ST AN1731 Application note

AN1731

APPLICATION NOTE

EMERGENCY LIGHTING APPLICATIONS

1. ABSTRACT

This application note shows the topologies implemented in the Emergency Lighting Applications and the
STMicroelectronics’s power bipolar transistors used.
Today, the Emergency Lamps market has grown considerably due to the new improved safety rules. In
fact, the Emergency Lamps are used in all public places and, also, in private homes replacing the
traditional lighting applications.

2. STSA851 DESCRIPTION

The STMicroelectronics's power bipolar transistor STSA851 is housed in the TO-92 package. This
device is manufactured in NPN planar technology using a 'Base Island' layout that involves a very high
gain performance and a very low saturation voltage.
The main characteristics of the STSA851 device are:

1) V

2) V

3) V

4) Ic = 5 A (continuous current);

5) Ib = 1 A (continuous current);

6) V

7) Hfe = 270 (typ) @ Ic = 2 A @ Vce = 1 V (typical conditions).

≥ 60 V;

ceo

≥ 150 V;

ces

≥ 7 V;

ebo

= 140 mV (typ) @ Ib = 50 mA @ Ic = 2 A (typical conditions);

ce(sat)

G. Consentino

3. HIGH EFFICIENCY DC-AC CONVERTERS

The part of the circuit used to drive the emergency lamp is composed of DC-AC converters. The DC-AC
converters transform the low DC input voltage in high AC output voltage required by the fluorescent tube.
Fluorescent tubes are employed in these applications because they are much more efficient at
converting electrical energy into light than conventional incandescent bulbs increasing the battery life.
Usually, DC-AC converters used in these applications are the Push-Pull switching converter forced to run
in synchronized mode by the inclusion of a supply inductor, and the Forward converter. Mainly, the DC­AC converters have suitable transformers that increase the output voltage and allow the electrical
isolation between the secondary and primary of the transformer, and suitable switches. Usually, the
switches are power bipolars driven by a third winding magnetically coupled to the transformer, like in the
PUSH-PULL current FED converter.

February 2004

1/35

AN1731 - APPLICATION NOTE

The power bipolar transistor collector current Ic depends on the load, turns rapport

N

2

K

=

N

1

(3.1)

2

where N2 is the secondary turn number and N1 is the primary turn number of the transformer, and it also
depends on the battery voltage.

Usually, in these applications the lamp power is in the range of 8-24W, and the turns rapport

N

2

N

1

2

(3.2)

is about 30, the current Ic is in the range of 1.5-3.0A. Furthermore, usually, the emergency lighting
boards are powered with an input voltage in the range of 3.6-6.0 Vdc so that the typical V
10-20 Vdc. The voltage and current values ranges, V

and Ic, are inside the SOA of the STSA851

ce_max

ce_max

is around

so that these devices can be used in all emergency lighting applications. The emergency lamp
applications drive a lamp up to 58W. Usually, these emergency lighting applications do not supply such
output powers but only 10-20 % of the nominal lamp power. Sometimes, such applications are used to
power lamps that commonly light up rooms. When the net voltage disappears the emergency lamp
switches on supplying around 10-20 % of the nominal lamp power just to light up the room.

4. FLUORESCENT TUBE CHARACTERISTICS

Fluorescent lamps are generally made with tubes filled with a gas mixture at a low pressure. The inner
sides of the tubes are covered with fluorescent elements. When the net voltage disappears, before the
tube lights on, the lamp has a higher resistance. In this moment, the electrodes voltage increases up to
around 500V and the electrodes start to warm up and emit ions. Figures 1 and 2 show the V-time and I­time waveforms and the V-I waveform respectively before the start-up of a 24W tube.

Figure 1: V-time and I-time Waveforms Before the Striking

2/35

Figure 2: V-I waveform before the striking

AN1731 - APPLICATION NOTE

As shown above, to strike the fluorescent tube the electrodes voltage reaches up to 505V of peak.
Furthermore, the current that flows through the lamp is very low, 56 mA, because the resistance
before the striking is high (around 10 KOhm).
When the fluorescent lamp lights on, the gas mixture inside is fully ionized, and an arc across the
electrodes occurs. In this new condition, the lamp resistance drops to around 1 KOhm value (Figures 3
and 4 show the V-time and I-time waveforms and the V-I waveform after the striking.

Figure 3: V-time and I-time waveforms after the striking

3/35

AN1731 - APPLICATION NOTE

Figure 4: V-I waveform after the striking

After the striking, the gas mixture emits radiations that excite the fluorescent elements inside the tube
producing the light in the visible spectrum. In this example, after the striking, the voltage across the
electrodes drops from 505V of peak to 220V of peak and the current increases from 56mA of peak to
158mA of peak.

Usually, after the striking, in order to increase the lamp efficiency up to 15%, the operation frequency is in
the range of 25-30KHz. Furthermore, as shown in Fig. 4, the waveform I-V has a linear behavior until the
established voltage value is kept. In fact, if the voltage across the electrodes overcomes this established
voltage value, the characteristic becomes flat because no ion can emit other radiations.

4/35

AN1731 - APPLICATION NOTE

2

2

5. PUSH-PULL CURRENT FED CONVERTER TOPOLOGY INTRODUCTION

As previously exposed, a topology solution for emergency lighting applications is the PUSH-PULL current
FED converter topology. This topology solution has a Push-Pull switching converter forced to run in
synchronized mode by the inclusion of a supply inductor.

Figure 5: PUSH-PULL current FED converter schematic circuit

The components values of capacitors, resistors, and inductors are selected operation on the input voltage,
power lamp, and operation frequency.

6. TRANSFORMER DESCRIPTION OF PUSH-PULL TOPOLOGY

In figure 5 the transformer named T1 has three windings. The primary winding vices are connected to the
collectors of the NPN power bipolar transistors Q1 and Q2. The same primary winding has a central vice
where the inductor L1 is connected. The secondary winding vices are connected to the load.

The third winding vices are connected to the base of the transistors Q1 and Q2 so that when the first is on,
the second is off and vice versa. During the Q2 on state, the current flows through the same device and
the respective half primary winding and vice versa. Usually the primary inductance LT of the transformer

T1 is much lower compared to the inductance L1. The resonance frequency of the PUSH-PULL converter
is also due to LT. N2 (secondary winding turns) and N1/2 (half primary winding turns) rapport is around 60,
while N1/2 and N3 (third winding turns) rapport is around 5. Considering a 6 Vdc input voltage, the voltage
v

(the max voltage across the vice of the primary winding central point and the reference) can be

1max

written as:

14.3

v

V

dcmax1

≅=⋅π=

(6.1)

V96

5/35

AN1731 - APPLICATION NOTE

2

v

(the max voltage across the secondary winding vices) can be written as:

2max

N

v

v

(the max voltage across the vices of the third winding) can be written as:

3max

V

dcmax2

N

2

v

1

(

V

dcmax3

2

14.3

2

2

)

2

N

N

14.3

3

1

2

6

2

V560606

≅∗=⋅π=

1

V2

≅∗=⋅π=

5

(6.2)

(6.3)

As exposed above, it is highlighted N1/2 and not N1. In order to understand the reason of it, it is
necessary to consider the graph below.

Figure 6: Particular of T1

When Q2 is on, Q1 is off and vice versa. Now, considering fig. 6 where T2 is on; the current ‘I’ flows
through the half primary winding 'b' and it generates a magnetic force (Hopkinson law):

N

1

Φ⋅ℜ=⋅I

(6.4)

Φ is the magnetic flux and ℜ is the magnetic reluctance of the T1 core;Φ can be written as:

N

1

⋅=ΦI

2

(6.5)

ℜ

ℜ can be written as:

(6.6)

Al⋅=ℜµ

µ is the core permeability, A is the core section and l is the core length. When T2 switches off, T
switches on, the current flows through the other half primary winding 'a' and the flux Φ inverts its

direction. Such flux flows into the transformer core creating a link with N2, N3 and also with the other

6/35

1

AN1731 - APPLICATION NOTE

∆Φ

=

=

=

turns N1/2, generating the voltages v2 and v3 (magnetic law-Lenz law):

Nv

−=

−=

v

2/1

v

N

v

2

2

v

1

,

N

v

1

2

;

22

t

∆

∆Φ

Nv

33

t

∆

N

∆Φ

1

−=
∆

2

N

3

3

N

1

1

2

(6.7)

;

t

v

2

c

2,

===

v

1

(6.8)

Furthermore, i2 (the current that flows through the lamp) can be written as:

N

1

1

2

Ii

2

I

==

K

N

2

(6.9)

In fact, the apparent input power can be written as:

in 1

IvA

(6.10)

The output power can be written as:

ivA

22

out

(6.11)

Considering an ideal transformer:

Iviv

122

(6.12)

i 1

I

2

12

v

2

===

2

kN

(6.13)

N

1

v

Before the lamp strike, or when the lamp is disconnected, the operation frequency (about 60 KHz) is due
to the resonance between C2 and the primary transformer winding inductance LT (see fig. 7).

Figure 7: Resonant Schematic Circuit Before the Lamp Strike

7/35

AN1731 - APPLICATION NOTE

=

fπ⋅=

221LTC

(6.14)

When the lamp is connected, the transformer circuit can be showed as in the graph below.

Figure 8: Ideal Schematic Circuit of the Transformer After the Lamp Strike

The input apparent power can be written as:

ivA

11in

(6.15)

Now it is possible to consider an equivalent circuit to fig. 8, as in fig. 9, where the apparent input power is
equal.

Figure 9: Equivalent Schematic Circuit of the Transformer After the Lamp Strike

Furthermore, after the lamp strike, the resonant schematic circuit can be represented as in figure 10
where, usually, the operation frequency is due to LT, C2 and C1K2 (25-30 KHz).

8/35

AN1731 - APPLICATION NOTE

k

k

Figure 10: Resonant Schematic Circuit of the Transformer After the Lamp Strike

In this transformer equivalent circuit the output impedance has been transferred from the secondary
winding to the primary winding.

2

(

2221

Lamp

1

)

jRiivIv

−==

C

⋅

ω

1

(6.16)

Iv

1

( k

2

i

2

Lamp

v

jR

1

z

1

I

Where:

is the primary equivalent resistance and where:

is the primary equivalent capacitance.
Now, the equivalent primary admittance (Y

Y

eq

) can be written as:

eq1

j

−

=

ω

LT

⋅

and where:

is the admittance of the series net

1

V

1

)

1

1

k

ω

Lamp

(

2

R

LAMP

C1K

2

ω

jCR

2

Cj

2

−

Lampeq

2

21

Cjk+⋅

Lamp

iIi

22

−==

+⋅+

1

)1(

ω

2

kC

1

C

⋅−=ω

R

2

N

v

N

I

(

1

jR

C

⋅

ω

v

2

1

21

===

2

1

I

)

2

)

1

(6.17)

(6.18)

(6.19)

(6.20)

2

ω

Cjk

1

)1(

ω

jCR

+

Lamp

(6.21)

(6.22)

(6.23)

9/35

AN1731 - APPLICATION NOTE

k

∆

Considering

R

Lamp

2

(6.24)

negligible compared to

(6.25)

deriving Y

compared to the pulsation and equal to zero, it is possible to achieve the frequency that

eq1

maximizes, the Y

1

2

kC⋅ω

1

(such frequency is the resonance frequency of the application during the lamps on

eq1

state).

2

≅ω

1

2

)(

CkCLT +

1

2

1

2

)(2

CkCLTf+⋅≅π

1

2

(6.26)

(6.27)

When the board is powered, R1 and R2 enable Q1 and Q2 and the lamp turns on. After the lamp start-up,
during the Q2 on state, the current flows through L1, the half primary winding transformer T1 and Q2, and
it increases as:

*Ltv

1

L

tg

=α

1

(6.28)

angular coefficient but, after a while, the current curves and it becomes flat. However, in the permanent
state, even if the current oscillates around its average value, there is a ripple of this same value. The
current ripple decreases increasing the inductance value L1.

Figure 11 shows the PUSH-PULL current FED converter schematic circuit with the theoretical waveform
of ‘I’.

Figure 11: PUSH-PULL Current FED Converter Schematic Circuit with the Theoretical Waveform

10/35

AN1731 - APPLICATION NOTE

After the strike, ‘I’ generates the current i2 and, at the beginning, the same i2 can be written as:

v

2

i

=

2

R

Lamp

(6.29)

because the capacitor C1 is discharged. Immediately after, C1 gets charged and i2 decreases to zero
until the voltage across C1 reaches the maximum value. A this time, the current i2 inverts its direction and
the capacitors C1 start discharging until the charge inside it becomes zero and the current i2 reaches its
maximum negative value. Furthermore, when i2 inverts itself, also the voltage across the third winding
inverts its direction so that Q2 switches off and Q1 switches on and ‘I’ flows through the other half primary
winding of the transformer T1 (see fig. 11).

Figure 12: PUSH-PULL Current FED Converter Schematic Circuit with vc1, vc2, i2, and v

t1b2

Theoretical Waveforms

In the above graph, vc2 is the voltage between the vices of the Q1 and Q2 collectors. The maximum value
of such voltage is twice v1, where v1 is the voltage between the vices of the central point of the primary
winding of T1 and the reference. The voltage v1 is a half positive sine wave and this reaches the
maximum value when Q1 or Q2 are on, while it drops to zero during the turn-off and the turn-on of the
same transistors (see fig. 13).

11/35