High-Speed Electronic Current Limiter Protection
Bourns® Transient Blocking Unit (TBU)
The Surge Threat
Surge protection is the process of protecting
electronic systems or equipment from voltages
and currents which are outside their safe operating
limits. ese surge voltages and currents can be
generated by short circuits, lightning or faults from
a power system and usually enter the electronic
system along inter-equipment wiring. e surges
may be galvanically coupled into the system as in
the case of a direct lightning strike, through an
inadvertent connection of the power system to
the wiring, or as a result of an earth potential rise.
ey may be capacitively coupled into the system
which may occur when a data system is used in the
vicinity of a high voltage power line. ey may be
inductively coupled into the system as may occur
if the wiring is run in parallel with large currents
running in a power circuit feeding a high power
motor.
e size and waveform of the transients which
can occur within a system are many and varied. In
general, however, the following will hold true:
1. Lightning - Although direct strike lightning
current can potentially generate transients in
the millions of volts and tens of thousands of
amps, electronic equipment is rarely exposed to
surges of this magnitude. e greatest exposure
in telecommunication systems is through inter-
connecting telecommunication transmission lines.
ese lines can only carry voltages up to 5 kV and
currents of the order of 1 kA. erefore, for the
vast majority of instances where the chance of a
lightning strike directly to the equipment is low,
5 kV and 1 kA is the limit of the direct strike or
inductively generated surges.
2. Power Induction - Although power induction
voltages can be quite high in voltage and current,
they are oen limited in duration. ese voltages
are caused by faults on the power system which
couple into the system (usually inductively as a
consequence of the surge causing a very large
fault current). In virtually all modern power
transmission systems, these faults are very quickly
terminated by circuit breaker and re-closer
equipment. is can occur in as short as a couple
of cycles of power frequency voltage and rarely
takes longer than a second. ese transients are
typically modeled as a 600 V
rms waveform lasting up
to a second.
3. Power Cross - Alternatively, power cross
voltages are low voltage events but the exposure
can occur for very long durations. ey are oen
caused by maintenance error or cabling faults and
can result in moderate currents (<25 A) owing for
a long period of time (15 minutes, for example).
ey are predominately at mains power supply
voltage levels (100-240 Vrms).
4. Earth Potential Rise (EPR) - EPR can be
categorized in two forms: 1) as a result of power
system faults and 2) lightning discharges. In
normal industry, where fault currents from the
power system are limited in magnitude by fuses
and circuit breakers, power system EPR is not
usually a considerable risk. EPR only becomes
a signicant risk when power earthing systems
are signicantly below standard or where high
power transmission systems are used such as
at power generation and distribution facilities,
within the high power industry, and in the vicinity
BACKGROUND
High-Speed Electronic Current Limiter
Transient Blocking Unit Introduction to TBU Protection
2
of electrical traction systems (electric rail). In
such circumstances, this type of surge needs to
be carefully managed and expert predictions
need to be made of the risk and size of events.
Lightning EPR can only result from a direct strike
to the building housing the equipment or in its
immediate vicinity. Such events are uncommon,
unless the installation is particularly vulnerable
due to location or extreme height (e.g. cellular
phone base-station antennae). e equipment
exposure as a result of EPR can be very high, and
at high earth resistance locations, may become a
signicant portion of the lightning current.
5. All other forms of transients tend to be lower
energy forms not posing any additional risk to
equipment if protection has been suitably designed
for the events detailed above.
Conventional Protection
ere are two primary methods for implementing
protection against surge threats, namely, blocking
the surge or diverting (shunting) the surge. Nearly
all conventional protection schemes used today
are based on shunting architectures because of two
fundamental assumptions:
1) Inexpensive devices exist which shunt faults of
hundreds of amperes and thousands of volts.
2) Inexpensive devices which block faults of
hundreds of amperes and thousands of volts are
rare, if they exist at all.
erefore, conventional protection relies on the
primary form of protection being shunt devices,
which divert current. However, shunt devices
have limitations which aect their ability to
protect electronic equipment by themselves.
Electronic equipment can be damaged by voltages
in the tens of volts and currents in the hundreds
of milliamperes if they persist for any signicant
length of time (more than microseconds).
erefore, equipment protection requires a
shunt protector that: 1) can react fast enough to
allow less than these values from reaching the
equipment, and 2) can subsequently short circuit
and protect the interface from a surge which may
peak on the order of 1000 A. Considering that the
surge may develop in a few microseconds, this is
quite a dicult task.
Some primary protective devices such as
semiconductor-based devices are fast enough
to react in time; however these devices tend to
have limited current handling capability. Also,
semiconductor devices rated for the primary
protection task tend to capacitively load a circuit
(due to their physically large size) resulting in
bandwidth limitations. Non-semiconductor
surge protectors, such as the Gas Discharge Tube
(GDT), do not capacitively load circuits and can
handle very large currents (tens of kilo-amperes);
however, these devices are slower to react and may
not keep the voltages suciently low to provide
successful protection by themselves. erefore,
conventional protection must be based on a
number of stages of such devices. ese stages
typically start with a GDT as the primary protector
for its current handling capability, followed by a
semiconductor thyristor protector for speed – the
secondary protector.
BACKGROUND
High-Speed Electronic Current Limiter
Transient Blocking Unit Introduction to TBU Protection
3
Coordination Protection
When GDT and thyristor shunt protectors are
used as primary and secondary protectors,
the protection coordination between them is
complicated in practice. When a surge event
occurs, the fast secondary protector will act to
limit voltage within the system rst (due to its
speed). Oen this protector will be rated to keep
the circuit voltages quite low in order to protect
the equipment. us, its action can prevent the
high energy primary protector (GDT), which
requires a higher voltage to operate, from working.
In this circumstance, damage is likely to occur to
the secondary protector before the GDT operates.
is problem is solved by the complex process
of inter-stage coordination. Coordination is
the process of placing impedance between the
primary and secondary protectors to ensure that
sucient voltage is generated across the primary
protector, resulting from current owing in
the secondary protector, to trigger the primary
device. Coordination is engineered properly
when the primary protector operates aer the
secondary protector operates, yet before the
secondary protector is damaged. e coordinating
impedance can be resistive, capacitive, inductive,
non-linear or a combination of all of these; proper
selection is critical.
Large resistance is the easiest choice to ensure
that only a small current in the secondary
device causes a signicant voltage across the
primary device causing it to operate. However,
large resistance introduces considerable loss
within the data transmission path which is oen
unacceptable. Capacitance and inductance are also
useful, but these impedances are frequency related
and so circuits coordinated with such will work
only for a band of surge frequencies. Non-linear
resistance can also be used to create dierent
coordinating arrangements based on the duration
of the surge - low level surges for example
which last a long time causing the coordinating
impedance to change to a high resistance state
choking further current ow and triggering the
primary protector (the basis of operation of the
PTC). A typical conventional POTS (Plain Old
Telephone System) protection design with inter-
stage protection is provided in Figure 1.
Figure 1. Conventional Protection -
Circled parts indicating the
secondary protection and the
coordinating impedances.
BACKGROUND
Interface
GDT Fuse Resistor Thyristor Diodes
High-Speed Electronic Current Limiter
Transient Blocking Unit Introduction to TBU Protection
4
In reality though, the process of selecting the
best coordinating arrangement requires an in
depth understanding of:
1. e surge threat;
2. e performance of all the protective
components’ shunting and coordinating
(voltage/current/power/energy/time)
characteristics in addition to the conditions
under which they trigger or not;
3. e interfaces resistibility to voltages and
currents (voltage/current/time);
4. e transmission system specications
(bandwidth, maximum allowable loss); and
5. e interaction between multi-stage
protection circuits.
And therein is the problem. e time
and frequency dependent aspects of the
coordination problem make these circuits
dicult to design with any degree of certainty.
Consequently, testing must be used to assess
the performance of a protection design within
a specic system - a requirement which has
spurred the vast array of testing and design
standards within the marketplace. In addition,
virtually all completed designs have some
inherent weaknesses – e.g. a weakness to sneak
currents, a weakness to high energy surges, or
a weakness to surges of a particular frequency
or at a particular rate of repetition - and these
weaknesses can cause eld failures, incurrence
of replacement costs and lack of reliability.
BACKGROUND
High-Speed Electronic Current Limiter
Transient Blocking Unit Introduction to TBU Protection
5
Recall that secondary protection is only required
to prevent the let-through energy of the primary
protector (the energy of the surge which gets past
the primary protector) from damaging the load.
By denition, the peak open circuit voltage of the
let-through energy past the primary protector
must be smaller than that of the initiating surge
– otherwise by denition, the primary protector
is providing no protection at all. erefore, a
secondary protector which blocks rather than
diverts this lower, manageable and largely
predictable voltage would be very eective.
e requirements for this ideal blocking device are
self-evident:
1. As the device needs to block the let-through
energy of the primary protector, it must be a series
component (in series with the transmission line),
located just aer the primary protector. As a series
component, the device will react to the current
through the device rather than voltage across the
interface.
2. is series device should have a predictable,
stable and low trigger current (current at which
the device changes between its conductive
and non-conductive state) to provide eective
protection for sensitive downstream equipment.
3. It should be very fast acting (less than 10
nanoseconds) to protect equipment from surges
which rise to 5 kV in a microsecond - as do direct
lightning strikes or lightning EPR;
4. It should have low impedance (resistive,
capacitive and inductive) so that it does not eect
normal circuit operation;
5. In the blocking mode, it should have very high
impedance so that it does not dissipate signicant
energy during long duration surges;
6. It should reset aer the surge to reinstate the
system and continue to allow normal system
operation.
In addition, for practical and economic reasons
it should be small in size and low in cost. e
Transient Blocking Unit (TBU) protector meets
these requirements.
IDEAL PROTECTION
High-Speed Electronic Current Limiter
Transient Blocking Unit Introduction to TBU Protection
6
General
TBU device protection takes a dierent approach
to that of conventional protection. It is still
based broadly on the assumption that a primary
shunt protector is needed as surge currents are
more manageable than surge voltages. However,
it eectively replaces the need for secondary
protection and coordination considerations, with
one component. e TBU device was developed
to meet the requirements of the ideal protector
described on the previous page. In order to
understand the operation, it is useful to picture the
TBU device as having a current limiting functional
block and a voltage disconnect functional block.
e TBU device responds to both overcurrent and
overvoltage faults as described in the following
paragraphs.
Overcurrent Faults
A short circuit event occurring at time 1 (see
Figure 4), raises the current to the current limiting
level of Iout time 2 (~10 nanoseconds). At this
point, the voltage disconnect portion of the circuit
operates and by time 3 (~1 microsecond), the
load is disconnected from the surge. During the
remainder of the surge (time 4), the TBU device
remains in the protected state of very low current
and voltage at the load.
HOW THE TBU DEVICE WORKS
LOAD
S
TBU
CURRENT
LIMIT
VOLTAGE
DISCONNECT
SS
CURRENT
LIMIT
VOLTAGE
DISCONNECT
CURRENT
LIMIT
CURRENT
LIMITVOLTAGE
DISCONNECT
Figure 3. Overcurrent fault diagram
LOAD
S
TBU
CURRENT
LIMIT
VOLTAGE
DISCONNECT
SS
CURRENT
LIMIT
VOLTAGE
DISCONNECT
CURRENT
LIMIT
CURRENT
SHORT
LIMITVOLTAGE
DISCONNECT
Figure 4. TBU device reaction to an
overcurrent fault
1
2
3
4
I
out
I
op
I
leak
V
load
V
source
time
current voltage
nanosecond
reaction time
~1µsec
Figure 2. TBU device functional block diagram
High-Speed Electronic Current Limiter
Transient Blocking Unit Introduction to TBU Protection
7
Overvoltage Faults
A lightning or power cross event at time 1
(see gure 6), raises the voltage and current
until the current limiting portion of the
circuit limits current to the level of Iout time 2
(~10 nanoseconds). At this point, the voltage
disconnect portion of the circuit operates and by
time 3 (~ 1 microsecond), the load is disconnected
from the surge. During the remainder of the surge
(time 4), the TBU device remains in the protected
state of very low current and voltage at the load.
Figure 5. Overvoltage fault diagram
Figure 6. TBU device reaction to an
overvoltage fault
HOW THE TBU DEVICE WORKS
1
2
4
3
time
current voltage
nanosecond
reaction time
I
leak
V
load
V
source
~1µsec
I
out
I
op
LOAD
S
TBU
CURRENT
LIMIT
VOLTAGE
DISCONNECT
SS
CURRENT
LIMIT
VOLTAGE
DISCONNECT
CURRENT
LIMIT
CURRENT
SURGE
LIMITVOLTAGE
DISCONNECT
High-Speed Electronic Current Limiter
Transient Blocking Unit Introduction to TBU Protection
8
Impulsive Surge Test Results
e following test results show the TBU devices
response to a fast-rising lightning impulse.
e test was done using a 10/1000 μs impulse
generator as a voltage source with the peak
voltage set to 1200 V. e test was conducted
without shunt devices (i.e. with no GDT, thyristor
or MOV).
Figure 8 shows the input waveform on the lower
trace in red (voltage on the TBU device input,
V
A) and the voltage at the TBU device output on
the upper trace in blue (VB). e TBU device
current limited to ~280 ma (i.e. 28 V across the
100 Ω resistor) and then disconnected the load
throughout the remainder of the surge. e peak
voltage across the TBU device was ~1200 V, just
aer it triggered to protect the load. e device
automatically reset aer the surge ended.
AC Overvoltage
e following tests show TBU device performance
against an AC overvoltage.
At the onset of the surge, the TBU device reacted
to the increasing current and triggered to protect
the load. e TBU device remained in the
protected state throughout the remainder of the
high voltage cycle. At each surge zero-crossing,
the TBU device reset in a microsecond, and
then retriggered to protect the load for each of
the remaining cycles of the test. Once the fault
was removed, the device reset to its normal, low
impedance state.
TBU DEVICE PERFORMANCE
Figure 8. Impulse test of TBU device
Figure 9. AC overvoltage test circuit
Surge
1200 V
TBU triggers at 28 V
V
A
V
B
3.50
Input A
100 µs/Div-204 µs
3.00
2.50
2.00
1.50 kV
1.00
0.50
0.00
-0.50
100
50
0
-50
-100 V
-150
-200
-250
-300
Figure 7. Lightning impulse test circuit
LOAD
100 ohms
Surge –1200 V, 10/1000 μsec
AB
TBU
S
Figure 10. TBU device blocking a 600 Vrms
power cross
LOAD
100 ohms
Voltage – 600 V
ac
AB
TBU
S
High-Speed Electronic Current Limiter
Transient Blocking Unit Introduction to TBU Protection
9
For more information on TBU devices
and other circuit protection products from
Bourns, please visit
www.bourns.com
COPYRIGHT© 2009 • BOURNS, INC. 03/09 e/FU0936
“TBU” is a trademark of Bourns, Inc.
“Bourns” is a registered trademark of Bourns, Inc. in the U.S. and other countries.
Americas: Tel +1-951 781-5500
Fax +1-951 781-5700
Europe: Tel +41-(0)41 768 55 55
Fax +41-(0)41 768 55 10
Asia-Pacific: Tel +886-2 256 241 17
Fax +886-2 256 241 16
Blocking, Not Shunting
In designing circuit protection, conventional
wisdom has held that blocking devices that can
cost-eectively block real world surges do not
exist. As a result, most circuit protection schemes
today rely on multiple shunt devices to protect
against surges. However, the currents that can
occur in such surges can be signicant (hundreds
of amps) and they can last a very long time (in
the order of one second to hours). is high
current coupled with signicant duration places
a very large stress on circuits protected by all
types of shunt protective devices including GDTs,
MOVs and thyristors.
However, the Transient Blocking Unit (TBU™)
Electronic Current Limiter has proven the
conventional wisdom to be outdated. As a
consequence, it is now possible to eectively
block surges and prevent sensitive load
electronics from experiencing high levels of
energy during surge events.
Conclusion
e TBU device provides blocking protection
for both power cross and lightning. Benets
of the TBU device include overvoltage and
overcurrent protection in one device, extremely
high speed performance, high blocking voltages
and currents, precise output current and voltage
limiting, very high bandwidth, and a small size.
ese advantages result in a protection device
which exceeds Telcordia GR-1089 and ITU
K.20/K.21 requirements, provides automatic
protection coordination, and is GHz data rate
compatible all within a minimum of printed
circuit board area.
e TBU protector provides the circuit protection
design community with a simple, superior
protection device.
SUMMARY