For example, when inspecting laser
welds on a highly reflective material
such as copper, laser welding increases
surface roughness in localized areas,
raising the emissivity of the welded
region. A multi-flash strategy [13] can
help mitigate reflections, provided suf-
ficient thermal energy is absorbed,
reaches the inspection depth of interest,
and is emitted. This is therefore only
practical for very thin welds with a rough
textured surface. An alternative approach
for inspecting the subsurface condition
of a weld is laser ultrasound [14–16].
Laser ultrasound is a powerful tool
however, the cycle times required are
often not suitable for a manufacturing
environment. Furthermore, achieving
an adequate signal-to-noise ratio may
require operating in the ablation regime,
which can result in marking the part.
Machine vision is another commonly
used technique for noncontact weld
inspection. While it is an effective, rapid
inspection tool, its limitations include
its inability to assess subsurface condi-
tions. Even so, many inferences about
joint quality can be drawn from surface
features [17], and machine vision can serve
as a powerful tool for weld inspection in
components such as laser welds found in
battery trays or electric motor stators.
Although 100% in-line inspection
is the ultimate goal, operational envi-
ronment constraints can render this
impossible. For this reason, audits
with offline nondestructive tools or
destructive testing still have their place.
Offline nondestructive tools (i.e., those
requiring cycle times not suitable for
in-line manufacturing) can be used,
with two common examples being
manual inspection or X-ray computed
tomography (CT). Manual inspection
may involve a straightforward visual
assessment by a trained operator or a
hands-on “screening” technique collo-
quially referred to as a “pick test,” where
an operator attempts to pry apart a weld.
The latter is, of course, destructive if the
weld is weak. X-ray direct radiography
(2D) or CT (3D) is used for inspecting
various assemblies such as battery dis-
connect units (BDUs) or power electron-
ics like inverters. X-ray can also be used
to inspect adhesive joints, such as struc-
tural adhesive in battery modules, to
determine whether they have achieved
wet-out (where “wet-out” refers to the
condition in which the adhesive achieves
complete contact with and uniformly
covers the surface to which it is applied.
See Figure 3). The effectiveness of X-ray
for this purpose depends on the module
geometry, location of the adhesive, and
wet-out requirements.
NON-PERMANENT CONNECTIONS
While the above examples focus on
permanent joints, there are also several
non-permanent joints in the vehicle that
require careful consideration during
inspections, such as bolted electrical
joints. Passive infrared thermography is
an effective method for inspecting bolted
joints in battery packs [18]. By energiz-
ing the pack electrically and using a
thermal camera, inspectors can identify
“hot spots” that indicate areas of high
resistance, often associated with faulty
joints. This technique enables detection
of potential issues, ensuring the integrity
and performance of the battery pack.
Bolted joints between the pack and
the motor inverter connect components
that span the vehicle and are designed
to allow for servicing and component
replacement. When considering the
joint’s electrical resistance, special con-
sideration must be given to the contact
surfaces of the busbar and the mating
cable lead. Both surfaces must be flat
and free of debris to ensure proper
contact. Additionally, the fastener torque
must be accurate. Traditional techniques
such as fastener torque monitoring
[19] can be conveniently applied since
fasteners must be torqued to specifica-
tion. This approach, however, does not
guarantee optimal mating contact at the
joint. Debris between surfaces or defor-
mation of one or more surfaces can still
lead to poor electrical performance. One
potential area for research is the appli-
cation of less-common production NDT
techniques such as laser ultrasonics [14],
where a change in fastener strain after
torquing can potentially be detected
through changes in travel time of the
acoustic signal. Moreover, as a noncon-
tact technique, it could prove suitable for
monitoring fastener torque during live
pack operation.
Ensuring the electrical integrity of
power and signal connectors in vehicles
is essential for reliable performance.
Apart from the common methods using
external sensors, signal conditioners,
and data acquisition hardware for non-
invasive testing, allowing for thorough
NDT TUTORIAL
|
ELECTRICVEHICLES
External structure
(transparent for
representation purposes)
Polymeric interface material
External structure
Unwetted region (e.g., void or delamination)
Figure 3. Schematic representation of imperfect wet-out conditions in which a localized unwetted region is present within the polymeric interface
material: (a) isometric view (b) cross-sectional view.
28
M AT E R I A L S E V A L U AT I O N • J A N U A R Y 2 0 2 6

evaluation without disassembling any
components, new methods are being
explored that leverage IoT (Internet
of Things) technology to continuously
monitor electrical system health in real
time. Using advanced data analytics,
machine learning algorithms can predict
potential failures by analyzing large
datasets with various sensor outputs and
test results.
Thermal Interface Materials
Thermal joints serve to either insulate or
conduct heat, depending on the appli-
cation [20]. Thermal interface materials
(TIMs) are crucial in EVs due to the need
to manage heat from battery packs and
power electronics, both of which are
central to the vehicle’s operation [21].
While TIMs also appear in ICE vehicles,
cooling systems that manage engine
heat (the primary heat source)—such as
radiators and coolant—play a more sig-
nificant role. ICE vehicles typically have
fewer high-power electronic components
compared to EVs, and most of these
components do not require intensive
thermal management.
For EVs, thermal joints predomi-
nantly take the form of mating surfaces
bonded with a paste. For example, a
“cold plate” serves as a heat exchanger
by circulating a coolant through internal
channels. The cold plate can then be
used as an effective thermal manage-
ment tool by coupling it to the compo-
nent(s) requiring cooling via a thermally
conductive paste or material. The quality
of the paste interface is often assessed
by verifying that wet-out conditions are
achieved (see Figure 3).
TIMs range from silicone‑based
greases and graphite‑enhanced pastes to
elastomeric or phase‑change gap pads
[22]. Low‑viscosity greases conform well
under compression but can pump out
over thermal cycles, while gap pads may
trap micro‑air pockets during assembly
[23]. Since thermal conductivity, elastic-
ity, and dielectric properties dictate how
voids or delamination affect heat flow,
they also determine the NDT method
used. For example, the use of flash ther-
mography with an insulating paste will
result in “hot spots” in regions of the
paste where the heat is trapped at the
surface, whereas a conductive paste will
propagate the heat. A similar principle
applies to ultrasound inspection, where
the paste may act either as an acoustic
attenuator or reflector, depending on its
properties. These material interactions
must be carefully considered to ensure
accurate and reliable inspection results.
TIMs may also serve to insulate
temperature and have a dual function
of mechanically securing components.
Polymeric interface materials in battery
modules are one such example, used
as potting material in battery modules.
This material not only serves as thermal
insulation between cells—important
for thermal runaway suppression—but
also helps secure the cells in position.
Potting materials therefore must satisfy
requirements related to allowable void
distribution and size. This can be a chal-
lenging inspection task depending on the
geometry of the structure to be inspected
and the region of interest. X-ray CT is
one offline tool that can be used for both
wet-out and void inspection however,
its effectiveness depends on the overall
geometry, the required resolution, and
the region of interest. Furthermore, data
analysis often requires manual inter-
pretation, since the exact CT slice plane
relative to the TIM layer is not always
reliably repeatable. When X-ray CT proves
unsuitable for the part geometry, destruc-
tive teardowns can be performed in an
audit fashion to ensure the manufactur-
ing process minimizes voids appropri-
ately. This inspection application remains
an active area of ongoing research.
Power electronics also use TIMs. NDT
for the thermal performance of TIMs in
power electronics involves techniques
such as high-resolution infrared thermal
imaging and transient thermal analysis
using thermal conductivity and resis-
tance measurements to understand TIM
degradation. Scanning acoustic micros-
copy [24] and X-ray CT are also used to
detect the discontinuities and voids in
TIM paste. Acoustic microscopy (e.g., fre-
quencies ≥50 MHz) can also be used to
image power electronic chips if the leads
are properly isolated, since it requires
submersion in a tank of liquid (coupling
fluid). Similar to CT, this is an audit-type
technique due to its longer cycle time.
Hermetic Sealing
Leak checking components that have
hermetic seals is important to ensure
that enclosures meet the required level
of waterproofing [25]. The consequences
of water intrusion into high-voltage
components—such as battery packs and
power electronics housings—can range
from slow degradation in the best case
to rapid performance loss in the worst
case. In pack enclosures, for instance,
leaks that allow water into the battery
pack can lead to high-resistance shorting
of the electrodes on a battery cell, result-
ing in slow discharge or, more severely,
an ionization path between high-voltage
terminals that ultimately renders the
battery inoperable [26]. Inspection
must therefore be performed on joints
that provide hermetic sealing, such as
welded metal joints, composite compo-
nent interfaces, and long multi-material
bolted connections that may also
include adhesives. With joints requiring
inspection coming from a broad set of
processes, the need arises for a multi-
stage approach to leak identification,
localization, and root-cause analysis in
real-world production settings.
The fastest and often first method of
identifying a leaking part is to use global
detection techniques, the most common
being pressure-drop testing [27]. Pressure-
drop testing consists of applying a small
positive pressure (~5 psi) and observing
the pressure relaxation over minute time
scales. This can be done either with an
absolute pressure measurement scheme
for higher acceptable leak rates or with
differential pressure measurement, in
which the pressure drop of a good part is
compared to the part under test for more
sensitive applications. Global leak testing
works especially well for larger parts like
battery packs because it is simple to set
up and allows many joints to be tested
at once.
For smaller parts with fewer joints—
or for larger parts where a leak has been
previously identified—several methods
can be used to identify a leaking part
and localize the source of the leak for
J A N U A R Y 2 0 2 6 • M AT E R I A L S E V A L U AT I O N 29