Keithley 237

SECTION2
Operation
Crimped copper to copper connections can be used to
minimize these effects.
2.33 ELECTROMAGNETIC
INTERFERENCE (EMI)
The electromagnetic interference characteristics of the
Source Measure Unit comply
with the electromagnetic
compatibility (EMC) requirements of the European
Union (EU) directives as
denoted by the CE mark. How-
ever, it is still possible for sensitive measurements to be
affected by external sources. In these instances, special
precautions
may be required in the test setup.
Sources of EMI include:
• Radio
and TV broadcast transmitters.
• Communications transmitters, including cellular
phones
and handheld radios.
• Devices incorporating microprocessors
and high-
speed digital circuits.
• Impulse sources as in the case of arcing in high-
voltage environments.
EMI effects can be seen as unusually large offsets
or, in
the case of impulse sources,
sudden, erratic variations in
readings.
The most convenient
method to minimize the effects of
EMI is to use the FILTER feature of the Source Measure
Unit.
Up to 32 readings can be averaged with this filter.
A detrimental effect of filtering is
that overall Source
Measure Unit speed is decreased.
2-114
The instrument and measurement setup should be kept
as far away as possible from any EMI sources. Addi-
tional shielding of the instrument, experiment,
and test
leads will often reduce EMI to
an acceptable level. In
extreme cases, a specially constructed screen room
may
be necessary to sufficiently attenuate the troublesome
signal.
2.34 GROUND LOOPS
In general, optimum safety, shielding and stability are
obtained
by connecting OUTPUT LO to earth ground by
installing the ground link on the rear panel of the Source
Measure Unit. However,
ground loops that occur in
multiple-unit test set-ups can create error signals that
cause erratic
or erroneous performance. The configura-
tion
shown in Figure 2-80 shows a ground loop that is
created
by connecting OUTPUT LO to earth ground at
two points. A large ground current flowing in the loop
will encounter small resistances, either in the conduc-
tors,
or at the connecting points. This small resistance
results in voltage
drops that can affect performance.
To prevent ground loops, the test system should be con-
nected to
ground at only a single point, as shown in Fig-
ure 2-81. Note that only the first Source Measure Unit is
connected directly to
power line ground (via ground
link). Experimentation is the best way to determine an
acceptable arrangement. For semiconductor character-
ization tests,
best stability is usually obtained by con-
necting the Source Measure Unit
used to supply power
to the base (or gate) of the OUT to earth ground.