Biomedical Engineering Reference
In-Depth Information
changing EM fields and frequencies, they execute the
footprinting procedure in the selected area immediately
prior to fingerprinting the DUT. Once the data from the
latest footprint has been stored, place the DUT on
a nonmetallic stand located as far as possible from any
conductive material. The standard antenna(s) is then
mounted on a tripod and, in a horizontally polarized
mode, placed 1 meter from the front of the DUT. As in
the footprint procedure, a series of 20-MHz-wide
spectrum sweeps are performed and the results are
recorded. This procedure is repeated with the anten-
na(s) vertically polarized. When using tunable standard
antennas, adjust them to the correct resonant length for
the center of each 20-MHz window. Upon review of the
data collected, any emissions attributable to the DUT
should be rescanned. To increase detail in the frequency
range in which the emissions were observed, the spectrum
analyzer window is narrowed to a 200-KHz/division or
smaller, (e.g., 5 KHz/division) sweep width.
Fingerprinting tests are conducted for two reasons.
First, compliance with hospital policy on acceptance
testing requires that representative samples of all devices
be tested prior to entering the clinical environment for
the first time. Second, both theory and experience have
demonstrated the need for device testing. A device
that radiates unintentionally can victimize other devices.
It may also be susceptible to victimization by ingress
at their egress frequencies and modulation parameters.
Those devices already in the environment are finger-
printed if there is reason to believe that they have the
potential to be an EMI victim or a culprit.
The procedure used to initiate and track an EMI in-
vestigation is shown in
Figure 5.6-5
. Members of the
clinical staff are instructed to call the biomedical engi-
neering department and television services group in any
case of a device malfunction not attributable to a routine
failure. In cases of a new device entering the clinical
environment for the first time or a new application for an
existing device, a request to test the device for compat-
ibility is generated. This call is referred to the clinical
engineer responsible for EMI investigations. When an
incident might be attributable to EMI, the engineer visits
the site as a part of the initial investigation. The possible
victim device is viewed and tested to determine whether
it might have been, or is being, affected by EMI. In-
terviews with the personnel responsible for the area and
the operation of the device(s) must be conducted. Based
on the results of this preliminary investigation, a decision
is reached as to the desirability or feasibility of further
investigation.
The decision to continue the investigation is based on
several factors: Is there a high probability of operator
error? Is this a very rare occurrence? Is this either a very
old device that might be reaching the end of its reliable
life cycle or a new device experiencing infant mortality.
As part of the preliminary investigation process, main-
tenance histories of the device are reviewed. Another
component involves the review of equipment added to
the environment that might have increased the overall
RF hostility in the area enough to cause interference.
Footprinting records of the area are reviewed as are fin-
gerprints of the victim devices(s).
The mode of device failure is an important part of the
evaluation. Is the victim device alarming? Is it operating
erratically? Is it changing its operational parameters either
temporarily or permanently? Is it shutting down? Is there
a latching change that created the alarm? The answers to
these questions can all point to the criminal device or
devices. If the failure mode can be duplicated and if the
failure appears to have been caused by an intentional or
incidental radiator, the culprit device can usually be
identified. If it is within the clinical environment, it is si-
lenced or removed. Many times, no further investigation
is required. An incident report is generated and filed. A
copy is provided to the department initiating the service
request. A copy, if deemed necessary, is given to the ap-
propriate reporting agency, such as the Food and Drug
Administration Center for Devices and Radiological
Health (FDA, CDRH).
If a complete testing procedure is required and a
recent footprint of the area exists, a new set is acquired
and compared to the older set. Changes in the area en-
vironment are noted and analyzed. This is also compared
with any existing fingerprints of the victim device. If no
fingerprints exist for a representative victim device, or
when a device shows signs that its ability to resist ingress
may be compromised, it will be fingerprinted. After
a cursory footprint of the quiet area, to ensure that no
significant changes have occurred in that environment,
the victim device, if practical (size and weight can affect
the test location), can be moved to this area for
fingerprinting.
Generally, the fingerprinting procedure depends on the
operating characteristics of the device. For example, in-
tentional radiators are tested for emissions at their oper-
ating frequencies, modulations, and at second and third
harmonic frequencies. Unintentional radiators (i.e., mi-
croprocessor controlled devices) are tested from 1/4 clock
frequency to 300 MHz. There are exceptions to this
guideline, such as when relatively strong emissions con-
tinue to the 300 MHz point. In these cases, readings are
continued to 1 GHz. The new digital cellular telephones
(GSM) are now in service. ThirdGeneration (3G) wireless
communications devices should be in service in the near
future. The Biomedical Engineering Department has al-
ready received reports of interference to clinical devices by
digital telephones.
As both the fingerprint and footprint tests proceed,
the records presented by the spectrum analyzer are
reviewed. Any RF emissions exceeding a predetermined
