J Med Syst (2006) 30:7181
DOI 10.1007/s10916-005-7982-3
R E S E A R C H A RT I C L E
Power Quality in Clinical Facilities
Antonio Moreno-Munoz Juan J. Gonzalez
Received: 10 May 2005 / Accepted: 24 June 2005


C Springer Science+Business Media, Inc. 2006
Abstract This paper presents preliminary results from a
power quality audit conducted at a clinical building over
the last year. Voltage and current were measured at various
laboratories; it was found that the main problems for the
equipment installed were voltage sags and surges. The paper
examines the causes and effects of power disturbances that
affect computer or any other microprocessor-based equip-
ment and analyzes the auto-protection capabilities of modern
power supplies. The convenience of enhanced power sup-
ply or low-cost customer-side protection solutions is also
discussed. Finally it is addresses the role of the Standards on
the protection of electronic equipment and the implications
for the final costumer.
Keywords Power-quality survey . Clinical facilities .
Veterinarian hospital . Health care equipment . Servicing
sensitive electronic equipment
Introduction
With the generalized use of sophisticated electronic med-
ical devices, almost all the hospitals are shifting toward
entirely electronic IT systems. As clinical facilities evolve
A. Moreno-Munoz (
)
Universidad de Cordoba, Area de Electronica, Departamento de
Electrotecnia y Electronica,
Escuela Politecnica Superior, Campus de Rabanales, Edificio A.
Einstein, C-2 E-14071, Cordoba, Spain
e-mail: amoreno@uco.es
J. J. Gonzalez
Universidad de Cadiz, Area de Electronica, Departamento ISA,
TE y Electronica,
Escuela Politecnica Superior Avda, Ramon Puyol, S/N. E-11202,
Algeciras-Cadiz, Spain
into highly automated digital hospitalsdependent on
computers and other susceptible devices for all aspects
of patient care and diagnosticthe great likelihood of
power disturbances can become serious problems. The
subject of power quality and its relationship to vulnera-
bility of these growing capital investments is becoming
a pressing concern for hospital administrators and clin-
ical laboratory practitioners alike [1]. In order to en-
sure patients well-being and business efficiency, power is
paramount to critical IT data, diagnostic systems, and clinical
veterinarian [2].
Although it is common that public opinion considers util-
ities as the source of power disturbance problems, they fre-
quently argue that there are circumstances beyond their con-
trol. Things like lightning, large switching loads, nonlineal
load stresses, inadequate or incorrect wiring and ground-
ing or accidents involving electric lines. These can create
problems to sensitive equipment if it is designed to operate
within narrow voltage limits, or it does not have adequate
ride-through capabilities to filter out fluctuations in the elec-
trical supply [3].
Instrumentation and control operations require high-
quality and ultra-reliable power in the quantities and time
frames that have not been experienced before. It has been
estimated that more than 30% of the power currently
being drawn from the utility companies is now heading for
sensitive equipment, and this percentage is increasing day
by day [4]. As veterinarian technologies and procedures
become more sophisticated, the potential problems and risks
to equipment intensify. For example, clinical lab instrumen-
tation has evolved from simple, low-speed, single-purpose
devices to complex, high-speed, integrated devices that
support multiple analyses. Internal components may include
laser light sources, mechanical turntables, heating/cooling
elements, pipetting devices, and numerous electronic
Springer
72
sensors and computing circuit boards. These high-end
instruments use components so miniaturized and sensitive
that they can easily falter and fail under power conditions
that earlier-generation equipment may have withstood.
As of today, no standard exists, which clearly defines
the roles and responsibilities of the energy provider, the
equipment supplier, or the high-tech facilities them-
selves in mitigation of PQ-event-caused losses [5]. Over-
all, as modern economies move into the twenty-first cen-
tury, high-tech users are expected to demand for high SQRA
(security, quality, reliability and availability) power supply.
In the new and dynamic deregulated electricity environ-
ment, this is opening business opportunities for acute energy
providers.
Reliability is the ability of the power system to supply
energy within accepted standards and in the amount desired.
Reliability is measured using various indices-characterizing
frequency, duration, and magnitude of adverse effects on the
electric supply. It has been recognized that measures of reli-
ability should include some other power-quality issues (such
as voltage sag and dip disturbances), which are becoming
increasingly significant in the digital age [6]. Contemporary
thoughts on this topic are quite differently focused; modern
power system reliability relates more to [7].
r security: Reducing the vulnerability of the informa-
tion technology equipment and electricity infrastruc-
tures,
r availability and reliability: Supporting extreme bus
voltage reliability, for example five nines (i.e.,
99.9999 availability), or six nines or even higher,
r quality: Assuring power quality for very large num-
bers of digital devices,
r self-healing and redundant: Using distributed energy
sources (DERs) to improve reliability.
The paper is organized as follows. In the following sec-
tion we present the typical power disturbances that affect
electronic equipments; the next section is devoted to the
methodological approach, discussion, and analysis of pre-
liminary results; following these, we describe the proposed
customer-side solutions; and finally we conclude the article.
Facility overview
The Rabanales Biotechnology Campus is supplied through
a 20-kV feeder emanating from the near substation of La
Lancha. This feeder serves 17 buildings which includes
the data center, the departmental R&D laboratories, the vet-
erinary hospital, the main library, the lecture hall building,
the Lucano residence hall, the sports facilities and the
train station. Each building has one 1000-kVA delta-wye
transformer inside, which steps-down the 20 kV to 230/400
Springer
J Med Syst (2006) 30:7181
V for the panel boards distributed through the building
floors.
While the campus server room is properly served via an
UPS, it is recognized that the normal electrical conditions
outside, in the rest of the campus, are subject to generalized
power perturbations. This situation has been the cause of
subtle problems such as computer lockups and electronics
component wear. Besides, it also has caused more devastat-
ing damage as data loss; disk crashes and burnt electronics
in almost all of the buildings.
The veterinarian hospital maintenance department re-
quired the collaboration of the local utility Endesa-Sevillana,
which monitored the service entrance of the campus, con-
cluding that the problems found was not under his respon-
sibility and that the electric service voltage variations were
within the EN-50160 standard limits. Thus, the hospital pro-
posed to the authors the analysis of power disturbances of
the Rabanales campus. Some of the major objectives of the
power quality study were
r correlating the dates presented in the utility report,
r identifying the power disturbances root causes,
r characterizing the electromagnetic compatibility
(EMC) level of equipments and installation,
r developing guidelines that help faculty and personnel
to understand the power quality concern, and
r providing recommendations for cost-effective solu-
tions.
It is possible to say that solving power-quality problems
within commercial consumers premises is a rather complex
task, involving, generally speaking, the following steps:
r Power quality monitoring in order to characterize the
main involved disturbances as well as the actual im-
munity level of the installation.
r Analyzing the sensitive processes and identifying the
critical parts.
r Choosing the adequate immunization techniques.
r Costs estimation: balance between the costs caused
by the disturbances and the investment and costs re-
lated to the immunization project.
r Decision of implementing the solution.
Power disturbances
One of the initial obstacles facing anyone attempting to un-
derstand power quality is the different terminology employed
by utility engineers and by manufactures and users of dis-
turbance analyzers. Although it is possible to follow proper
definitions as they are commonly understood by these two
groups [8]. this paper uses the electromagnetic compatibility
approach to describing power disturbances as classified in
[9].
J Med Syst (2006) 30:7181 73

Table I. Disturbances on computer-related equipment

Power disturbance Typical equipment Typical cause of Typical threshold level Typical duration of
problem disturbance of disturbance disturbance

Impulsive and Lock Miscellaneous Lightning, power 0 to 4 p.u. rated RMS 50 ns to 1 ms and
oscillatory soft errors Hard disk network switching voltage oscillatory 0.3 ms to 5
transient crash Circuit board (large capacitors or s
failure inductors), SCR
controlled loads,
variable speed drives,
photocopiers and
operation of other load

Voltage dip (sag) Miscellaneous soft
errors Resetreboot
Voltage swell Miscellaneous soft
errors Power supply
failure Circuit board
failure
Interruption Hard disk crash
Power supply failure
Power system fault, large
load start-up, faulty
circuit breakers and
loose wiring
Power system fault and
large load disconnect
Power system faults,
local circuit breaker trip,
loose wiring and
equipment failure
0.9 p.u. to 0.1 p.u. rated Short duration: 0.530
RMS voltage cycles 30 cycles -3 s 3
s1 min Long
duration: >1 min
1.1 to 1.8 p.u. rated RMS Short duration: 0.530
voltage cycles 30 cycles -3 s 3
s1 min Long
duration: >1 min
0.1 p.u. rated RMS Short duration: 0.53 s
voltage 3 s1 min Long
duration: >1 min

Clean, uninterrupted power is critical for clinical equip-
ment and laboratory IT systems. If power fluctuates for just
a few moments, data can become corrupted or lost. Internal
system communications can lock up and require a reboot,
damaging sensitive components and interrupting crucial
procedures. A computer or any other microprocessor-based
equipment is said !to be subject to the power disturbances
summarized in Table I [1012].
George W Allen and Donald Segall conducted one of the
most respected studies [13], and, although based in the USA
in 1974, their report is still quoted today. They monitored
AC power to IBM equipment at 200 locations in 25 cities
across the United States, and recorded the various AC power
anomalies that disrupted the equipment operation during a 2-
year time span. The AllenSegall study concluded that 88.5%
of AC power disturbances were transient related. Allen and
Segall found that the most disruptive (49%) of power prob-
lems stemmed from oscillatory, decaying transients. Today,
most electronic equipment uses switch mode power supplies
whose susceptibility to transients and common mode noise is
far greater than traditional linear power supplies. Other stud-
ies [14] recorded nearly double the frequency of disturbances
at Bell Telephone sites.
The well-known ITIC (Information Technology Industry
Council) curve [15], formerly named CBEMA (Computer
Business Equipment Manufacturers Association) curve, is
generally used to evaluate operational voltage limits for elec-
tronic equipment power supplies. The curve is considered to
be a typical design objective for computer hardware design-
ers. The curve establishes magnitude and duration limits
within which input voltage variations do not affect the re-
liability of the electronic equipment. As can be seen, the
steady-state tolerance envelope is in the range of 10%
from the nominal voltage. Within this range, the equipment
will behave properly. For shorter time event the tolerance
is expanded. For example, voltage sags down to 70% of
nominal is permitted for up to 0.5 s, while on the oppo-
site, voltage swells can be permitted rises of up to 120%.
This sensitivity curve only applies to IT equipment; other
equipment generally does have an entirely different sen-
sitivity characteristic, unfortunately unknown for the ma-
jority of equipment [2]. ITIC curve were specifically de-
rived for use in the 60 Hz 120 V distribution voltage sys-
tems. The guideline expects the European user to exercise
their own judicial decision when translating that curve on
equipment operating under 50 Hz 230 V distribution voltage
system [16].
Methodology and results
In order to identify the most likely causes of problems
detected, on-site inspections of equipment and installa-
tions were conducted over the first week of this study.

Springer
74 J Med Syst (2006) 30:7181

Table II. Power line monitoring thresholds this time. As expected the dominant harmonic is the fifth
Disturbance type Threshold setting harmonic, this follows the typical daily load pattern (Fig. 1).
Sag (Dip) <10% 230 V rms It has its maximum of 2% at about 16.16 (THD = 2.41%),
Swell >6% 230 V rms which was measured on Friday (Fig. 2). However, the 6%
Interruption 0 V rms compatibility limit for that harmonic order established in
EN-50160 standard was not violated. For this period, the 7th
harmonic maximum is 0.91. The most likely cause for the
The power site survey followed well-known approaches
level of the 5th harmonic would be the electronic equipment
[7,17,18]. Many power quality problems can be resolved
power supplies. The general rule is that work activity pro-
with an appropriate compliment with the latest edition of the
vides, roughly, the largest pollution levels at mid-afternoon
Reglamento electrotecnico para baja tension (REBT), the
while during the morning harmonic pollution is lower; since
Spanish national electrical code. In addition, in the power
measurement shown were performed in summer, this is likely
site visit, we followed the guidelines proposed in [6,8,19].
to be addressed to the incidence of air conditioner actuation
This process included
between 12 p.m and 19 p.m. and the lack of lighting loads in
r A walk-down of the facilitys electrical system to the morning.
inspect the condition of equipment and becoming fa- Figures 3 and 4 shows the current 3rd, 5th and 7th har-
miliar with the electrical system. monic levels and the current THD. These harmonic levels are
r Interviewing facility electrical personnel and end- under the UNE-EN-21000-3-4 standard limits, but the THD
users on failure of equipment. maximum is 27.54, which violates the 20% established limit
r Identifying and collecting the electronic equipment for the entire installation.
that is most sensitive to power disturbances. However, a harmonic source can be traced from mul-
r Requesting and reviewing equipment literature and tiple measurements in the system. The closer one gets to
electromagnetic compatibility characteristics. the source, the higher the current distortion level will be.
For example, the highest degree of current distortion has
Even though the installation is not really new, a detailed
been found in the organic-chemical laboratory. There has
on-line diagram of the electrical system was not available.
been continuous disconnection of the Residual current cir-
Thus, a complete schematic of the system was developed.
cuit breakers (RCCB) caused by the impact of spectrometers
In each building the air conditioning system and the eleva-
and chromatographers current (Fig. 5 and 6).
tor are powered via separate branch circuit while in each
The current THD maximum for the complete laboratory is
floor, sub-panel serves both the fluorescent lights and the
58.98. Meanwhile, the maximum value for the neutral current
electronic equipments separately. These power circuits are
is 83.51. The phase current 3rd, 5th, and 7th harmonic levels
likely to contain more than 120 PCs and terminals, several
have maximum of 20, 11, and 5%, respectively. The neutral
printers, copiers, facsimile machines, but in addition, chro-
current 3rd, 5th, 7th, and 9th harmonic levels have maximum
matographer, spectrometers, ultra violet spectrophotometers
of 35, 30, 12, and 23% respectively. Figure 7 shows the
and numerous microprocessor-based control and instrumen-
typical neutral current daily pattern of the laboratory.
tation devices.
In Fig. 8 a sample of sites rms voltage level as the function
The monitoring device selected was a portable, stand-
of the time of day is presented.
alone, three-phase power quality analyzer [20]. Some of
Figure 9 indicates that, after a monitoring period of one
the key monitor requirements were: ability to transfer
year, disturbances fall within the voltage tolerance envelope
the surveyed data to in-house computer program, appropriate
of the ITIC curve. Thus, voltages were stable, with a daily
numerical storage and inexpensive and easy to use. Once the
fluctuation, and occasionally minor sag and swells, but in
sites were selected, the monitor was connected to the low-
most of the cases outside of equipment requirements [21].
voltage building entrance and to the different floor subpanel
Although no voltage outages were recorded, such find-
circuits.
ings over a relatively short power audit are not conclu-
The types of disturbances and threshold levels used in the
sive indications of long-term facility or utility continuity of
power quality survey study are listed in Table II. Voltage and
service. During the monitoring time, and after some recent
current values as well as power were logged on the more
power disturbance, computer power supplies were damaged
problematic buildings for periods of 1 day, 1 week, and 1
(most likely due to transients as shown in Fig. 10) and had to
month.
be replaced. Unfortunately our power quality analyzer soft-
For a typical building entrance, after a 13-day period mea-
ware is not able to identity, classify, or highlight these events.
surement, it was evident that the voltage harmonics level
For this purpose we used of a low cost Fluke 43B analyzer.
was relatively high. The campus starts activity at 7.00, this
Identification of transients involves the manually reviewing
is characterized by the rise in the harmonic level at about
all event data and visually scanning all waveforms.

Springer
J Med Syst (2006) 30:7181
Fig. 1. Voltage 5th harmonic
week pattern at LV typical
building entrance.
Fig. 2. Voltage THD week 2.65
pattern at LV typical building
entrance.
2.12
1.59
1.06
0.53
0
thdV2 Avg ()
Proposed solutions
Table III provides an example of cost for some general miti-
gating technologies used for end-user equipment protection
[22, 26].
For protection of high-cost diagnostic equipment, the
most cost-effective technology is online double-conversion
UPS technology. As opposed to an offline or line-interactive
UPS (only recommended for outages), an online UPS pro-
tects against all nine types of power contamination and cre-
ates clean, perfect sine wave power for downstream systems.
Critical clinical equipment and hospital IT systems are com-
pletely isolated from raw utility power and all its irregular-
ities, using the ACDC dual conversion schema: incoming
utility AC power is conditioned and converted to DC power
through the rectifier, a small portion of which is used to
charge the UPS battery; the remaining DC power travels
to the inverter, which produces new, perfect sine wave AC
power to deliver to the instruments.
75
6.00
Analyze/File: 210807.HED
4.80
3.60
2.40
1.20
0
21/08/03 11:36:00 Resolution 1 : 8 03/09/03 08:01:00
h05V2 Avg ()
Analyze/File: 210807.HED
21/08/03 11:36:00 Resolution 1 : 8 03/09/03 08:01:00
For high value critical-process or instrumentation pro-
grams, the availability in nines, as shown in Table IV, is
often stated as being the adopted reliability. Even with a
uninterruptible power supply (UPS), it takes careful con-
figuration planning to achieve the five to six nines of
reliability demanded by high-tech facilities. The main el-
ement that contributes to system reliability is redundancy.
In its basic form, distributed redundancy involves cre-
ating two (redundant) UPS system buses and redundant
power distributed systems. This eliminates as many sin-
gle points of failure as practical, all the way up to the
load equipments input terminals. In order to provide fault
tolerance, some method of allowing the load equipment
to receive power from both UPS power buses must be
provided.
The most advanced online UPS are scalable, with ad-
vanced options that are designed for maximum uptime,
even those applications where the power simply cannot
be interrupted under any circumstances, including routine
maintenance. These designs include parallel redundancy,
Springer
76 J Med Syst (2006) 30:7181

Fig. 3. Current 3rd, 5th, and 7.15

Analyze/File: 210807.HED

7th harmonics at LV typical

building entrance.

5.72

4.29

2.86

1.43

0
21/08/03 11:36:00 Resolution 1 : 8 03/09/03 08:01:00

h03I2 Avg () h05I2 Avg () h07I2 Avg ()

Fig. 4. Current THD at LV 30.29

Analyze/File: 210807.HED

typical building entrance.

24.24

18.18

12.12

6.06

0
21/08/03 11:36:00 Resolution 1 : 8 03/09/03 08:01:00
thdI2 Avg ()

Fig. 5. Voltage and Current of

operating spectrometer.

microprocessor-controlled system, power factor correction
rectifier, internal bypass, and multiple monitoring and com-
munication options.
To protect against fast power system failures, such as
circuit breaker trips or a power system fault, you need
a fast switching approach. Static transfer switches (STSs)
have been applied to accomplish very fast break-before-
make transfers between two AC power sources. It are
important that the two AC power sources be designed
as independent as practical to eliminate any common
failures.
Overall, the factors that improve reliability are

Springer
J Med Syst (2006) 30:7181 77

Fig. 6. Current harmonics of

operating spectrometer.

Fig. 7. Neutral current. Analyze/File: 140104.HED

31.86

25.83

19.80

13.78

7.75

1.73
13/01/04 12:25:00 Resolution 1 : 1 14/01/04 12:30:00

h03I2 Avg () h05I2 Avg () h07I2 Avg () h09I2 Avg ()

Fig. 8. Anomalies detected at 200

LV buildings entrance over the
web on ITIC curve. 180

160
Unacceptable Region
140
Voltage magnitude %

120
110
100
90
80

60
Unacceptable Region
40

20

0 -4 -2 0 2 4 6
10 10 10 10 10 10
Duration in sec.

r Module redundancy r IGBT inverter technology (due to lower parts

r Increased UPS bypass source reliability count)
r Individual battery systems r Use of recognized and agency-listed standard com-
r Simplified operator interfaces and procedural safe- ponents
guards
By the contrary, factors that decrease UPS reliability are

Springer
78 J Med Syst (2006) 30:7181

Fig. 9. RMS voltage log graph Analyze/File: 210807.HED

245.00
at LV typical building entrance
over the week.

237.00

229.00

221.00

213.00

205.00
21/08/03 11:36:00 Resolution 1 : 8 03/09/03 08:01:00

V2 Avg (V) V2 Max (V) V2 Min (V)

Fig. 10. Individual LV

transient detected at equipment
power supply.

r Complicated switchgear systems r UPS topologies with narrow input voltage and fre-
r Shared controls or single points of failure quency windows or topologies
r Common battery systems
r System complexity
r Poor environmental conditions

Table III. Comparative cost of mitigating equipment

Mitigating equipment Typical cost (E) Operating and maintenance costs (%

of initial costs per year)
Transient suppressor 15 100 5
UPS 500/kVA 1000/kVA 25
Standby UPS 100/kVA 1000/kVA 25
Shielded isolation transformer 20/kVA 60/kVA 15
Line conditioner 300/kVA 10
Written pole MG set (15 s) 1100/kVA 5
Dynamic sag corrector 184/kVA 5
CVTs 1000/kVA 10
Low-speed flywheel (15 s) 265 400/kVA 5
High-speed flywheel (15 s) 750/kVA 7
Static transfer switches (STSs) 150.000 (10 MVA) 5
Ultracapacitors (10 s) 1000/kVA 5
Fuel cells 1500/kVA 10

Springer
J Med Syst (2006) 30:7181 79

Table IV. Comparative cost of mitigating equipment interruptions of under one second duration [1]. Typical ride-
through capability of power supplies rage from 10 to 30 ms.
Nines Availability (%) Unavailability Lost time per This time is really too short to be of much help, if switching
year power supply where modified to one second of ride-through
1 90 10% 876 h capability, the over kill installation of UPSs can be substi-
2 99 1{% 87,6 h tuted by a lower-cost power conditioning system [4]. It have
3 99.9 1,0E-3 8,76 h
been demonstrated that there is a large opportunity to embed
4 99.99 1,0E-4 0.876 h
solutions to voltage-sag problems into advanced electronic
5 99.999 1,0E-5 5,3 min
6 99.9999 1,0E-6 31.8 s tools without requiring large-scale mitigation solutions at
7 99.99999 1,0E-7 318 ms the utility or facility-wide level. Basically, an embedded
8 99.999999 1,0E-8 31.8 ms solution is the result of a product re-engineering effort un-
dertaken by the manufacturer to reduce the sensitivity of
the equipment to variations in the quality of the supplied
Installation recommendations power. These measures can range from fairly minor design
substitutions for the most sensitive components in a system
The general recommendation is to follow well-known ITE to major product revisions that incorporate new technologies
installation general guidelines [27, 28], from which the most and possibly require revamping the power scheme inside the
generally forgotten are equipment.
r Each electronic branch circuit should have individual It is of common sense that equipment that is designed with
phase, neutral, and ground conductors. The use of built-in power quality immunity is the best way to achieve
shared neutral conductors as with multi-wire branch system compatibility at lower cost, lower complexity, and
circuits is not recommended. higher performance compared to the hindsight practice
r Connect laser printers and heavy-duty equipment of retrofitting existing equipment with power-conditioning
on individual 20 A branch circuits. Laser printers equipment. Specifically, there are three types of embedded
and heavy-duty copiers produce high current surges solutions.
which cause an increase in the neutral to ground volt- r Component Replacement. Replacing the typical
age. This action can potentially damage other elec- weak-link components on the system with more
tronic equipment localized on the same branch circuit. robust devices in order to improve its immunity.
r Install transient voltage surge suppression (TVSS) r Modified Designs. Designing the system to make it
devices at the service entrance to protect it from utility more power-quality robust.
power problems and lightning strikes. r Supplemental Energy Storage. Embedding on-board
r At electrical distribution panels, where hard-wired energy storage to hold in weak components or sub-
devices help suppress internal voltage transients from systems during voltage sag.
spreading to other circuits in your facility.
r At telecommunications and cable local area network However, manufacturers of digital equipment may well
circuits, which are extremely vulnerable to voltage underinvested in PQ, primarily because it would increase the
transients. cost of their products in highly competitive markets whose
r Install individual TVSS devices at the point of use, customers have not yet shown a willingness to pay extra for
where sensitive equipment connects to electrical out- better power conditioning.
lets. Broadly speaking, there are three basic options to improve
r Specify harmonic canceling transformers in the de- the robustness of DC power supplies in a system [29, 30].
sign of high-tech facilities. r Upsize existing power supply. Because the amount of
voltage sag ride-through time available from a typical
Embedded solution for low-cost equipment linear or switch-mode power supply is directly related
to the loading, power supplies should not be running
Voltage sags at the terminals of sensitive equipment are often at or near their maximum capacity. Upsizing by at
due to faults occurring at a much higher voltage level. Even least two times the nominal load will help the power
though the load current is small compared to the fault cur- supply to ride-through voltage sags. This can also be
rent, the changes in load current during and after the fault still accomplished by adding another identical supply and
strongly influence the voltage at the equipment terminals. It sharing load with the existing unit.
has been discovered that the 85% of computer malfunctions r Change to three-phase input DC power supplies.
attributed to poor Power Quality are caused by voltage sag or Have been demonstrated robust responses from

Springer
80
Table V. Main power protection parameter of computer power supply
Parameter Recommended value
Input range (or operating range) 90 264 VAC
Frequency 47 63 Hz
Output over-voltage protection +12 5% VDC
+5 5% VDC
+3.3 5% VDC
12 10% VDC
5 5% VDC
Output over current protection 240 VA
Hold-up time (or ride-through 15 30 ms
capability)
Mean time between failures 100,000 h
(MTBF)
standard linear and switch-mode DC power supplies
that utilize a three-phase input scheme.
r Change power supplies to use universal input switch-
ing power supplies in every location possible. Typi-
cally, the universal input type power supply has a volt-
age range of 90264 V AC. When connected phase-
to-phase in a 208 V AC system, the power supply
can continue to operate for voltage sags as low as 41
percent of nominal (even at full load). This type of
supply should be specified for DC-powered instru-
mentation and I/O control voltage. Using this type of
power supply for workstation PCs, controllers, and
I/O will lead to excellent voltage sag ride-through.
Computer manufacturers and their local distributors do
no appear to have suitable power-quality information readily
available. Any information that can be obtained is limited
and seems to require a lot of effort to obtain [31].
As a consequence, the almost only source of informa-
tion is the power supply specification guidelines developed
by the owner of the microprocessor system. For example,
in high-end power supplies from the vendors that INTEL
recommends for his microprocessor system [32], the pa-
rameters of interest found on its specification sheets can be
listed as shown in the following Table V. These power sup-
plies are designed to provide protection from higher-than-
normal voltages and currents, and provide a limited amount
of power-line noise filtering or power factor correction PFC
capability.
Most of the inexpensive aftermarket power supplies do not
have this sort of protection; concretely, in this study all the
computer power supplies damaged belongs to inexpensive
clone systems.
What is more, in general, ITE and medical equipment
share connections to supply mains and the provision for
input/output data connections. Thus, a crucial design factor
that demands caution when one considers an off-the-shelf
ITE power supply for a new device is that medical equipment
regulatory requirements must also be met [33].
Springer
J Med Syst (2006) 30:7181
In our case the first recommendation was to replace
the damaged power supply by one with the characteris-
tics cited above, and besides to install individual TVSS
devices.
Conclusion
The massive penetration of electronically controlled devices
and equipment in low voltage distribution networks (the
digital society) could be responsible for the increasing of
power quality problems; particularly, we have found that the
electronics utilized by most computerized diagnostic instru-
ment, found in biochemical R&D laboratories, are becoming
more sensitive to even minor power perturbation.
It have been seen [24] that the level of immunity for
some power quality phenomena would be insufficient to ade-
quately protect terminating equipment from the disturbances
defined in EN 50160. The 95% per week basis for assess-
ing most parameters means that actual power quality could
result in considerable disruption in equipment performance
and yet meet EN 50160.
As stated above, this level of reliability should consider
not only service interruptions but also power quality condi-
tions that may cause process disruptions.
The power supply is a component many users ignore when
acquiring for a microprocessor-based system and it is there-
fore one that some system vendors might choose to skimp on.
After all, a dealer is far more likely to be able to increase the
price of a computer by spending money on additional mem-
ory or a larger hard drive than by installing a better power
supply. Using ITE power supplies in medical equipment re-
quires additional measurements and testing that may reveal
areas of non-compliance. By specifying IEC 60601-1 power
supplies, manufacturers can greatly facilitate validating that
their medical equipment complies with IEC 60601-1.
Thus, hospital and other larger consumers, purchas-
ing large numbers of computers and related electronic
equipments, should always specify the withstand character-
istics of the equipments being purchased.
Finally, for the authors, the European increasing trends
in the consumption of power disturbance mitigation equip-
ment reflects the growing consumer implication in this is-
sue, which must be encouraged with the education through
the distribution of guidelines and related information. Be-
fore considering any further levels of power protection, con-
sumers should know that the power supply in his system
could already afford him a substantial amount of protection.
What is more, consumer must have a basic knowledge about
the power-protection devices available and under what cir-
cumstances he should use them.
Educating customers about PQ is vital to the inter-
ests of energy providers. Facility personnel have access
to a great deal of PQ information, but not many of them
J Med Syst (2006) 30:7181
are equipped to find and filter the most relevant informa-
tion pertaining to their particular facility or their energy
providers T&D system. Surveys have shown that there
is usually a difference between what the customer wants
and what he actually needs. Since an energy provider can-
not provide perfect power, it must educate customers
about this while also helping them understand that there
are several options available for mitigating PQ problems.
Energy providers must help customers explore the avail-
ability and suitability of various mitigation devices, and en-
courage them to work with their equipment manufactures to
make the industrial processes more robust under PQ event
conditions [5].
References
1. Hanada, E., Takano, K., and Kodama, K., Electromagnetic noise
superimposed on the electric power supply to electronic medical
equipment. J. Med. Syst. 27(4):381392, 2003.
2. Appelt K., Power Protection Systems: Protecting Investments in
Capital Health Care Equipment, 2004.
3. Gulachenski, E. M., The low cost alternative to UPS. Electro95
International Proceedings, pp. 97107, 1995.
4. Ward, D. J., Power quality and the security of electricity supply.
Proc. IEEE 89(12):18301836, 2001.
5. Parihar, P., and Liu, E., Power quality services: Technologies and
strategies for energy providers in the deregulated market. Electric.
J. 12(9):7984, 1999.
6. Burns, R. E., Potter, S., and Witkind-Davis, V., After the lights
went out. . Electric. J. 17(1):1115, 2004.
7. Heydt, G. T., Grand challenges in electric power engineering: Ex-
treme system reliability. Power Eng. Soc. Summer Meet., 2002
IEEE 3:16951697, 2002.
8. IEEE guide for servicing to equipment Sensitive to momentary
voltage disturbances. IEEE Std. 1250, 1995.
9. IEEE Recommended Practice for Monitoring Electric Power Qual-
ity. IEEE Std. 1159, 1995.
10. Field handbook of power quality analysis. Dranetz-BMI. 1998.
11. Cumbria, N., Deregt, M., and Rao, N. D., Effects of power distur-
bances on sensitive loads. Proc. Can. Conf. Electric. Comput. Eng.
pp. 11811186, 1999.
12. Koval, D. O., Computer performance degradation due to their sus-
ceptibility to power supply disturbances. Ind. Appl. Soc. Annu.
Meet. 2:17541760, 1989.
81
13. Allen, G. W., and Segall, D., Monitoring of computer installations
for power line disturbances. In IEEE PESC74 pp. 199205, 1974.
14. Goldstein, M., and Speranza, P. D., The quality of U.S. Commercial
AC power. INTELEC, CHI818-4/82-0000-002B, 1982.
15. Information Technology Industry Council ITIC Curve Ap-
plication Note. [Online] Available at http://www.itic.org/
iss pol/techdocs/curve.pdf
16. Arrillaga, J., Bollen, M. H. J., and Watson, N. R., Power quality
following deregulation. Proc. IEEE 88(2):246261, 2000.
17. Michaels, K. M., Sensible approaches to diagnosing power quality
problems. IEEE Trans. I. A. 33(4):11241130, 1997.
18. Martzloff, F. D., and Gruzs, T. M., Power quality site surveys:
Facts, fiction, and fallacies. IEEE Trans. I. A. 24(6):10051018,
1988.
19. IEEE Recommended Practice for Powering and Grounding Sensi-
tive Electronic Equipment. IEEE Std. 11001992.
20. GSC-57., HT Instruments. Available at www.htinstruments.com
21. Varian 1200 Liquid Chromatographer/Mass Spectrometer Pre-
installation Instructions. Available at www.varianinc.com
22. Dugan, R. C., McGranaghan, M., Santoso, S., and Beaty, H. W.,
Electrical Power System Quality, McGraw-Hill, New York, 2003.
23. McGranaghan, M., and Roettger, B., Economic evaluation of power
quality. IEEE Power Eng. Rev. 22(2):812, 2002.
24. Standler, R. B., Protection of small computers from disturbances
on the mains. IEEE IAS Conf. 2:14821487, 1988.
25. Stebbins, W. L., Power line disturbances: A users perspec-
tive on the selection and application of mitigation equipment
and techniques. IEEE Textile Ind. Tech Conf. pp. 4/14/7,
1989.
26. Schenung, S., Overcoming Cost Reduction Barriers for Advanced
Power Quality Mitigation Systems, EPRICSG, Palo Alto, CA. TP-
114370, 1999.
27. Online: http://www.cpccorp.com/tips.htm
28. Three Phase Power Source Overloading Caused by Small Com-
puters and Electronic Office Equipment. ITI Information Letter.
[Online]. Available at http://www.itic.org/technical/3phase.htm
29. Donmez, A., Stephens, M., Thomas, C., and Banerjee,
B., Embedded Power Quality Solutions for Advanced Ma-
chine Tools. Power Quality 2001 Conference. Available at
http://www.f47testing.com/cncpq/files/PQA2001paper.pdf, 2001.
30. Stephens, M., Soward, J., Johnson, D., and Ammenheuser, J.,
Power quality Solutions for Semiconductor Tools, equipment de-
sign solutions. [Online]. Available at http//www.powerquality.com,
2000.
31. Annex B2, [Online]. http//www.iee.org.uk/PAB/EMC/core.htm
32. Online: http://www.formfactors.org/
33. OBrien, F., ITE Power Supplies and Medical Equipment,
2000.
Springer