high voltage between current carrying
and non-current carrying components of the equipment under test.
In order to ensure the electrical safety of any
medical device, there are several tests that must
be performed on each device before it can be
shipped. In most cases these tests are required in
order to comply with safety agency regulations.
This article deals with two of the most widely
performed production line safety tests commonly
required on medical devices.
The first and most prevalent test is the
Dielectric Voltage Withstand Test, commonly
known as a hipot test. Hipot testing is the
deliberate application of high voltage between
current carrying and non-current carrying
components of the equipment under test (EUT).
This test is designed to stress the insulation far
beyond what it will encounter in normal use. If
the insulation can withstand the higher voltage
for a given period of time, it is assumed that it
will be able to adequately protect medical
personnel and patients from electrical hazards at
normal operating voltage. This is the origin of
the term voltage withstand.
In addition to overstressing the insulation, the
test also has the ability to uncover material and
workmanship defects which result in conductor
spacings that are too close. The appearance of
humidity, dirt, vibration, shock, or contaminants
in normal use tends to close these small gaps and
allow current to flow. This creates a shock
hazard later if the defects are not found and
corrected at the factory. The most effective way
to locate these potential problems is to apply high
voltage across the conductors to verify that they
will not allow current to flow. As simple as the
description of this test may sound, it can have its
complexities and problems. Our intent is to
cover these problems as well as some of the new
technology that has been developed to solve
them.
Compliance agencies such as UL, CSA, IEC,
VDE and others specify a minimum amount of
high voltage to be applied to each device coming
off a production line. The test voltage is seldom
less than 1000 volts, and for some medical
devices the specified voltage can be in excess of
4000 volts. A rule of thumb that is often
employed by some agencies is to apply two times
the normal operating voltage plus 1000 volts.
Special consideration is given to the type of
service intended for the product. For example, it
is not uncommon to see medical devices that will
be used in patient connected applications
subjected to test voltages well in excess even of
the rule of thumb level.
The amount of time the high voltage has to be
applied is also specified in most standards. The
most common time duration for the application of
high voltage is either one second or one minute.
In order to move through the test as quickly as
possible most manufacturers select the option of
a 1 second test time with a slightly higher output
voltage. Maintaining the correct test voltage
becomes critical since the operator cannot
monitor what voltage was actually reached
during a 1-second test.
Unfortunately, there are uncontrollable factors
that can externally affect the output voltage of the hipot. A hipot transforms the input voltage of
115 or 230 volts AC to several thousand volts at
the output. Therefore, stability of the output
voltage is linked to the level of the input voltage.
Figure 1a: Line Variation vs. Hipot Output
Figure 1b: Load Variation vs. Hipot Output
in most cases, line voltage fluctuations are very
common.
This is particularly evident in manufacturing
environments where other machinery could be
running off the same input line circuit.
This uncontrollable variation in input voltage
can cause drastic changes in the output voltage.
In cases where the line voltage has dropped, the
output voltage can actually fall below agency
requirements, If line voltage increases, the output
voltage can actually reach higher voltage levels
possibly causing damage to the EUT. For
example, consider a product that has an
operating voltage of 120 volts. A compliance-
agency-required test voltage. using rule of thumb
calculations (i.e., 2X + 1000), would be 1240
volts. Let's assume that this voltage was set with
the line input voltage at 115 volts. This would
mean that once the output of the hipot was
adjusted it will produce approximately 10.78
volts of output for each volt of input. Now let's
assume that the line voltage has dropped for
whatever reason to 105 volts. The hipot is still
set to produce 10.78 volts for each volt of input
so the output voltage now drops to about 1132
volts. This, of course, means that any device
being tested while the input voltage is low is
actually being tested at 108 volts below the
required test voltage. (Figure 1a) shows the effect
that a variation in line voltage can have on the
output test voltage of a hipot. This is a key
reason why hipot instruments are now available
with regulation circuitry that maintains the
specified output voltage setting. This regulation
circuitry monitors the input voltage and
electronically makes adjustments to assure that
the preset voltage level is maintained.
Another factor that can greatly affect the
output voltage applied to the EUT, would be
loading conditions. Many manufacturers have a
standard procedure for setting the hipot voltage
while the EUT is connected. This ensures that
the proper test voltage will be reached when the
hipot tester is operating under a loaded condition.
Unfortunately, this setting-up operation becomes
a little more complicated in some manufacturing
environments where different products run down
the assembly area and into the testing area.
These different products could represent different
loads from the one that was used when the hipot
was originally set up. Since it is impractical to
expect the operator to monitor changes in the
hipot's output voltage during a one-second test,
we could again see tests either being performed
with more or less voltage than required by agency
specifications. (Figure 1b) shows the types of
changes in output voltage that could be seen if
the load is varied from 120 K2 to 2 M2.
Compliance issues have made it necessary to
have hipot instrumentation that can maintain
output voltage even when the load varies. The
only way to deal with this problem in the past was to have hipot instruments that could provide
extremely high current levels without collapsing
under load. In some cases, these instruments
could have output current capabilities in excess
of 100 milliamps. This approach may have
solved this problem but tended to create another
one. Instruments with an output current this high
created an
Figure 2: High Frequency Spikes Created By An Arc
unnecessary safety risk for the test equipment
operator. A load-regulated instrument can
electronically monitor the loading effect on the
hipot and compensate for these load variations to
maintain the preset output voltage. This
approach does not require the hipot to have
excess output current capability so it ensures
compliance with agency requirements while not
making any concessions to operator safety.
Up until now we have covered the correct
application of the high voltage. An equally
important aspect of hipot testing is the
determination of what is considered to be an
electrically unsafe product. The first and most
evident type of failure is an electrical break-down
or arcing between the current and non-current
carrying components of the medical device.
Failure detection systems that truly detect arcing
conditions monitor the high frequency energy
created by an arc. A filter in the hipot monitors
the duration of this high frequency and if it
exceeds the maximum allowable level for a long
enough time, an indication of an arc failure is
displayed. This technique allows for detection of
sporadic arcing which in many cases, is an
indication of an unsafe

electrical condition in the
EUT. Figure 2 shows the high frequency signal
that appears on the voltage sinewave when an
arcing condition occurs.
Another failure condition is the detection of
excessive leakage current. The maximum level
of acceptable leakage current is normally
specified by the compliance agency. Many
testing specifications allow only AC testing.
This can present a problem when testing highly
Figure 3a: Current Vector
Figure 3b: Sinewaves in Phase
Figure 3c: Sinewaves 90
°
°
Out of Phase
Figure 3d: Sinewaves Out of Phase
capacitive medical devices. The application of
AC test voltage to a capacitive item causes capacitive leakage current to flow. Such current
is often much greater than the real current that is
flowing due to resistive leakage. Many hipots
can only detect and display total leakage current
which is actually the vector sum of the resistive
and capacitive leakage current (Figure 3a).
Unless the real component of total leakage is
separated out, a doubling or more of the real
leakage current can go undetected. The
technology is available today to allow accurate
monitoring of real leakage current.
To understand how this works you must first
have an understanding of what effects a
capacitive and a resistive load can have on the
measured current sinewave. A pure resistive
load causes a current sinewave (Figure 3b) that is
in phase with the output voltage sinewave. On
the other hand a pure capacitive load will cause a
sinewave (Figure 3