You are on page 1of 62

Komal Charan Garlapati

ADVANCED TOPICS OF SEMICONDUCTOR


19th March 2017 DEVICES
1

Reliability

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


2

Reliability of Semiconductor Devices


Reliability is the probability a semiconductor device will perform its specified function for a
specified time period under specified environmental conditions.
Bathtub Curve: The bathtub curve is typically used as a visual model to illustrate the
three key periods of product failure rate.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


3

Reliability of Semiconductor Devices


Early Failure is defined as a failure that occurs at an early stage after operation of
semiconductor devices is initiated. In addition, the early failure rate tends to decrease
with the passage of time. Latent failures that were not removed in the screening process
are included in early failures and become manifest within a short time under stresses
such as temperature and voltage after operation starts. In the case of semiconductor
devices, most early failures are due to defects that form during the production process
(for example, microdust adhering to wafers) and defective materials.
Random failures occur once devices having latent defects have already failed and been
removed. In this period, the remaining high-quality devices operate stably. The failures
that occur during this period can usually be attributed to randomly occurring excessive
stress, such as power surges, and software errors. This group also includes devices
susceptible to remains of early failures.
Wear-out failures occur due to the inherent lifetime of the device due to wear and fatigue.
When a device enters the wear-out period, the failure rate tends to increase rapidly. This is
influenced by the usage conditions.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


4

Reliability of Semiconductor Devices

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


5

Calculating Reliability:
Two approaches to establishing a reliability level is to evaluate either the probability of
survival or the probability of failure. Either method is equally effective, but the most
common method is to calculate the probability of failure or Rate of Failure ().

The values most commonly used when calculating the level of reliability are FIT (Failures
in Time) and MTTF (Mean Time to Failure) or MTBF (Mean Time between Failures)
depending on type of component or system being evaluated.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


6

Calculating Reliability:
Accelerated Testing:
Stresses are used to accelerate failure. For example, temperature cycling is used to
accelerate mechanical failure.
Temperature Acceleration
Voltage Acceleration
Current Acceleration
Humidity Acceleration

Temperature Acceleration can be used with many physical or chemical failure mechanisms.
These mechanisms are often described by the Arrhenius Equation:
t = to exp [Ea/KT]
where R is the Response, Ro is the pre exponential term, Ea is the activation energy in
electron volts, K is Boltzman constant (8.6e-5 eV/K) and T is temperature in degrees
absolute.ms may be accelerated by different amounts for the same stress.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


7

Example to find the Activation Energy.


Example: 10% of metal lines fail due to electromigration failures in 200 hours at 175 C, in
32 hours at 200 C and in 4 hours at 250 C. Find the activation energy. (Note: K=8.6
105 eV/K)

Time for 10% failures = to exp[Ea/KT]

200 = to exp (Ea/K(273+175)) (1)


32 = to exp (Ea/K(273+200)) (2)
Solving (1) and (2)

Ea = K ln (200/32)/((1/448 - 1/473))
and
Ea = 1.34 eV

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


8

Calculating Reliability:
Failure Rate () is calculated by dividing the total number of failures or rejects by the
cumulative time of operation. In the HTOL model, the cumulative time of operation is
referred to as Equivalent Device Hours (EDH):

EDH=

Where D= Number of Devices Tested.


H= Test Hours per Device.
= Acceleration Factor derived from the Arrhenius equation.

The Failure Rate () per hour is shown by


= =

Where r = number of failures or rejects.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


9

Calculating Reliability:

Acceleration Factor (Af) is the test time multiplier derived from the Arrhenius equation.
This equation calculates the time acceleration value that results from operating a device at
an elevated temperature.
1 1

=
Where Ea = Activation Energy (eV) of the failure mode.
k (Boltzmann's Constant) = 8.617 x 10-5 eV/K.
Tuse = Use Temperature in Kelvin.
Ttest = Test Temperature in Kelvin.
K (degrees Kelvin) = 273 + C.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


10

Calculating Reliability:

Activation energy (Ea) is an empirical value that is the minimum energy required to
initiate a specific type of failure mode that can occur within a technology type.
Oxide defects, bulk silicon defects, mask defects, electro-migration and contamination are
some examples of such failure modes, each with an unique associated Ea.
In lieu of using empirical data for each individual failure mode, it is generally accepted that
a standard single value of Ea = 0.7 eV provides a reasonably accurate average Ea value
for diode type semiconductors.

FIT (Failures in Time) is a standard industry value defined as the Failure Rate () per
billion hours.
= = 109

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


11

Calculating Reliability:

MTTF (Mean Time to Failure or ) is another standard industry value which provides the
average time to failure of Non-repairable Items such as light bulbs and diodes or
unserviceable systems such as satellites or other unmanned space craft. For items with
long life expectancies, it is often a more useful to report MTTF in years rather than hours.

1
=

1
=
24365
MTBF is the average time between successive failures. MTBF is used for repairable
systems. It is truly a Mean Up-time Between Failures since it does not include the time to
repair.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


12

Electromagnetic Compatibility(EMC)

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


13

Need for EMC


Electromagnetic compatibility (EMC) requires that each system perform its intended
function in the electromagnetic environment, and that the operation of a given system
does not degrade the performance of other systems.

An electronic device is said to be electromagnetically compatible with other electronic


devices if it satisfies these three criteria:
- It does not interfere with the regular operational function of other systems.
- It is not susceptible to emissions from the other systems.
- It does not interfere with its own operation.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


14

Assurance
Assurance: EMC Assurance depends on four areas.
- Radiated Susceptibility: It is the immunity of the system to electromagnetic disturbance
radiated through free space.
- Radiated Emissions: It is the emissions of broadband or narrowband electromagnetic
disturbance radiated from the system to the outside environment.
- Conducted Susceptibility: Conducted Susceptibility is concerned with the ability of an
electronic circuit, or a system to operate acceptably when subjected to radio frequency voltage or
current on interconnecting conductors.
- Conducted Emissions: Conducted emissions concern with the propagation of electromagnetic
disturbance generated by an electronic/electrical device into the other electrical systems through
the interconnecting wires.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


15

Need for Electromagnetic Compatibility


Undisturbed continuous operation of the equipment has always been the main goal for
design engineers.
The equipment should not disturb the others while performing its normal operations.
Many interfering systems operating in the same environment. (Ex: Radio Communication
equipment).
The electromagnetic compatibility gains importance day by day as the words such as
safety-related, high-reliability, mission-critical comes forward.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


16

EMC Design features


EMC is concerned with the generation, transmission and reception of electromagnetic
energy.
The source produces the emission energy which is transferred to the receiver by means
of a coupling path. This transfer of electromagnetic energy usually occurs via unintended
coupling paths. This energy is said to be interference if it adversely affects the operation
of the receiver.
It is now apparent that there are three ways to prevent interference:
- 1. Suppress the emission at its source.
- 2. Reduce or eliminate the effectiveness of the coupling path.
- 3. Decrease the susceptibility of the receiver.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


17

Approach for achieving EMC


There are mainly two approaches in achieving the EMC. They are:
- 1. Crisis approach.
- 2. System approach.
Crisis approach: It is often referred as Band-Aid approach. In this approach, the EMC
is totally disregarded until the design of the product is finished. It makes the EMC
solutions complex and expensive which results in undesirable add-ons.

Systems approach: In this approach EMC is considered throughout the design. EMC
becomes the integral part of the electrical and mechanical design. EMC is designed into
the product and it is more cost effective.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


18

Electromagnetic Interference
Electromagnetic Interference (EMI) or Radiofrequency Interference (RFI) can be
described as the undesirable effect due to high frequency electromagnetic emission or
disturbance from a source due to its operation regardless whether intentional or not upon
another piece of operating equipment.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


19

Sources of EMI
Natural Sources:
- Electrostatic discharge(ESD)
Electrostatic discharge is defined as the transfer of charge between bodies at different electrical potentials.

Electrostatic discharge can change the electrical characteristics of a semiconductor device, degrading or
destroying it. Electrostatic discharge also may upset the normal operation of an electronic system, causing
equipment malfunction or failure.

ESD Discharge:

An ESD event can occur when any charged conductor (including the human body) discharges to an ESDS
device. The most common cause of electrostatic damage is the direct transfer of electrostatic charge from
the human body or a charged material to the ESDS device.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


20

ESD Levels for a human body


Human body can charge and the potential varies depending on the work doing and the environmental
conditions. Table shows the electrostatic potential developed by the human body at different conditions.
ESD Discharge:

Means of static generations Electrostatic potential with 10-20 % Electrostatic potential with 65-90%
Relative humidity relative humidity

Walking across a carpet 35000 V 1500 V


Walking on a vinyl floor 12000 V 250 V
Picking up a polythene bag 20000 V 1200 V

Getting up from a polyurethane foam 18000 V 1500 V


chair

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


21

Manmade Sources
Intentional Sources:
- radio communications, Wi-Fi, mobile networks, satellites etc., in which the high frequency signals
are intentionally used for communication.

Unintentional Sources:
- computers, electrical drives, transmission lines, fax machines etc., where the main purpose
doesnt require any use of radio frequency signals but they come as the byproducts of the
applications.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


22

EMI Coupling Modes


Interference is transmitted from source to the susceptible equipment. There are 4 types of
coupling mechanisms. They are:
- 1. Galvanic Coupling.
- 2. Inductive coupling.
- 3. Capacitive coupling.
- 4. Radiative coupling.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


23

Galvanic Coupling
This is also called as conductive coupling and it occurs through the conduction of high
frequency currents through conductors. And main cause of this type of EMI coupling
occurs when many circuits share a common source with common impedance or when
many circuits are grounded through a common ground.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


24

Inductive Coupling
Inductive coupling is caused by common inductances between a source of interferences
and an interfered circuit. Intensity variations in any cable modifies the magnetic field
distribution around it, which in turn, the field variation induces electromotive forces in
close circuits.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


25

Capacitive Coupling
Two neighboring conductors have a parasitic
capacitance. If the voltage in one conductor
changes, an interference current flows via the
parasitic capacitance Cp to the neighboring
conductor and causes change an interference
voltage in the measuring resistance.
The following factors influence the interference
currents.
Input resistance: The greater the input resistance,
the greater the interference voltage caused by the
interference current.
Spacing of the conductors: The greater the
spacing the smaller the parasitic capacitance and
smaller the interference currents. The parasitic
capacitance increases with the smaller conductor
spacing and with the length over which the
conductors lies in parallel

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


26

Radiative Coupling
Interference in a conductor can also be transmitted into a circuit by electromagnetic
radiation. The conductors and circuits acts as transmitting and receiving antennas for
the electrical or magnetic component of the field.
Signals are radiated in an increasing amount at higher frequencies and propagate
through space in the form of a wave. Higher the frequency of the produced signal, the
smaller the volume expansion of this wave (wavelength ). There is the following
relationship between the wavelength and signal frequency f:
=c/f
It is known from radio technology that the optimum radiation is achieved with the dipole of
length equal to /4. The conductor starts emitting the radiation components even if its
length is equal to /10.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


27

Radiative Coupling

Frequency Wavelength()

50Hz 6000km

100Hz 3000km

1kHz 300km

10kHz 30km

1MHz 300m

100MHz 3m

1GHz 30cm

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


28

Forms of EMI currents


EMI currents are decomposed into two auxiliary currents, referred to as the differential-
mode (DM) current, idm and the common-mode (CM) current icm. DM and CM currents
are shown in Figure. DM currents flow in two conductors connected in a loop while CM
currents flow in a conductor with the ground forming the return path

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


29

Typical EMI filters


The decomposition of EMI currents into CM and DM components is useful in determining
which elements in a general EMI filter are essential and which ones are expendable. This
can be explained by means of Figure which depicts a generic EMI filter topology.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


30

EMI Filters
The capacitors CDL and CDR are used to divert differential-mode EMI while CCL and
CCR are used to divert common-mode currents. The coupled inductors represented by L
and mutual inductance M form a common-mode choke. These, along with the inductor
Lgw increase the high-frequency impedance of the ground loop and thus reduce
common-mode currents. If it is known that DM currents are dominant in the total EMI, the
common-mode choke M and the capacitors CCL and CCR can be left out. Similarly, if it is
known that CM currents are dominant, then the capacitors CDL and CDR can be left out.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


31

Aero Standard (DO-160G)

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


32

EMI Limits: Conducted

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


33

EMI Limits Radiated

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


34

EMI Issues in Power Converters


Rapid changes in voltage and current of a switching power converter produce EMI noise.
The radiated EMI can usually be shielded by metal cabinets used for housing the power
converter
Conducted EMI is reduced by using filter to block its transmitting paths

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


Signal Spectra:
Time domain to frequency domain:

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


36

Sources of EMI in Motor Drive System


Parasitic Capacitances and Inductances:

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


37

DM and CM currents in Drive Applications


Common mode currents are formed because of the current flow through the parasitic
capacitances formed between lines and ground because of high dv/dt.
Differential mode currents are formed because of the current flow through the parasitic
capacitances between the phases.
The parasitic inductances of switches accounts for the voltage Ldi/dt which will again
increase common mode noise.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


38

EMI in converter circuits


dv/dt and di/dt effects:
The high frequency components because of large dv/dt and di/dt induces the current flow
in capacitors and voltage drop across the parasitic inductances.
The noise current magnitudes are proportional to the dv/dt and di/dt values at the
parasitics.

Heat sink role:


Heat sink is one of the main cause of the parasitic capacitances as the heat sink is
attached to the semiconductor switch with a thermal grease in between them.
If any dimension of the heat sink is between and /20, radiated EMI can be expected.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


39

Common Mode Noise Propagation


CM noise is generated from the different coupling paths:
1. Idg due to the dv/dt imposed across the parasitic capacitance (Cdg) between the switch
drain and the grounded heat sink.
2. Iwg due to the dv/dt imposed across the parasitic capacitance between the output phase
cable w and the earth (Cwg).
3. Ing due to the dv/dt imposed across the parasitic capacitance between the neutral point
of the motor windings and the motor frame (Cng).

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


40

DM Noise Propagation
DM noise is generated from the following coupling paths:
1. Ips due to the dv/dt imposed across the parasitic capacitance in parallel to the lower
switch (Cps).
2. Iuv due to the dv/dt imposed across the parasitic capacitance between the output phase
cables (Cuv).
3. Iug due to the dv/dt imposed across the series connection of line-to-ground parasitic
capacitance (Cug) and back to the phase w through the drain-to-ground parasitic
cacitance (Cwg).

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


41

Filter Configurations

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


42

Selection of filters
Filters are selected based on the parameter Insertion Loss(IL)
required.
IL= 20 log(V1/V2)
V1= Voltage appearing before application of Filter.
V2= Voltage appearing after the application of Filter.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


43

Impedance Mismatch Condition

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


44

Advisable filter configurations

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


45

Common Mode Chokes

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


46

Filter with Chokes

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


47

Common Mode Filter

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


48

Differential Mode Filter

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


49

Thermal Analysis

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


50

Need for thermal analysis


Designing a cost competitive power electronics system.
Over designing the system adds unnecessary cost and weight; under designing the
system may lead to overheating and even system failure.
The efficiency of a power converter depends on the switching and conduction power
losses of the semiconductors used in the system.
The thermal and power losses are interrelated and thermal analysis helps designer to
design optimum heat sink

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


51

Junction Temperature
The term junction temperature became commonplace in the early days of semiconductor
thermal analysis when bipolar transistors and rectifiers were the prominent power
technologies. Presently the term is reused for all power devices, including gate isolated
devices like power MOSFETs and IGBTs.
The rated maximum operating junction temperature Tvjmax is used to determine the
maximum allowable power dissipation of a continuously turned on IGBT (i.e. static
operation). For switching operation (also for the short period of time during turn-off of the
IGBT) it has to be ensured that the device safely operates under high dynamic stress,
short dynamic temperature transients and operational chip and module temperature
inhomogenities. Hence, the maximum calculated virtual junction temperature under
dynamic operation is limited to a value lower than Tvjmax.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


52

Power Losses at switching device


The total power loss in an semiconductor is composed of switching power losses and
conduction losses.
Ptotal = Pswitching + Pconduction
Switching power losses consists of turn on and turn off losses.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


53

Turn On loss
Eon: Turn-on switching energy. The energy dissipated during a single turn-on event.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


54

Turn Off loss:


Eoff: Turn-off switching energy. The energy dissipated during a single turn-off event.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


55

Losses in an IGBT
The instantaneous conduction losses are a combination of the collector current and the
collector-emitter voltage, Vce, which is dependent on the collector current, ic, its zero-
current collector-emitter voltage, Vceo, and its collector-emitter on-state resistance, rc as
follows:

The average conduction power loss in an IGBT is:

Switching power loss in IGBT:

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


56

Thermal Analysis (Basic Principles)


Analogy to Electrical system

Using temperatures and the rate of heat dissipation, a quantitative measure of heat transfer efficiency
across two locations of a thermal component can be expressed in terms R, defined as

=

where T is the temperature difference between the two locations. The unit of thermal resistance is in
C/W, indicating the temperature rise per unit rate of heat dissipation.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


57

Parameters
Q: Total power or rate of heat dissipation in Watts, represents the rate of heat dissipated
by the electronic component during operation. For the purpose of selecting a heat sink,
the maximum operating power dissipation is used.
Tja,max : Maximum junction temperature of the device in C. Allowable Tja values range
from 115C to as high as 150 C for some devices.
Tc: Case temperature of the device in C. Since the case temperature of a device
depends on the location of measurement, it usually represents the maximum local
temperature of the case.
Ts: sink temperature in C. Again, this represents the maximum temperature of a heat
sink at the location closest to the device.
Ta: Ambient air temperature in C.

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


58

Thermal Circuit with Heat sink


=
=


= =


= =

Here

= + + =

Required Heat Sink thermal resistance

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


59

Example for selecting the Heatsink


Calculate the maximum heat sink resistance for the safe operation of the IRF540
MOSFET which is dissipating 10 Watts. Ambient temperature is 25 C.

Required Heat Sink thermal resistance

= (175-25/10) 1 0.5
= 14 C/W

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


60

Questions?

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.


61

Thanks

Honeywell Confidential - 2017 by Honeywell International Inc. All rights reserved.

You might also like