UPS 3

Eng nginee ineerin ring g Enc Encycl yclop ope edia Saudi Sa udi A ramco DeskTop Standards

Selectin Se lectin g Typ Types es and Size Sizes s of IInv nverters erters

Note: The source of the technical material in this volume is the Professional Engineering Development Program (PEDP) of Engineering Services. Warning: The material contained in this document was developed for Saudi Aramco and is intended for the exclusive use of Saudi Aramco’s employees. Any material contained in this document which is not already in the public domain may not be copied, reproduced, sold, given, or disclosed to third parties, or otherwise used in whole, or in part, without the written permission of the Vice President, Engineering Services, Saudi Aramco.

Chapter : El Electrical File Reference: EEX21103

For additional information on this subject, contact W.A. Roussel on 874-1320

Engineering Encyclopedia

Electrical Selecting Types and Sizes of Inverters

C O NT E NT S

P AG E S

INVERTERS: INVERTERS: BASIC FUNCTIONS FUNCTIONS AND OPERATION................................................ OPERATION................................................ 1 DETERMINING TYPES OF INVERTERS FOR USE IN SAUDI ARAMCO ELECTRICAL ELECTRICAL INSTALLATIONS.................................................................. INSTALLATIONS.................................................................. 9 FACTORS THAT THAT EFFECT THE SIZE OF AN INVERTER INVERTER .......................................... 22 WORK AID 1: PROCEDURES AND TECHNICAL AND ECONOMIC REQUIREMENTS FROM SADP-P-103 AND ESTABLISHED ENGINEERING PRACTICES FOR DETERMINING THE APPROPRIATE TYPE OF INVERTER FOR USE IN SAUDI ARAMCO ELECTRICAL INSTALLATIONS.................................................................................29 GLOSSARY ..................................................................................................................... 30

Saudi Aramco DeskTop Standards



C O NT E NT S

P AG E S

INVERTERS: INVERTERS: BASIC FUNCTIONS FUNCTIONS AND OPERATION................................................ OPERATION................................................ 1 DETERMINING TYPES OF INVERTERS FOR USE IN SAUDI ARAMCO ELECTRICAL ELECTRICAL INSTALLATIONS.................................................................. INSTALLATIONS.................................................................. 9 FACTORS THAT THAT EFFECT THE SIZE OF AN INVERTER INVERTER .......................................... 22 WORK AID 1: PROCEDURES AND TECHNICAL AND ECONOMIC REQUIREMENTS FROM SADP-P-103 AND ESTABLISHED ENGINEERING PRACTICES FOR DETERMINING THE APPROPRIATE TYPE OF INVERTER FOR USE IN SAUDI ARAMCO ELECTRICAL INSTALLATIONS.................................................................................29 GLOSSARY ..................................................................................................................... 30




INVERTERS: BA SIC FUNCTIONS AND OPERATION OPERATION An inverter is defined as a device that changes Direct Current (DC) into Alternating Current (AC). The inverter can be constructed of a rotary element or Silicon Controlled Rectifiers (SCRs) to produce the AC output. The function of the inverter is to provide a smooth, regulated AC sine sine wave to the load from from a DC input. The ability of an inverter inverter to provide provide a smooth, regulated AC sine wave output is vital in the application of an inverter to a critical load. The operation of a static inverter can be explained through the use of mechanical switches as shown in Figure 1. Mechanical switches switches 1 and 1' are operated operated in unison, and mechanical mechanical switches 2 and 2' 2' are also operated operated in unison. Source current (I s) is DC and will always flow in the same direction. direction. The load that is supplied from the the mechanical switches that are simulating an inverter requires an alternating current. The operation of the mechanical switches will produce the required load alternating current. Switches 1 and 1' are closed, and switches 2 and 2' are open initially. Load current (I L) will flow from point A to point B. To reverse the load current flow, switches 1 and 1' are opened and switches 2 and 2' are closed. IL will flow from point B to point A. Continual reversal of the switch positions will result in an AC current flow across the load.


1



Simulated Static Inverter Using Mechanical Switches Figure 1

The mechanical switches that are shown in Figure 1 are not static switches because they contain parts that move. Figure 2 shows a bridge SCR static inverter. The mechanical switches that were shown in Figure 1 have been replaced with electrical switches that are called silicon controlled rectifiers (SCRs). Figure 2 also shows commutating inductors (L) and a commutating capacitor (C). The inductors and the capacitor are used to turn off one pair of SCRs as the other pair of SCRs is turned on. SCR 1 and SCR 1' replace switches 1 and 1' of Figure 1, respectively, and SCR2 and SCR2' replace switches 2 and 2' of Figure 1, respectively. The SCRs are turned on to conduct current flow and turned off to stop current flow. The SCRs are turned on through the application of a gate pulse while the SCR is forward biased. The SCR's are turned off through commutation.


2



Bridge SCR Static Inverter Figure 2

Commutation is the process of turning on one SCR while turning off another SCR. Commutation can be accomplished through use of a capacitor-inductor type commutation circuit as shown in Figure 3. The following conditions of the commutation circuit are shown in Figure 3A: _150 VDC power is supplied to the commutation circuit. _SCRA is conducting current through winding T A (indicated by the solid black SCR symbol). _Capacitor C is charged to 300 volts (twice the value of the line voltage). Saudi Aramco DeskTop Standards

3



_Current flow through the load is from point 1 to point 2.


4



When the circuit conditions that are shown in Figure 3A are established, SCR B is gated on at the desired moment of commutation as shown in Figure 3B. The following sequence of events occur when SCR B is gated on: _Capacitor C discharges through Inductor L, SCR A, and SCR B. _This discharge path applies -300 volts to the cathode of SCR A. _This application of voltage in the reverse direction drives the forward current of SCRA to zero, which turns off SCR A. _Inductor L is in the circuit to limit the rate at which capacitor C discharges. The conditions of the circuit after commutation has occurred are shown in Figure 3C as follows: _SCRB is conducting current through winding T B (indicated by the solid black SCR symbol). _Capacitor C is charged to 300 volts in preparation for the next commutation cycle; however, the polarity of the change is reversed. _Current flow through the load has reversed so that the current flow now is from point 2 to point 1. This process of commutation continues to alternate such that one SCR is turned on and the other SCR is turned off. This process results in an alternating current flow to the load.


5



Simplified Inverter Figure 3A


6



Simplified Inverter Figure 3B


7



Simplified Inverter Figure 3C


8



DETERMINING TYPES OF INVERTERS FOR USE IN SAUDI ARA MCO ELECTRICAL INSTAL LATIONS There are various types of inverters on the market that will perform the same function. Each type of inverter has certain advantages and disadvantages in the inverter's operating characteristics. The Electrical Engineer must understand the operation and the advantages and disadvantages of each inverter type to determine the proper inverter to be installed in a particular Saudi Aramco installation. This section provides information on the following topics that pertain to the determination of types of inverters for use in Saudi Aramco electrical installations: _Ferro-Resonant _Pulse Width Modulation _Advantages and Disadvantages Ferro-Resonant The ferro-resonant inverter is a combination of solid-state and iron core components. Ferroresonant inverters use several self-contained inductors and a capacitor to convert a square wave into a sine wave. This section will discuss the following topics concerning a ferroresonant inverter: _Basic Characteristics _Operation Basic Cha r acteristics

The ferro-resonant inverter does not directly convert DC to AC. Before the signal can be applied to the ferro-resonant inverter, the DC must be converted into a square wave AC. The conversion to a square wave AC will generally be accomplished through use of SCR's. The ferro-resonant inverter is a way to obtain a stable AC voltage at relatively low cost. Ferro-resonant inverters have various operating characteristics. The ferro-resonant inverter can be designed so that the output voltage is isolated from the input voltage and so that the output has a different nominal voltage than the input. The output windings of the ferroresonant transformers may be connected in single-phase or three-phase wye-connected circuits. The ferro-resonant transformer may have a delta-connected output if interphase transformers are used. A basic equivalent circuit for a ferro-resonant inverter is shown in Figure 4. The basic ferroresonant inverter is comprised of three components: a linear reactor, a capacitor, and saturable transformer. If isolation is not required, the saturable transformer can be replaced by a saturable reactor. The ferro-resonant inverter receives a square wave input and provides a sine wave output to the critical loads. Saudi Aramco DeskTop Standards

9




10



Equivalent Circuit of a Basic Ferro-Resonant Inverter Figure 4

Operation

In reference to Figure 4, the saturable transformer tends to maintain a constant volt-second integral. At a constant frequency, any increase in voltage will result in increased exciting current for the saturable transformer. At low input voltages, the series combination of the capacitor and linear reactor cause the voltage across the capacitor and saturable transformer primary to be above the input voltage. As the input voltage rises, the saturable transformer inductance drops; and, at high input voltages, the parallel combination of the saturable transformer and the capacitor becomes a net inductor. Also, the circuit becomes an inductive divider. The output voltage will rise with the frequency of the input voltage because the voltage must rise to maintain the same volt-second integral as the period is reduced.


11



Figure 5 shows a block diagram of a simple UPS system that uses a Ferro-Resonant Inverter. The power source provides a 3_ AC input to a 3_ rectifier that converts the 3_ AC input to a DC output. The DC output of the 3_ rectifier supplies power to the battery DC bus, the ferroresonant inverter, and the logic circuits. The logic circuits will provide the gate signal to the SCRs in the fixed frequency inverter to produce an AC square wave output with a frequency of 60 Hz. The 60 Hz AC square wave output of the fixed frequency inverter is supplied to the input of the ferro-resonant transformer. The ferro-resonant transformer will filter the output to produce a resultant 60 Hz AC sine wave.


12



Ferro-Resonant (Cont'd)

Block Diagram of a Simple UPS System That Uses a Ferro-Resonant Inverter Figure 5


13



Pulse Width M odulation Inverters that convert DC to AC through use of pulse width modulation are among the most advanced inverter technologies in use today. Pulse Width Modulated (PWM) inverters switch the power elements to approximate sine wave outputs. This section will discuss the following topics about pulse width modulation: _Basic Characteristics _Operation Basic Cha r acteristics

The ability to control the output of the PWM inverter is the basic operating characteristic that sets the PWM inverter apart from the ferro-resonant inverter. In order to provide a more controlled output, the construction and design of the PWM inverter is much more complex. The pulse width modulated inverters use more than one input signal to produce a sine wave output. The use of more than one input signal allows for more potential problems that are caused by a loss of one of the input signals. The needs for a more complex control system and for more than one input signal cause the pulse width modulated inverter to be more expensive than an equivalent ferro-resonant inverter. The increased cost of the PWM inverter can be justified if the critical loads require the controlled output. Operation

Figure 6, a block diagram of a PWM inverter, shows the following blocks: _Reference sine wave generator _Amplitude control circuit _Summing amplifier _Current limit circuit _Voltage regulator circuit _Triangle waveform generator _Comparator #1 and #2 _Inverter power stage _Filter The reference sine wave generator produces a reference sine wave at the fundamental power frequency. The output of the reference sine wave generator is supplied to the amplitude control circuit.


14



The amplitude control circuit also receives a signal from the summing amplifier. The signal from the summing amplifier is for use in the adjustment of the amplitude of the reference sine wave. Such adjustment is based on the magnitude of the inputs to the summing amplifier from the current limit circuit and the voltage regulator circuit. The sine wave output of the amplitude control circuit is supplied to comparator #1 and to comparator #2. The triangle waveform generator produces two triangle waveforms that are 180 o out of phase. One triangle waveform is supplied to comparator #1, and the other triangle waveform is supplied to comparator #2.


15



PWM Inverter Figure 6


16



The comparators compare the sine wave from the amplitude control circuit with the triangle waveforms from the triangle waveform generators, as shown in Figure 6. The output of the comparators is switched each time the instantaneous voltage of the triangle waveform is equal to the instantaneous voltage of the sine wave. The switching of the comparator's outputs produces a modulated square wave output from each comparator that is also shown in Figure 7.


17



Pulse Width M odulation (Cont' d)

Comparator Waveforms Figure 7


18



The modulated square wave outputs are supplied to the inverter power stage to fire the power SCR's in each leg of the inverter power stage. The firing of the power SCR's produces a modulated square wave inverter output that is shown in Figure 8. The modulated squarewave output is supplied to a filter that produces a sine wave inverter output for use by the critical loads. The sine wave inverter output also is shown in Figure 8.

Inverter Output Waveforms Figure 8


19



Advantages and Disadvantages The ferro-resonant and pulse width modulated inverters have advantages and disadvantages that cause one inverter to fit an application better than the other inverter. To determine the correct type of inverter for installation, the Engineer must understand the advantages and disadvantages of each type of inverter so that the different types can be compared. Also, the Engineer must understand the needs of the installation to match the proper inverter to the installation. This section will discuss the advantages and disadvantages of the following types of inverters: _Ferro-Resonant _Pulse Width Modulation Ferro-Resonant

The main advantage of a ferro-resonant inverter is the cost of the inverter. Because of the simple design and operation of the ferro-resonant inverter, the cost of this type of inverter is much lower than the cost of any other inverter. An inherent disadvantage of ferro-resonant inverters is the phase shift between input and output voltage and current. This phase shift is caused through the use of the inductors and capacitors in the circuit. A slight phase shift occurs at light loads, but the phase shift increases to 90 o lagging at full load. Consequently, an inverter that is synchronized to the alternate power source will receive a sync pulse when the inverter output is crossing the zero axis. This disadvantage can be offset by elaborate sync control circuits in single phase inverters, but use of these circuits causes three phase voltages to be unequally displaced from their normal 120 o phase separation. The remaining disadvantgages of ferro-resonant inverters are that these inverters are less efficient and produce more ambient noise than do other types of inverters. Pulse Width M odulation

Pulse width modulation (PWM) has the following advantages over the ferro-resonant inverters: Voltage regulation is accomplished within the inverter section that eliminates the need for a separate voltage regulator section. With PWM circuits, the AC waveform shaping requirements are much simpler because of the sine wave approximation on the inverter output. Fault protection is enhanced. Because of the built in feedback loop, the PWM circuits can quickly cause the inverter to "fold back" and decrease its output voltage to a safe level in the event of a load fault.


20



Because of the smaller output filter, the output impedance is lower than with the ferro-resonant inverter. Very close voltage regulation is possible with the PWM inverter. There are two main disadvantages of a PWM inverter: The cost of a PWM inverter increases over a ferro-resonant inverter due to the increased complexity of the inverter and control circuitry. An inverter that uses PWM logic that is controlled by 1.2 kHZ internal oscillator is susceptible to false signals that are caused by operation of walkietalkie radios in the area of the inverter.


21



FACTORS THAT EFFECT THE SIZE OF AN INVERTER An inverter must be properly sized to provide an AC output to the critical load or loads that are connected to the inverter. An inverter that is too small to handle the requirements of the critical load will not maintain an AC output during critical times. An oversized inverter will increase the cost of the inverter in two ways. The initial cost of an oversized inverter will be greater than the properly sized inverter. Second, oversizing of an inverter will cause the inverter's efficiency to decrease and its operating cost to increase. This section will discuss the following factors that affect the size of an inverter: _Combined Critical Load Power Factor _Combined Inrush Current _Critical Load Growth Factor _Harmonics Combined Cr itical Load P ower F actor It is essential for the UPS manufacturer and the Design Engineer to know the power factor of the combined critical load because the inverter must supply the reactive power and the real power that are required by the load. In rotating equipment inverters, the inverters supply the reactive component of power from energy that is stored in the magnetic fields. In static inverters, the static inverters must use commutating energy from inductors and capacitors to supply the reactive component of power and must be designed to provide a specific power factor range. The value of the combined critical load power factor will have a direct effect on the size of the inverter. For example, a UPS inverter that is rated for 50 kVA can supply 50 kW of real power at a unity power factor of one to operate critical loads. If the same inverter operated at a .8 power factor, the inverter could only supply 40 kW of real power to operate critical loads. If the actual real power that is required by the critical loads is 50 kW, a UPS inverter that operates at a .8 power factor must be rated for 62.5 kVA. Figure 9 is a table of the power factors of typical Saudi Aramco critical loads. The first column lists the typical types of equipment (loads) that are found in Saudi Aramco installations. The second column lists the power factor that is associated with the load. All of the power factors that are given are lagging unless otherwise noted. The power factors of all the loads that are supplied to an inverter would be combined to produce the combined critical load power factor.


22



Power Factors of Typical Saudi Aramco Critical Loads Figure 9


23



Combined Inrush Curr ents Inverters should be sized to deliver one half of the combined inrush currents for the critical load. The inverter size is determined on the premises that some equipment turn-ons will be delayed; that the inrush current demand will be shared between the UPS circuit breaker, battery charger, and inverter in one supply path; and that the alternate source circuit breaker and transformer will be in the other current supply path. An inverter installed in a parallel redundant system without a bypass source to which the inverter can be transferred requires that each inverter be sized to deliver the entire inrush load. Large load inrush currents increase the size of an inverter for a given installation. Figure 10 shows a list of the inrush current levels and the duration of the inrush currents for typical Saudi Aramco critical loads. The first column lists typical Saudi Aramco types of equipment. The second column lists the inrush current that is associated with the start up of a piece of equipment. The inrush current is listed as a percentage of full load current. For example, the amount of inrush current for an induction motor is 500 to 800 percent of full load current, or five to eight times the value of full load current, as shown in Figure 9. The third column in Figure 10 shows the typical time of duration of the inrush current for each piece of equipment. The inverter must be sized large enough to deliver the required inrush current for the required number of seconds or cycles.


24



Inrush Current Levels and Durations for Typical Saudi Aramco Critical Loads


25



Figure 10


26



Cr itical Load Gr owth Fa ctor The critical kVA load on a UPS system can change because of the addition of new equipment. A UPS installation is designed to last between ten and twenty years and must be flexible enough to support system growth without excessive oversizing of the inverter that causes the cost to rise. The base size of an inverter at the time of installation is equal to the combined kVA of the critical loads plus any allowances for inrush current and power factor. This base size should then be multiplied by a growth factor to determine the actual size inverter that should be installed. This growth factor will allow for system load growth without a prohibitive increase in inverter cost. The following growth factors that are based on the kVA of the critical loads should be used to account for future system expansions: Size of the Critical Load Under 50 KVA Over 50 KVA

Growth Factor 1.25 1.10

Harmonics A harmonic is a sinusoidal component of a periodic wave or quantity that has a frequency that is an integral multiple of the fundamental frequency. The creation of a 60 Hz sine wave output of an inverter will produce various harmonics of the 60 Hz signal. The effects of these harmonics on the load will vary with the construction and operation of the load. The need to remove or limit harmonics is based on the load's ability to withstand the harmonic content of the 60 Hz signal. Harmonics can produce harmful effects on traditional electric components, or modern static electronics. The high frequency harmonics that are carried through use of the 60 Hz signal will cause an increased heating of iron core devices. Also, certain relay devices will become more sensitive because of the skin effect of the higher frequencies. Sensitive static electronic devices can also overheat because of the high frequencies of the harmonic content of the 60 Hz signal. Many static electronic devices will develop data errors that are caused by the high frequencies of the harmonic contents.


27



To protect the critical load from the harmful effect of harmonics, a filter must be placed between the inverter source and the critical loads. The addition of a filter to remove the harmonic content of the 60 Hz signal causes an increase in inverter cost. A filter that removes the harmonic content is constructed through use of inductors and capacitors that are connected to the transformer secondary. The values of inductance and capacitance should be established to provide a total harmonic distortion of less than five percent and a single harmonic distortion of less than three percent. The source of harmonics is not always the power source (inverter). Certain types of loads will produce their own harmonics during operation. The harmonics that are produced by the load are called reflected harmonics. The filter that is installed to limit source harmonics will reduce reflected harmonics to help prevent damage to the inverter and the other loads of the system.


28



WORK AID 1: PROCEDURES AND TECHNICAL AND ECONOMIC REQUIREMENTS FROM SADP-P-103 AND ESTABLISHED ENGINEERING PRACTICES FOR DETERMINING THE APPROPRIATE TYPE OF INVERTER FOR USE IN SAUDI ARAMCO ELECTRICAL INSTAL LA TIONS Wor k Aid 1A: Pr ocedur e This procedure is for use in the determination of the type of inverter for installation in Saudi Aramco installations.

1.

Determine if the output impedance requirements of the system high or low. Determine which inverters will supply the necessary impedance.

2.

Determine the importance of the voltage and frequency regulation requirements of the load. Determine which inverters will provide the level of regulation required.

3.

Determine if the cost is a major concern in the selection of this inverter. Determine if the cost consideration overrides all other requirements.

4.

Select the least expensive inverter type that fits the requirements of the load.

Wor k Aid 1B: Technical and Economic Requir ements -

Standard three phase, ferro-resonant inverters are not permitted for use in Saudi Aramco applications.

-

Scott "T" transformer ferro-resonant inverters are used for the following conditions: -

-

High output impedance is allowed. Regulation requirements of the load are low. Cost is the prime concern.

Pulse width modulated inverts are used for the following conditions: -

Low output impedance is required. Close regulation of output voltage is required. Excess cost is justified by the need for higher levels of control.


29

UPS 3

Recommend Documents