Improving Power Quality Immunity in Factory Automation

Power quality events are of increasing concern because today's computers, PLCs, adjustable-speed drives (ASDs) and automated manufacturing devices, are sensitive to such imperfections. Among all categories of electrical disturbances the voltage sag (dip) and momentary interruption are the nemeses of the automated industrial processes. ^[1] Voltage sag is commonly defined as any low voltage event between 10% and 90% of the nominal RMS voltage lasting between 0.5 and 60 cycles. Momentary voltage interruption is any low-voltage event of less than 10% of the nominal RMS voltage lasting between 0.5 cycles and 3 sec. In medium-voltage distribution networks, voltage sags mainly are caused by power system faults.

Fault occurrences elsewhere can generate voltage sags affecting consumers differently according to their location in the electrical network. Even though the load current is small compared to the fault current, the changes in load current during and after the fault strongly influence the voltage at the equipment terminals. It has been discovered that the 85% of power supply malfunctions attributed to poor power quality are caused by voltage sag or interruptions of fewer than one second duration. ^[2] Starting large motors can also generate voltage sags, although usually not so severe. In comparison with interruptions, voltage sags affect a larger number of customers and for some customers voltage sags may cause extremely serious problems. 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 electrical supply. ^[3] (Fig. 1)

The Samafrava plant is an up-to-date semiautomatic factory devoted to the mass production of plastic bags. The plant is supplied through a 25-kV feeder emanating from a distant substation. The facility has its own 1,000-kVA delta-wye transformer that steps-down the 25 kV to 30/400 V for the panelboards distributed through the plant. The low-voltage side of the installation is designed as a TN-S system. The primary distribution is a compromise between radial and shunt schemes. The installation has grown in a haphazard way, without a consistent structure. In such a scheme it is imperative that the neutral conductor is connected to earth only once, at the main earthing terminal. Otherwise, the benefits of the TN-S wiring configuration — improved EMC and power quality — are lost. The most critical loads in the plant are extruder process machines used to produce plastic bags. In a blown film extrusion process, film is extruded as a continuous tube, air-cooled, collapsed, and wound onto rolls as bags into single-layer widths. (Fig. 2)

On the other hand, the Infrico facility is supplied through a 25-kV feeder emanating from a distant substation. The facility has its own 630-kVA delta-wye transformer that steps-down the 25 kV to 230/400 V for the panelboards distributed through the building. The Infrico plant is an up-to-date semiautomatic factory devoted to the manufacturing of refrigerators and cold systems for the hostelry and pastry industry. Recently, it has incorporated a flexible manufacturing system (FSM) from one of the world leading supplier; this is a modular sheet metal FSM for the assembly of the stainless steel cabinet manufactured for the different refrigerators included in the Infrico catalog.

METHODOLOGY AND RESULTS

It is possible to say that solving power quality problems within commercial consumers' premises is a rather complex task. Some main objectives of the power quality study included:

1. Detecting the main involved disturbances by power quality monitoring.

2. Identifying the power disturbance's root causes.

3. Characterizing the electromagnetic compatibility (EMC) level of equipment and installation.

4. Analyzing the sensitive processes and identifying the critical parts,

5. Developing guidelines that help faculty and personnel to understand the power quality concern.

6. Choosing the adequate immunization techniques.

7. Providing recommendations for implementing cost-effective solutions.

PROPOSED SOLUTIONS

In order to identify the most likely causes of problems detected and prior to the power quality monitoring, on site inspections of equipment and installations were conducted over the first week of both studies. The power site inspection followed well-known approaches, which included: ^{[4], [5], [6]}

1. A walk-down of the facility's electrical system to inspect the condition of equipment and become familiar with the electrical system.

2. Interviewing facility electrical personnel and end-users on failure of equipment.

3. Identifying and collecting the electronic equipment most sensitive to power disturbances.

4. Requesting and reviewing equipment literature and electromagnetic compatibility characteristics.

In both cases, current installation lacks organization and rationality in its approach, probably due to the numerous modifications since the original project. The monitoring device selected was a portable, stand-alone, 3-phase power quality analyzer. Some of the key monitors requirements included the ability to transfer the surveyed data to an in-house computer program, appropriate numerical storage, and inexpensive and easy to use. The monitors were connected to the low-voltage building entrance and to the sites selected simultaneously. Although the device installed can monitor various power quality disturbances, only RMS variations (sags and swells), interruptions and outages are considered.

Figs. 1 and 2 illustrate all the RMS variations that were recorded at the facility (all locations, all phases) during the monitoring period. This kind of diagram is known as a magnitude-duration scatter plot. It also translates information from the well-known ITIC (Information Technology Industry Council) curve, formerly named the CBEMA (Computer Business Equipment Manufacturer's Association) curve.^[7] The curve is considered to be a typical design objective for computer hardware designers. The curve establishes magnitude and duration limits within which input voltage variations do not affect the reliability 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 events, the tolerance is expanded. For example, voltage sags down to 70% of nominal are permitted for up to 0.5 sec. Conversely, 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 sensitivity characteristic, unfortunately unknown for the majority of equipment.???^[2] The ITIC curve was specifically derived for use in 60-Hz, 120-V distribution voltage systems. The guideline expects the European user to exercise his or her own judicial decision when translating that curve on equipment operating under 50-Hz, 230-V distribution voltage systems.^[8] We consider this curve only as a reference.

While the Samafrava study shows that the blown film extrusion system will experience about four voltage sags and two momentary interruptions every month, the Infrico study indicated that an average site will experience about 45 voltage sags and five momentary interruptions every month — which is quite dramatic.

A common way of presenting voltage-dip survey results is from a density table. Following the method recommended in IEEE Std. 493 ^[9] and IEEE Std.1346 ^[10], it breaks the sags and interruptions down into count versus magnitude versus duration. Each bin of the table gives the (average) number of voltage dips within the given range of magnitude and the given range of duration. A further step of presenting the results is through a cumulative table. It shows the number of events per site per year that are more severe than the given magnitude and duration. The values obtained in the cumulative voltage-dip table can be interpreted as function values of a two-dimensional function that gives the cumulative dip frequency as a function of magnitude and duration. For the figure, intermediate function values have been calculated by using an interpolation algorithm. The voltage dip contour chart corresponding to Fig. 3 is shown in Fig. 4, with contours indicated for dip frequencies equal to 5, 10, 50, 100 and 200 events per year.

REFERENCES

The sensitivity of equipment (the power supply in Fig. 4) to voltage sags is usually expressed only in terms of the magnitude and duration of the voltage sag. For this purpose, a “rectangular voltage-tolerance curve” is used. This curve indicates that voltage sags longer than the specified duration and deeper than the specified voltage magnitude will lead to a malfunction. The information obtained from a device can be directly related to the voltage dip contour chart corresponding to the facility. Thus, in this case, if a power supply can tolerate voltage dips down to 37% for a duration of up to 20 ms, then the equipment will trip 53 times per year.

The general recommendation is to follow well-known ITE installation general guidelines.^[11] It is possible to correct voltage sags using conventional technologies, such as uninterruptible power supply (UPS) systems. However, these are universal correction devices because they have typically been designed for the correction of all types of voltage disturbances. This implies the amount of energy that a UPS is required to store is based upon the long duration of a typical voltage outage. However, this may not be necessary in the case of a voltage sag. In addition, a UPS would need to be able to withstand not only the load current, but also the full load voltage.

Recently, new technologies such as custom power devices based on power electronic concepts have been developed to provide protection against power quality problems. Generally, custom power devices are divided into three categories, such as a static series compensator like the DVR, a static shunt compensator like the distribution static compensator (DSTATCOM), and static series and shunt compensator like the Unified Power Quality Conditioner (UPQC).^[12]

The essence of DVR is a voltage that is injected via a series-connected boost transformer to achieve the load voltage previous to the voltage sag. This power electronic converter comprises a capacitor bank, pulse-width-modulated (PWM) inverters, ac harmonic filters, injection transformers and an optional energy storage system. Within the family of DVRs there exists various compensation techniques.^[13] However, distressed by the frequency of voltage sag occurrences and wishing to avoid an overwhelming capital expenditure in power conditioning equipment, the audit addresses embedded solution approaches.

It's common sense that equipment designed with built-in power quality immunity is the best way to achieve system compatibility at lower cost, lower complexity and higher performance compared to the hindsight practice of retrofitting existing equipment with power-conditioning equipment. Specifically, there are three types of embedded solutions:

1. Component replacement. Replacing the typical “weak link” components on the system with more robust devices in order to improve its immunity.

2. Modified designs. Designing the system to make it more power quality robust.

3. Supplemental energy storage. Embedding on-board energy storage to hold in weak components or subsystems during voltage sag.

All power electronic supplies are supported by an ac-dc diode rectifier followed by a dc-dc voltage regulator. The latter transforms the nonregulated dc voltage from the dc-link into regulated levels. A capacitor is connected to the dc-link to reduce the voltage ripple at the input of the voltage regulator. If the ac voltage suddenly drops, the capacitor is discharged not only for half a cycle (as in normal operation) but also for a longer period. This drop will continue until a new equilibrium is reached and the dc voltage is lower than the input ac level. The new operation point will have a lower voltage on the dc bus. The duration of the discharge of the capacitor is directly dependent to the magnitude of the voltage sag, the size of the capacitor and load current. A larger capacitor or higher operating voltage will increase the ride-through capability of the power supply (typically ranging from 10 ms to 30 ms).

Three basic options to improve the robustness of dc power supplies in a system include:^[1],[14]

1. Upsize existing power supply. Because the amount of voltage sag ride-through time available from a typical linear or switch-mode power supply is directly related to the loading, power supplies should not be running at or near their maximum capacity. Upsizing by at least two times the nominal load will help the power supply to ride-through voltage sags. This also can be accomplished by adding another identical supply and sharing load with the existing unit.

2. Change to 3-phase input dc power supplies. It has been demonstrated the robust responses from standard linear and switch-mode dc power supplies that use a 3-phase input scheme.

3. Change power supplies to use universal input switching power supplies in every location possible. Typically, the universal input-type power supply has a voltage range of 87 Vac to 264 Vac. This solution usually has a single-switch PFC boost follower pre-regulator at the front end of the diode bridge. This is because the boost converter topology has continuous input current that can be shaped through the use of a multiplier and average current-mode control to achieve near-unity power factor.

The boost converter is traditionally designed to have a fixed output voltage greater than the maximum peak line voltage. However, the boost voltage does not have to be well regulated or fixed because in the next stage, the step down converter or the inverter can be designed to handle the voltage variations. As long as the boost voltage is above the peak input voltage the converter will regulate properly ^[15].

As a consequence, when connected phase-to-phase, the power supply can continue to operate for voltage sags as low as 41% of nominal (even at full load). In addition, its typical hold-up time or shutdown-time off can range between 20 ms and 170 ms.

1. Moreno-Muñoz, A. and Redel, M.D. (2005). “Calm in the Campus: Power Disturbances Threaten University Life,” IEE Power Engineer, Vol. 19, No. 4, August 2005, pp. 34-37.

2. “Recommended Practice for Monitoring Electric Power Quality, 1995,” IEEE Std. 1159.

3. Gulachenski, E.M. “The Low Cost Alternative to UPS,” In Electro95 International Proceedings, 1995, pp. 97-107.

4. Moreno-Muñoz A.; Redel, M. D; González, M. “Power Quality in High-tech Campus: A Case Study,” Proceedings of the Institution of Mechanical Engineers, Part A: Journal of Power and Energy 2006; 220 (3): 257?269.

5. Michaels, K. M. “Sensible Approaches to Diagnosing Power Quality Problems,” In IEEE Trans. on I. A., 1997, Vol. 33, No. 4, pp. 1124-1130.

6. “IEEE Recommended Practice for Powering and Grounding Sensitive Electronic Equipment,” IEEE Std. 1100-1992.

7. Information Technology Industry Council ITIC Curve Application Note (online) http://www.itic.org/iss_pol/techdocs/curve.pdf.

8. Arrillaga, J.; Bollen, M.H.J.; Watson, N.R. “Power Quality Following Deregulation.” In Proceedings of the IEEE, 2000, Vol. 88, Issue 2, pp.246-261.

9. “IEEE Recommended Practice for the Design of Reliable Industrial and Commercial Power Systems.” IEEE Std 493-1997.

10. “IEEE Recommended Practice for Evaluating Electric Power System Compatibility with Electronic Process Equipment,” IEEE Std. 1346-1998.

11. Moreno-Muñoz, A.; Redel, M.D.; Prieto, A.L.; Plaza, A.; González, M. and Luna, J. “Power Quality in a University Campus: The User'sPerspective,” ICREPQ'04 Congress, Barcelona, March 31 and April 1-2, 2004.

12. Hingorani, N.G. and Gyugyi, L. “Understanding FACTS: Concepts and Technology of Flexible AC Transmission Systems,” New York: IEEE Press, 2000.

13. Moreno-Muñoz A.; Oterino D.; González M.; Olivencia F.A.; De-la-Rosa, J.J.G. “Study of Sag Compensation with DVR.” In: Proceedings of the IEEE MELECON 2006, 990-996.

14. McGranaghan, M. and Roettger, B. “Economic Evaluation of Power Quality,” In IEEE Power Engineering Review, 2002, Vol. 22, Issue 2, pp.8-12.

15. Moreno-Muñoz A. and De-la-Rosa J.J.G. Analysis of Voltage Dips in PWM AC-DC Converters,” In: Proceedings of the International Symposium on Power Electronics, Electrical Drives, Automation and Motion, SPEEDAM 2006; S41: 6?9.

16. “DC Buffer Block Deeps Supplies Up,” http://www.electronicstalk.com/news/oro/oro105.html.

17. http://www.phoenixcontact.com.

18. Donmez, A.; Stephens, M.; Thomas, C.; Banerjee, B. “Embedded Power Quality Solutions for Advanced Machine Tools,” Power Quality 2001 Conference, http://www.f47testing.com/cncpq/files/PQA2001paper.pdf.

19. Stephens, M.; Soward, J.; Johnson, D; Ammenheuser, J. “Power Quality Solutions for Semiconductor Tools, Equipment Design Solutions,” 2000, http//www.powerquality.com.

20. Annex B2, http//www.iee.org.uk/PAB/EMC/core.htm.

A. MORENO-MUÑOZ, A. R. GIL DE CASTRO, V. PALLARÉS and J. J. G. DE LA ROSA*
Universidad de Córdoba. Departamento A.C., Electrónica y T.E. Escuela Politécnica Superior. Campus de Rabanales. E-14071 Córdoba.
*Universidad de Cádiz. Área de Electrónica. Dpto. ISA, TE y Electrónica. Escuela Politécnica Superior Avda. Ramón Puyol, S/N. E-11202-Algeciras-Cádiz, (Spain).