. There are a number of converters which belong to this family. Most power
converters belong to the isolated converters family for several reasons. One major reason
is that it is necessary to supply power at a potential which is different from the source. So,
you have to have an arbitrary potential isolation between them. Another reason is safety.
In many applications, you would like the load circuit load side circuit to be totally isolated
from the source side circuit for safety of equipment or people who are handling that. These
are the reasons where requiring isolation. In power converters, the most common isolation
is the electromagnetic isolation what we have written here.
(Refer Slide Time: 04:19)

So, in this lecture, we will try to look at the isolated power converter circuit topologies and
how they got evolved from the non-isolated circuit topology. Here, you can identify that
the source plus the load plus the single pole double pole switch identical to the non isolated
buck converter. What has now been added is another switch SPDT which is identical to S
and 𝑆̅ . It switches with the same relationships, on- off relationship as this S and 𝑆̅ . But,
in between these two switches we have introduced an electromagnetic isolation. There are
several important points, which have to be considered whenever we provide
electromagnetic isolation. One of the most important issues is known as flux balance. In
any electromagnetic transformer, it is necessary that the transformer operates with stable
flux which is within the saturation limits of the core, electromagnetic core that is used for
the transformer. This principle is that of flux balance.
(Refer Slide Time: 05:49)

We will see in the next few slides what this flux balance is and how do we achieve the
required constraints, required conditions, how are they achieved. So, we briefly review
magnetic circuits which are essential to understand the operation of a simple
electromagnetic power transformer. The transformer has a primary and a secondary. This
primary and secondary are wound on a common core and the core has certain flux. This is
an electromagnetic power transformer. It is isolated with; it is excited with certain voltage
on this side. The same frequency voltage of a different magnitude is obtained on V2. There
is a relationship between the exciting voltage and the secondary voltage or the primary
voltage and the secondary voltage through the number of turns N1 and N2 of the magnetic
core.
(Refer Slide Time: 06:53)

So, let us look at some of these ideas and then see what is meant by flux balance. The
fundamental relationship in an ideal transformer is that the voltages are related to the
𝑉
number of turns 𝑉1 . The ratio of the voltage V1 to V2 is the same as the ratio of the number
2

𝑁 𝑉 𝑁 𝐼
of turns 𝑁1. 𝑉1 = 𝑁1. In a same way, the ratios of the currents 𝐼2 is equal to the reverse ratio
2 2 2 1

𝐼 𝑁 𝑉
of the number of turns. For example, 𝐼2 = 𝑁1. 𝑉1 is the direct ratio of the number of turns.
1 2 2

𝐼2
is the inverse ratio of the number of turns. This is very simple to understand because in
𝐼1

a transformer where there are no losses V1I1 =V2 I2 that is, input power is output power.
𝑉 𝐼
On account of that, you will see that the ratio of 𝑉1 will be the inverse ratio of 𝐼2. That turns
2 1

𝑁
out to be the number of turns ratio or 𝑁1. In general, this is simply called turns ratio of the
2

transformer. The VI relationship of a transformer is given by the number of turns in the

primary and in the secondary.
(Refer Slide Time: 08:16)

There is also a relationship between the excitation voltage and the flux in the core. If we
represent the core flux as ∅ which is linking both the primary and the secondary and if this
transformer is ideal where there are no losses; in such a case V1 is the primary voltage is
𝑑∅
equal to N1 into rate of change of the flux in the core 𝑁1 𝑑𝑡 . The secondary voltage is N2

times the rate of change of flux. This is the relationship which is called the Faraday’s law
𝑑∅
of electromagnetic induction. 𝑉 = 𝑁 𝑑𝑡 when applied to this case. On the primary side,
𝑑∅
𝑉1 = 𝑁1 𝑑𝑡 . Flux is common for both primary and secondary. On the secondary side, 𝑉2 =
𝑑∅
𝑁2 𝑑𝑡 ; ∅ being the common link common linking flux between the primary and the

secondary. This relationship is the voltage to flux relationship.

(Refer Slide Time: 09:23)

If we go to the next step, what is the relationship between flux to voltage? That is inverse
relationship of ∅ and V. We try to invert this relationship and that turns out to be that the
1 𝑡
flux is nothing but 𝑁 ∫0 𝑉1 𝑑𝑡 .This relationship is simply obtained as inverse relationship
1

of the Faraday’s Law. The core flux can be found out by integrating the primary voltage
and dividing by primary turns or integrating the secondary voltage and dividing by the
secondary voltage over the period of integration 0 to t. This relationship relates how from
V1 or from V2 you can reach the flux in the core. This is the flux to voltage relationship.

(Refer Slide Time: 10:18)

Now, if a transformer is operating in the AC standard application where the excitation is
periodic, the primary voltage is a periodic voltage. Normally, it is a periodic sinusoidal
𝑑∅ 𝑑∅ 1 𝑡
voltage 𝑉1 = 𝑁1 𝑑𝑡 and 𝑉2 = 𝑁2 𝑑𝑡 . The flux is 𝑁 ∫0 𝑉1 𝑑𝑡 . This is the flux swing in a
1

transformer core. The core flux is related to the volts second integral of the excitation
voltage. For periodic operation, one of the important thing for periodic operation is that
the flux swing is symmetrical. It starts from the same point and at the end of the cycle it
comes back to the same point if it is a stable periodic operation. In such a case, to obtain a
stable flux swing under a periodic condition, the volt second integral of the excitation
voltage has to be 0. Now, this is the principle of flux balance-if the flux has to remain
stable then, over a period the volt second integral applied to that transformer has to be 0.

(Refer Slide Time: 11:53)

Now, let us see how this is seen in the standard sinusoidal excitation case. For example, if
I excite it with a primary voltage V1, which is a sinusoidal voltage; the sinusoidal voltage
integrated is the flux which you see here or the derivative, differentiation of this ∅ will
𝑑∅
give you V1. In this curve, you see that wherever ∅ is maximum, where = 0, the voltage
𝑑𝑡

is 0. At all other points, the derivative the slope of this flux function is the voltage function
or the integral of the voltage function is a flux function. These are the two relationships
between V and ∅.
(Refer Slide Time: 12:46)

If you see this curve, you can see the swing of the flux from the minimum to the maximum.
This swing is decided by the volts second balance or the double hatched area that I show
here. The swing in the next half cycle from maximum flux to minimum flux is equal to the
volts second area that I am just now seeing in the double hatched line. So, if the positive
swing in the flux and the negative swing in the flux have to be equal, if the positive swing
and the negative swing have to be equal under steady state; what it means is that the overall
cycle volt second integral has to be 0 or the area under the positive curve, positive half and
the area under the negative half have to be equal to each other. The same thing can be
mathematically expressed in another way; that the excitation voltage the average value of
the excitation voltage as to be 0. This is what we call as the principle of the flux balance.
(Refer Slide Time: 13:53)

But, in a power converter, in a switched mode power converter, we do not apply sinusoidal
voltages to transformers. Invariably, we apply voltages which are square wave in nature;
certain voltage for some time and some other voltage for some other time and so on. Here
what I show is a square wave excitation of a transformer V1, the primary voltage V1 is a
square wave of some frequency a square wave has positive half equal to the negative half.
This is T s / 2 and this is T s / 2 for half the period the voltage is plus V1 and for the next
half period the voltage is - V1.

(Refer Slide Time: 14:46)

So, in such a case if we try to see what is the flux? That flux is nothing but the integral of
the voltage. So, you see that the flux is now a ramp. It is a triangle wave which is nothing
but the integral of this V1 d t or V1 is the slope of the flux curve. In this half period, flux
is constantly rising. So, the straight line; the slope is constant and equal to V1 here. In the
negative half, the flux has a negative slope and a constant slope and you see a negative
voltage applied to that. If the average voltage is 0 and if the time period is half then, V1
and - V1 are equal. So, the net average voltage applied to the transformer is 0 over a period,
and because of that, the flux swing is symmetrical in such a transformer. In such a
transformer, the flux swing is + ∅ /2 and - ∅ / 2 and it is symmetrical.

(Refer Slide Time: 15:51)

The integral of one full cycle what you see as the hatched voltage here in one half is the
same as the hatched voltage here in the other half with the result that the average voltage
is 0. The flux has a stable periodic value with + ∅ / 2 as the maximum flux and - ∅ / 2 as
the minimum flux.
(Refer Slide Time: 16:23)

Supposing if the voltage applied is not symmetric for example, if this voltage what you
apply during the positive half and the duration of that; the product of that is not equal to
what is applied in the negative half. If the average, if the volt second over a period is not
equal to 0, in such a case what happens? If the average voltage is not 0, what will happen
to the flux? This is the question which will give us the precautions to be taken if you wish
to operate a switched mode power converter with transformer isolation.

(Refer Slide Time: 16:58)

You see here this positive half voltage and the negative half voltage-volts second area are
not equal. On account of that, in every cycle there is a small accumulation of voltage flux.
There is a small accumulation of flux. In the first cycle, during positive half flux increases
and during negative half, flux decreases. Because the positive volt second and the negative
volt second are not equal, flux reaches a net value at the end of one cycle and in the next
cycle it acquires a little more of flux and so on, and slowly the flux builds up. You can see
that the flux is now asymmetric and it builds up cycle by cycle. In every cycle, a little bit
of flux is accumulated and over a period of time the accumulated flux shifts, moves the
flux in the transformer to one side either positive or negative and this asymmetric operation
is not desirable.

(Refer Slide Time: 18:06)

What happens because of this asymmetric operation? The flux builds up, flux builds up in
the core asymmetrically and once it reaches the saturation level, the flux cannot build up
any more because every magnetic material has a saturation flux density, beyond which the
material is not capable of supporting additional flux. So, in such a case, as the flux is
building up, you can see that eventually, it hits a maximum and saturates. We say that the
core has now saturated or the flux has reached a limit beyond which the core is not capable
of sustaining additional flux.
(Refer Slide Time: 18:53)

What is the consequence of this? The consequence of this is saturation in the core.
Saturation in the core will make the transformer to act like a short circuit. A large current
will be drawn. The source will be overloaded and so on. Now, in this particular case we
are looking at the isolated converter which was obtained by simply introducing an
electromagnetic transformer to a standard source and our standard buck converter with
isolation in between. So, this is what we proposed as a circuit topology for isolated power
converter. Notice that during the on time when the switch is on, the voltage across the
transformer is VG same as supply voltage.

During the ON time, the primary of the transformer through this switch S is being
subjected to a voltage of VG. During the time when the switch is connected to S ̅, the
switch is connected to S ̅ and the transformer primary end is connected to 0 and we see
that when the switch S ̅ is on the voltage V1 or the primary voltage to the transformer is
0. We are applying during a part of the cycle, a positive voltage and during the next part
of the cycle is 0 voltage. It is very obvious that over one cycle, there will be a net volt
second in the circuit.

𝑡
You will have ∫0 𝑉𝑑𝑡 ≠ 0 because on one side it is positive volts second and in the other
side it is 0 volt second. So, there is nothing to balance and so this transformer will saturate
eventually. So, in a standard converter if we simply introduce an electromagnetic isolation
without taking adequate measures to balance the flux in the core; in such a situation this
transformer will be subjected to asymmetric flux excitation.

(Refer Slide Time: 21:12)

The transformer will saturate which is not a very desirable thing. If you build such a
convertor and operate it at a certain switching frequency with S ̅ and S as shown here in
this circuit, you see that the voltage of light to the transformer is unipolar. It is positive
during the switch S on time and it is 0 during the S ̅ on time. Because of this, the net volt
second across the transformer is only positive. It is not 0 and because of that the core flux
is building up every cycle. It builds up during the positive voltage during the 0 voltage
flux does not die, d ∅ / d t = 0 and because of that ∅ remains constant.

Next cycle again, current flux builds up and the off period flux remains constant. It is very
easy to see that over several cycles, the flux will reach higher level where it will saturate.
Core saturation is certain to take place and so this type of operation is not permitted.
Simply taking these switches, introducing transformer will not provide us the isolation
features. Then, how does one handle this? So, let us look at the mechanism of reducing the
flux to 0.
(Refer Slide Time: 22:51)

We saw in the previous one, the flux remains at high level during the off period because
we are not applying any voltage to the primary but if you want to reset the core; the core
as built up during on time flux has built up and during the off time instead of allowing it
to remain there, we would actively bring it down to 0 and keep it at 0, then start the next
cycle. If such an operation is possible then, you can see that in every cycle, the core flux
builds up and dies to 0, remains at 0 then, again builds up and so on. In such a situation,
the core flux will not build up eventually and the core will not saturate eventually. This is
something which we have to provide in the circuit topology.

We have to provide a mechanism of core flux reset or bringing the core flux back to 0 at
the end of every cycle. Now, how this, how do you do that? From this core flux, you can
see that if I want to bring the flux down, if I want to have a negative change in flux then,
we know that d ∅ / d t is relating to the voltage, so I have to apply some negative voltage
to the primary during the time I wish to reset the core, if the core as to be reset, apply a
negative voltage to the transformer primary during the off time. See if that is done during
the on time because of this positive voltage, the core flux is building up this positive
voltage results in a positive slope and during the negative time, negative off period, this
negative voltage will provide a negative slope to ∅ and we have an opportunity to reset the
flux all the way down to 0. The appropriate reset process is by providing here reset voltage
to the transformer.
(Refer Slide Time: 24:45)

Now, in the same circuit, what has been done is a reset source as been added. I call that as
- VR. The source voltage is VG. When the switch is connected to S, the transformer will
result in flux build up during the on time because of the voltage VG. During the off time
because the switch is now connected, here the transformer is subject to a negative voltage,
negative voltage of magnitude VR. So, you see that during the on time, a positive voltage
is supplied to the transformer and during the off time, a negative voltage is supplied to the
transformer. Now, we have an opportunity to balance the flux because in one direction,
flux is increasing and in the other duration, flux is decreasing. So, if we correctly select
the durations, it is possible to make the flux come back to 0 at the end of every cycle.
(Refer Slide Time: 25:44)

So, what we see is that process-how is this reset been done? Let us look at the transformer
𝑉 𝑁 𝐼 𝑁
with one non ideality. The ideal transformer has the relationship that 𝑉1 = 𝑁1 and 𝐼2 = 𝑁1 .
2 2 1 2

Now, in the ideal transformer, 𝑁1 𝐼1 = 𝑁2 𝐼2 . What you see from here? The product 𝑁1 𝐼1 and
the product 𝑁2 𝐼2 . are equal. On account of that, in an ideal transformer, there is no net
flux because the magnetizing in the core takes place because of 𝑁1 𝐼1 in the direction shown
here - 𝑁2 𝐼2 . in the direction shown there. So, the net ampere turns in the core is 𝑁1 𝐼1 .

The secondary ampere turn is 𝑁2 𝐼2 .. Because these both are, one of them is flowing away
from the dot and the other is flowing towards the dot. The net magnetizing ampere turns
is a difference between 𝑁1 𝐼1 and 𝑁2 𝐼2 . or we say that the magnetizing ampere turns is
𝑁1 𝐼1 − 𝑁2 𝐼2 . Now, this can be seen as an equivalent magnetizing current on the primary
side or on the secondary side. The magnetizing current is nothing but the net magnetizing
ampere turns divided by the turns in the primary if you want primary current or secondary,
if you want the secondary referred magnetizing current. You can see that this current is
𝑁1 𝐼1 −𝑁2 𝐼2
called the magnetizing current Im which is . This N1 is to refer the total
𝑁1

magnetizing current as if it is a current flowing through N1 only. So, this is the magnetizing
current refer to the primary. If we wish to get Magnetizing current refer to the secondary,
𝑁1 𝐼1 −𝑁2 𝐼2
it will be .. Okay?
𝑁2
(Refer Slide Time: 28:10)

Now, the flux is decided by whatever is the ampere turns divided by what is known as the
reluctance of the machine. So, this quantity which is in the denominator, the magnetic core
𝐿
length L divided by the magnetic area A and then the magnetic permeability 𝜇;𝐴𝜇; we have

seen this as the reluctance of the magnetic circuit R and the net flux in the core is related
to the magnetizing ampere turns divided by the reluctance of the circuit. In an ideal
transformer, this quantity will be 0 because N1 I1 and N2 I2 are equal. In a real
transformer, these two will not be equal and the difference between the primary ampere
turns, N1 I1 and the secondary ampere turns, N2 I2 is what is establishing the core flux ∅.
So, this is a net ampere turns in the core 𝑁1 𝐼1 − 𝑁2 𝐼2 .. This is the reluctance of the
magnetic circuit and we see that flux is the ampere turns divided by the reluctance of the
magnetic circuit.
(Refer Slide Time: 29:27)

𝑁1 𝐼1 −𝑁2 𝐼2
Now, if we say that the magnetizing current referred to the primary is ., this is the
𝑁1

primary reflected magnetizing current. We also have another relationship which is the
electrical side relationship of the link flux linkage and the magnetic side relationship of
the flux linkage. Flux linkage in electrical side is given by LI or magnetizing inductance
L m multiplied by the magnetizing current I m is the flux linkage in the core. It is also the
same as N1∅ which is the flux linkage in the core multiplied by the number of turns in the
primary.

Here, we have taken the Im as the magnetizing current referred to the primary. It will be
equal to N1 ∅. If you replace ∅ as 𝑁1 𝐼1 − 𝑁2 𝐼2 by reluctance, this multiplying by N1 and
then replacing this entire thing by I m N1 will give rise to a relationship here, which is the
𝑁12 𝜇 𝐴𝐶 𝐼𝑀
total flux linkage . This is the total flux linkage and if you wish to find out what
𝐿𝐶

is L m, divide this quantity divided by I m-what is left with is the magnetizing reluctance.

Magnetizing reluctance is nothing but if it is referred to the primary side, it is 𝑁12 divided
by reluctance. Reluctance is a common quantity both for primary and secondary. N1
belongs to the primary turns. N2 belongs to the secondary turns. So, if you want
magnetizing current refer to primary, it is 𝑁12 by reluctance and if you want the
magnetizing inductance on the secondary side, it will be 𝑁22 divided by reluctance. This is
a relationship between the various quantities of interest to us. Lm magnetizing inductance
related to the number of turns on the primary and the reluctance of the magnetic circuit.

(Refer Slide Time: 31:53)

So, what we can do is that the entire magnetization process?

(Refer Slide Time: 31:59)

What we had seen in the previous slide?

(Refer Slide Time: 32:06)

As L m I m is what the flux linkages, it can be separated as a separate part of the equivalent
circuit. What we have here is the ideal transformer where this 𝐼2′ and I2 are the ideal
primary current and secondary current. 𝑁1 𝐼2′ = 𝑁1 𝐼2 . Because this is ideal transformer,
there is no flux in this ideal transformer core. All the core flux we have separated in a
magnetizing inductance Lm into which a current of Im is flowing.

This product L m I m is setting up a flux in the core which is ∅. So, when we represent
this magnetic process in a transformer, as an ideal transformer plus a magnetizing
inductance then, it is much easier for us to develop many of the new topologies that we are
going to look at. The magnetizing current is represented as a shunt non-ideality of the
transformer. Now, this transformer because it has no flux accumulation and we can have
any type of switching even if the volt second is not balanced for this ideal transformer, it
is ok. This is the equivalent circuit.
(Refer Slide Time: 33:18)

So, let us look at now the convertor which is called the forward convertor. This circuit
topology is called the forward convertor. It is one of the very first isolated power
convertors which we are now considering. It is a forward convertor, we have a buck
convertor up to this point. Then a transformer isolation has been introduced. This
transformer has both the non ideality as well as the ideal property. We represent the ideal
property of the transformer by this and then connect the magnetizing inductance as a
parallel non-ideality as a shunt non ideality, to this we connect a voltage of VG and VR, a
voltage which is applied to the transformer during non time VG and a voltage that is
applied to the transformer during the off time called V R because V G and V R are of
opposite polarity, you can see that now we have introduced a mechanism of resetting the
core.
(Refer Slide Time: 34:28)

We will have to now work out what the other quantities of interest are...

(Refer Slide Time: 34:35)

Now, the circuit realization of this circuit VR is not provided by a source because in this
circuit, in the previous circuit that we had seen, the magnetizing energy eventually flows
to flows through this VR, during the time S is connected to the reset circuit. See in such a
case, power is being supplied into VR. Power is not being drawn out of VR.
(Refer Slide Time: 34:59)

It is dissipated in VR. If that is the case, VR can be a dissipated circuit element and the
way to realize this circuit is to connect to the transformer parallel a circuit which has a
constant voltage. This circuit, when current is flowing through this magnetizing through
this reset branch, you will find that the current is in the direction of diode, so no voltage
drop. The current is against the direction of the zener, so you have a zener voltage drop
with a - sign and that will be equal to our reset voltage and supply voltage is VG. The
transformer has turns ratio of N1 to N2 and the usual reacting elements of Lc are there and
this is the load resistance R. Now, this is the circuit realization of a very simple forward
convertor.
(Refer Slide Time: 35:45)

So, let us look at how the voltage relationship between the input and the output will occur
in this case. We go back to the old method. Whenever we want to find out the relationship
between VG and V 0, we carry out a volt second balance condition on the inductor or we
find out what is inductor voltage during the positive half, and what is a inductor voltage
during the negative half.

If the inductor has to have a sustained operation, this positive half must be equal to the
negative half. If such is the case, then the inductor will have flux which does not keep
accumulating or keep dropping down. So, for steady state operation, the voltage across the
inductor must satisfy the condition that over one full cycle, the volt second balance across
the inductor will be 0. In this case during the on time, the voltage across the inductor is
𝑁
VG. We consider all the switches to be ideal.𝑉𝐺 𝑁2 is the voltage at the secondary point
1

and that voltage on the left hand side of the inductor and V 0 is there on the voltage on the
right hand side of the inductor. So, during on time the voltage across the transformer
𝑁
primary is 𝑉𝐺 𝑁2 − 𝑉0, but during the off time the voltage across the transformer across the
1

inductor is just V 0.
(Refer Slide Time: 37:20)

If we balance the volt second between these two, you find that T ON multiplied by the ON
state voltage - T OFF multiplied by OFF state voltage has to be 0. If this equation is 0,
then we get our famous voltage conversation ratio relationship. The output voltage is
related to the input voltage through a turns ratio which is corresponding to the transformer
multiplied by the duty ratio which we had seen in the non isolated power convertor.

This volt second balance, if we carry it out on the inductor we will get the relationship
between V 0 and VG. This is as before; nothing new in this. But, because the transformer
has flux build up and we have mechanism; we need mechanism not to build up the flux.
We want to find out the process of magnetization of a transformer and then modify the
circuit so that there is no core flux saturation build up is taking place.
(Refer Slide Time: 38:28)

Now, the relationship between IG and I 0 does not change. The reason is, you see that
everything is been ideal. This I m is also 0. We are assuming that current I m is neglected.
See in such a case, the primary current average is nothing but the secondary reflected
𝑁
current 𝐼𝐿 𝑁2 𝑖𝑠 multiplied by the duty ratio D. It is really the average current IG and from
1

𝑁 𝑉
this, you can write the current ratio is duty ratio multiplied by 𝑁2. This is the same as 𝑉0 .
1 𝐺

We have seen that under ideal conditions, the forward voltage transfer ratio and the reverse
current transfer ratio will be identical. In the non isolated convertor, it was D. In the
isolated convertor, we have a additional term which is the turns ratio of the transformer.
Now, we have added one more degree of freedom; one more degree of design freedom.
(Refer Slide Time: 39:42)

For example, from a given voltage of V 0, it is V given voltage of VG, it is possible to

obtain the same V 0 with two different quantities for the turns ratio and the duty ratio of
𝑉
operation. Now, it is possible to delink the ratio of 𝑉0 from just D. It is now a function of
𝐺

𝑁2
and D. So, it is possible to operate such a converter at a very low duty ratio with large
𝑁1

turns ratio or with the higher high duty ratio with a low turns ratio. So, we have added an
additional degree of freedom for the designer. You can get the same voltage conversion
ratio from two different converters employing two different turns ratio and two different
duty ratios.
(Refer Slide Time: 40:39)

The current ratio in the ideal converter is the same as the reverse of the voltage ratio.

(Refer Slide Time: 40:53)

When you multiply both of them, you get the efficiency which is 1. In an ideal converter,
𝑉0 𝑁 𝐼0 1𝑁
the efficiency is 1 because the = 𝐷 𝑁2 and = 𝐷 𝑁1 . You multiply them, you get
𝑉𝐺 1 𝐼𝐺 2

𝑉0 𝐼0
= 1 and this is nothing but efficiency. What we define, the input power or output
𝑉𝐺 𝐼𝐺

power divided by the input power is the efficiency. That efficiency is 1. You know ideal
switched mode converter with isolation as shown in this case.
(Refer Slide Time: 41:29)

In such a converter, what will be the current ripple in the conductor? This is something
which we have done several times before. Inductor current ripple is always related by the
inductor voltage. If the inductor voltage is integrated, that will give us what is the current
𝑁
ripple on the inductor current. Here, the inductor voltage is VL and that is equal to 𝑉𝐺 𝑁2 −
1

𝑉0 during the on time. It is - V 0 during the off time. If you integrate this quantity, you
should be able to get how the current rises during the on time which is this and how the
current falls during the off time which is what I have shown by the arrow here. The IL is
V 0 / R which is the output ratio and Δ L can be written based on the slope.

Here it is voltage is V 0 inductor is L; V 0 / L is the slope and the duration is (1 – D)TS.

So, this relationship is Δ IL represented as a function of V 0, L, D and TS. This relationship
is the same as the load current which is V 0 / R. In a dc to dc converter, the output voltage
divided by the output resistance decide the output current and output current in this
converter is identical to IL or the inductor current. Now, if you add, combine these two you
get the current ripple in the inductor is now a function of TS /( L/R) and multiplier of 1 -
D or what it says in other words that the switching frequency determines the current ripple.
The natural time constant inversely determines the current ripple and 1 - duty ratio directly
relates to the current ripple.
(Refer Slide Time: 43:33)

If current ripple has to be small, we need this quantity to be very much less than 1 and that
is obtained by getting this dominant ratio. Here, TS has to be very much less compared to
L/R. If TS is very much less in comparison with the circuit time constant L/R then, in such
a converter you will find the current ripple in the inductor will be negligible.

(Refer Slide Time: 44:05)

Now, just as we have been doing in all the other converters, it is only a matter of keeping
account of the various voltage drops, if we wish to know the non ideality of the switches.
The switches have a voltage drop. For example, this switch will drop by voltage of VT
whenever it is on. Whenever it is off, no current is flowing through that and this switch
whenever it is on, it drops a voltage of VT which is VD in one direction and this switch
drops a voltage of VD in the other direction during the off time.

So, all the voltage drops in the converter. If we neglect all the other things, you will find
that the volt second balance on the inductor is determined by VG - VT which is the voltage
across the transistor, the difference between them is now across the transform primary. So,
the primary voltage into N2 / N1 will be the voltage across the secondary. From that, you
subtract one diode drop VD and then on the other side is voltage V 0. So, you subtract
another V 0. So, this quantity, this is voltage applied across the primary for the on duration.

Similarly, the voltage applied across the secondary during the off duration across the
inductor during the off duration. So, from this it is possible to relate VG in terms of V 0.
VG is now a function of V 0 through several variables. They are duty ratio, turns ratio N2
/ N1 and then the conduction ratio of the transistor. Similarly, conduction ratio of the diode
VD by ideal output Voltage. So, this is the non ideal performance of an isolated
transformer which has a voltage drops across the switches.

(Refer Slide Time: 46:10)

This quantity can be simplified, so that the ratio of this voltage is the ratio of this quantity
𝑉 𝑉
1 − 𝑉𝑇 − 𝑉 𝐷 . This represents the non ideality of the switches.
𝐺 𝑂𝑡
(Refer Slide Time: 46:30)

Now, we can try to see what happens to the current conversion ratio. See, earlier all the
converters that we looked at, the current conversion ratio did not change at all because we
have never come across up to now any converter which has a shunt non-ideality. We had
a series non ideality of switch, diode resistance, resistance, resistance and so on but we
had not had a converter which had a shunt non ideality. Forward converter is a first
converter which we come across which has a shunt non-ideality which is I m. On account
of that, this 𝐼2′ h T will be the same. 𝐼2′ is the secondary current. 𝐼1′ is the primary current I
0. I 0 is what is flowing through that.

If you reflect to the primary you get a steady current of I2. This is what is flowing in the
idealized transformer but the magnetizing inductance we have separated as a separate
process. The establishment of flux in the core is decided by a certain current in the
transformers. We have defined all those currents 𝑁1 𝐼2 − 𝑁2 𝐼2 is the net magnetizing
ampere turns. The net magnetizing ampere turns divided by the flux density is the net, net
ampere, net magnetizing ampere turn is divided by the reluctance is the net flux. Now, if
you see the magnetization process because a positive voltage is applied to the transformer
VG during the on time, the transformer magnetizing current increases and corresponding
to this I m peak, the magnetizing current peak, certain amount of flux is established in the
core.
Then, when we switch this off this magnetizing current comes to 0 because this has been
switched off but the inductor current can never come to 0. The switch current can come to
0 because switch has no associated stored energy. So, when this current comes to 0 and if
the inductor current is continuing where does the current go? It has a path here through
this and then back here. So, you see that because the main switch has completely switched
off, it is not supporting any energy in the inductor. Inductor finds an alternate path which
can keep the energy flowing or will conserve the energy in the core and this will be the
path for that.

So, if we now write the equations for the primary current I m it is this; VG divided by Lm
multiplied by the on duration DTS. If this is the volt second balance divided by mutual
inductance, what we have is the peak magnetizing current seen in this picture. But, now
during the off time the current through the inductor continues to flow through the
demagnetizing circuit. In the demagnetizing circuit the voltage is constant because it is
made with the zener.

(Refer Slide Time: 49:54)

So, during this demagnetizing interval, voltage across the transformer remains constant.
So, that will result in the flux being reset gradually. So, in the current conversion ratio
because this is the additional current which is been drawn from the source to magnetize
𝑁
the transformer. The actual input average current is the load current reflected 𝐷𝐼0 𝑁2and
1

the average contributed by the peak magnetizing current of the transformer and this is very
easy to find. The area of contribution by this is I peak * DTS, which is this duration; that is
the total ampere turns divided by because the average value is one half of this current is a
triangle one half and this base and that multiplied by D will be the actual reflected current
on IG.

𝑁
So now, if we find the ratio of IG to I 0; it has the ideal gain 𝐷 𝑁2. But now, all the other
1

𝑁 𝑉 𝐷𝑇𝑆
variations have been combined into a single correction term, which is 1 + 𝑁1 2𝐿𝐺 . So,
2 𝑀 𝐼𝑂

from this you can see several things. For example, if supply voltage is high, VG is high
then, you will find that I G’s component of magnetizing current is high. So, also duty ratio
and switching time T-all these quantities directly relate the average input current. On the
other hand, there are four other quantities which indirectly determine the magnetizing
current-N2 two times L m and then multiplied by I 0; this division of I 0 is to normalize
the magnetizing current with reference to I 0. So, this relationship is the current conversion
ratio and because the converter has non idealities both in the shunt sense and in the series
sense, you see that there are now two correction factors.

(Refer Slide Time: 52:09)

The shunt non ideality has a correction factor because of this magnetizing current. The
switch non ideality has a correction factor as shown here. The ideal part remains the same.
Ideal part remains the same. So, we can say that this is the fraction by which the voltage
conversion has changed from 1. This is the ratio by which the current conversion has
changed from 1. What you notice here is that the dc current on account of the
magnetization current is going up on the input side. The dc voltage on account of the
several switches is going down on the output voltage. The efficiency is now the product
of these correction terms.

(Refer Slide Time: 53:09)

You will find that the efficiency is now a correction factor because of the switches. The
𝑉 𝑉
switches contribute a correction factor 1 − 𝑉𝑇 − 𝑉 𝐷 by ideal output voltage. The
𝐺 𝑂𝑡

𝑁1 𝑉𝐺 𝐷𝑇𝑆
magnetizing current of L contributes to 1 + 𝑁 . So, efficiency is this composite
2 2𝐿𝑀 𝐼𝑂

number here.
(Refer Slide Time: 53:38)

If there are resistors on the parasitic paths, for example, primary has a resistance.
Secondary winding of the transformer has a resistance and the inductor has a resistance. If
such a resistive paths are there, all that we have to do is only to write the volt second
balance equation again and keeping in this account, the drop in voltage on account of R P,
R S, R L and so on. Once that is done, you will have additional terms which are
contributing to the efficiency being less than 1.

This numerator quantity is less than 1 - some quantity and the denominator is 1 plus some
quantity. In an ideal converter, this x and this 𝑋 ′ both are 0 but in a real converter this has
some effect/penalty; the output voltage drops because of the series voltages of the devices.
There is a similar penalty on account of the conducting paths all over the place, so this is
in the denominator with 1 + 𝑋 ′ . So, whatever is the additional term other than the ideal
𝑁
gain; this is our ideal gain- 𝐷 𝑁2 , duty ratio. All the other terms are contributing to
1

deterioration of voltage gain and deterioration of current gain.

(Refer Slide Time: 55:22)

If you put both of them together, what you get is the efficiency in the power converter.
Then, we have the loss in the power converter also can be found out because every time
the switch is turning on, certain amount of energy transfers from the source to the inductor
from here to this point. But then, during the off period this energy is completely transferred
to VR or the zener diode. Whatever energy was first built up 0.5 Lm I m Peak square
multiplied by switching frequency; this is the total energy. It has been taken from the
source. Every cycle this is a total energy but in every cycle 0.5Lm Im2 is taken in. In one
one second, there are FS number of cycles.

If you multiply these two, this is the power loss or the energy loss per second which is
equal to the power loss. This can be modified and put in slightly different form. This relates
how much energy loss is there in the circuit on account of the magnetizing inductance,
switching frequency, operating duty ratio, operating source voltage. If all these numbers
are known, it is possible to find out the power loss in the device.
(Refer Slide Time: 56:46)

Now, what we have done is we have introduced here reset voltage against supply voltage
of VG. In most converters, the preferred value of VR is the same as VG. We tend to make
designs where VR and VG are equal. In such a situation, the core flux increase and the
core flux decrease; will take the same time if you do not want the flux to build up. So, in
such a case the maximum duty ratio of operation is limited to 0.5 in forward converters.

(Refer Slide Time: 57:27)

The voltage waveform across the switch if you see during the time the flux is building up,
the voltage across the switch is 0 because it is conducting and during the off time this
device is losing its energy in the zener. So, the voltage at this point is VG - V R or total of
VG +VR; that is additional voltage during the reset time. After the reset is over, there is
no current in the inductor. The voltage at this point is the same as the voltage at this point
which is - VG, which is this quantity. So, you can see that one of the important
consideration is that during the off time the device is experiencing double the voltage.

(Refer Slide Time: 58:18)

On account of this, the primary and secondary conduction times are forced to be equal.
(Refer Slide Time: 58:29)
You cannot have a forward converter whose duty ratio is more than 0.5. So, what we will
do now is we will conclude at this point with one type of forward converter where
maximum duty ratio is limited to 0.5. Magnetic core is utilized only in one direction.
Again, 0.5 on account of that the conductors are utilized only half the time, because the
windings are carrying the current only for 50 percent of the time.

The circuit voltage is limited to half, because the maximum voltage on the device is VR +
VG which is 2 VG. So, all these on every count a forward converter is only half utilized
in effect half, half, half, half; only one sixteenth of the potential utilization and further
there is an additional magnetization loss. We will stop here with this, and then in the next
lecture continue with many other isolated converters, where many of these problems are
addressed.