|
|
|
DspEdu 2.1 2. And What is so Different About a DG Set ? 3. Harmonics in Plant Distribution Systems 4. Characteristics of Nonlinear Loads 4.1 Single Phase Uncontrolled Rectifier with Capacitance Filter 4.2 Single Phase Fully Controlled SCR Bridge with D.C Side Smoothing Inductor 4.3 Three-phase Uncontrolled Rectifier with Capacitor Filter 4.4 Three-phase fully controlled SCR Converter with d.c side Inductor. 4.5 Single-phase Versus Three-phase Nonlinear Loads 4.6 Other more troublesome AC-DC Converters which no one should make anymore ! 5. Consequences of Harmonic Currents and Voltages Generated by Nonlinear Loads 5.2 Effect on Capacitors and Motors 5.3 Generator Voltage Regulation Problems 5.4 Generator Speed Governor Problems 5.5 Instability of Thyristor Firing Controls 5.6 Some UPS Units may conclude that DG Set frequency is wrong ! 5.7 Some UPS-DG Set Specific Interaction Issues 5.7.1 UPS units do not like rapid and large frequency variations ... 5.7.2 DG Sets dislike the Inrush Power from UPS 5.7.3 DG Sets do not like those LC Filters at the input side of UPS Units 5.7.4 Automatic Transfer Switch may create problems... 6.1 Harmonics When the System is on Grid Supply |
Suresh
Kumar. K.S.
Department of Electrical Engineering National Institute of Technology Calicut Calicut-673601, Kerala State , India [This article was prepared for a short-term course on Standby Power Systems offered to Working Professional from Industry by Department of Electrical Engg., NIT Calicut in March 2004.The HTML version was uploaded to this server on 22nd March 2004 ] Many modern electrical power applications require continuous and high quality power. Standby Diesel-Electric Generating sets (DG Sets) are integral parts of the solution. Standby genset sizing requires an understanding of the genset characteristics and the connected load. Nonlinear loads, including uninterruptable power supply (UPS), variable frequency drives (VFD), adjustable speed drives (ASD), and switched mode power supplies present a special problem to such DG Sets and a full understanding of the deleterious interaction between Non-linear loads and DG Sets will be needed to choose and apply the right DG Set for the plant. Non-linear loads draw distorted currents from supply. Some non-linear loads also generate voltage harmonics at the point of connection i.e they distort the a.c voltage at their terminals directly.The non-sinusoidal currents flowing through the system impedance (which includes cable/OH line impedance , tranformer impedance etc) create distorted voltage drops across these impedances. The generated voltages are usually pure sinusoidal.With a sinusoidal source and non-sinusoidal impedance drops the load bus voltages get distorted. This indirect voltage distortion due to non-sinusoidal currents flowing in system impedances (in the case of all non-linear loads) and direct voltage distortion by some non-linear loads will be present always - i.e even when the plant is drawing power from Utility grid too. But they assume greater significance when the plant is on DG Set. 2. And What is so Different About a DG Set ? An AC Voltage
Source is a good source if (i) if its source impedance approaches
zero (ii) its frequency is a constant independent of everything (iii) its
voltage magnitude is a constant independent of everything and (iv) its
open circuit voltage waveshape is pure sinusoidal. A DG Set is bad on all
four counts. Utility Supply Point is much better on all the four counts.For
all practical purposes the frequency of Utility Supply may be considered
constant whereas it changes in response to every minor or major load changes
in the case of DG Set.And so does the terminal voltage of a DG Set. And
the source impedance of DG Set for sub-cycle (i.e for harmonic frequencies
and sudden transients) i.e the sub-transient reactance of a DG Set is usually
three to four times that of a similarly rated power distribution transformer.
For example a 500kVA 11kV/440V Distribution Transformer may have 4.3% leakage
impedance whereas a 500kVA G Set will have about 16% Sub-transient Reactance.
And that is really bad. If one wants to go for a DG Set which will have
same reactance as a 500 kVa transformer , it has to be 2MVA DG Set Unit
! No one would want to spend that kind of money !
3. Harmonics in Plant Distribution Systems The common sources of Harmonic Currents in Plant Distribution System are
4. Characteristics of Nonlinear Loads The front-end converter of a on-line UPS or Variable Speed Drive or Battery Charger or DC Drive or Rectifier Supply converts a.c line to d.c inside and delivers power to the subsequent power electronic stage. This front end can be of different types as explained below. 4.1 Single Phase Uncontrolled Rectifier with Capacitance Filter In this case the d.c side control ,if any , is exercised after the capacitor filter by means of a thyristor buck stage with smoothing inductor. The current drawn from the power line will be very much distorted and will be rich in third harmonic. Also the line power factor will be low. Hence this kind of front end is not advisable for units above 1kVA rating. One finds this kind of input converter in SMPS Units, Electronic Chokes , Single Phase UPS Units etc. A schematic diagram of this kind of front-end and the related waveforms are shown below.
Vs is the open circuit at the Point of Common Coupling (PCC) and Ls represents the short circuit reactance at the bus.The line current drawn by the system is harmonic rich with third harmonic being the dominant one.The THD of this kind of current may be as large as 130%.In addition to drawing a highly distorted current , this kind of rectifier also produces a kind of distortion in PCC Voltage called "Voltage Clamping". This is illustrated in the third plot in the figure above.When the capacitor voltage is above the source voltage the diodes do not conduct and the rectifier terminal voltage will follow the source voltage.When the diode conducts the rectifier voltage follows the d.c side voltage. At the end of diode conduction (which comes after the sine wave has gone over the peak) the rectifier terminal voltage suddenly starts following the sine wave source rather than the d.c output voltage.This accounts for the sudden dip in the rectifier terminal voltage as depicted in the figure above.The harmonic current drawn by the rectifier will cause harmonic distortion of voltage of voltages everywhere upstream ; but the PCC voltage is decided positively by the clamping mechanism described above. One solution to this voltage clamping distortion is to introduce a reactor between the PCC and the Rectifier input terminals.With a reactor in between , the PCC Voltage will face less clamping though the Rectifier termina voltage will be clamped.The ability of such a reactor to improve the PCC voltage will be obviously a direct function of the ratio - L/Ls - where L is the inductance of reactor added and Ls is the source inductance. And witha larger value of source inducatnce in a DG Set an existing reactor will be less successful in improving the PCC voltage clamping when the plant is on a DG Set. 4.2 Single Phase Fully Controlled SCR Bridge with D.C Side Smoothing Inductor Here phase control of a fully controlled thyristor single phase bridge is employed to control the current in a d.c side choke which feeds the d.c side load. The harmonic performance tends to be much better than in the case of uncontrolled rectifier. However, being a single-phase system, the line current drawn will contain third harmonic even in this case. This converter is preferred in the 1-2kVA range. Above 2kVA level ,a three phase input front-end is preferable from triple-n harmonic current and neutral overloading point of view.It is possible to use a d.c side smooting inductor in the case of uncontrolled bridge rectifier too ; performance will be similar to a fully controlled SCXR bridge with zero firing delay. This kind of AC-DC Converters do not cause voltage clamping at the PCC. But they bring in another serious kind of voltage distortion at the PCC - called "Voltage Notching".Of course they also draw harmonic currents ; but much less than the converter using a capacitor filter. Schematic diagram of the converter and related waveforms are shown in Figs. 2a and 2b below.
The waveforms show that this converter
does three things - (i) It draws a near square wave as its input current
(greater the value of smoothing inducatnce closer the current waveform
to square shape ) and thus the input current THD will be around 30%.However,
being a single-phase load, its input current will have triple-n harmonics.
How do these notches and ringing come up ? Consider the events in positive half cycle of input. The thyrisors X1 and X4 get triggered on only at 60 degree position. However the large d.c side smoothing inductor will keep the d.c side more or less constant always. This means that the other thyristor pair - X2 and X3 - must be carrying the current in the 0 to 60 degree ranfge in the positive half cycle.At 60 degree the two thyristors X1 and X4 get triggered on thereby applying reverse voltage across the previously conducting two thyristors X2 and X3. But then they have been conducting before and can not become blocking instantaneously - they can stop conducting only if the trapped charges inside them are swept out by a reverse current flow - till then they remain switched on. This results in two conducting thyristors putting a dead short across the converter terminals and that results in the notch in the voltage.The supply voltage gets shorted through supply inductance and the current in this shorted inductance ramps up thereby reversing the current in the line and transferring the cyrrent from X2-X3 pair to X1-X4 pair.The supply inductance and the snubber capacitors (or the device capacitances if there is no snubber) ring at the time of X2-X3 pair recover their blocking capability and that produces ringing in line current and converter terminal voltage. The solution again is line reactors in series with converter input. And the effectiveness of such line reactors will be compromised when the converter is fed from DG Sets due to higher source impedance level of DG sets. 4.3 Three-phase Uncontrolled Rectifier with Capacitor Filter Similar to 4.1 , but three-phase input. The line current will be distorted. But it will not contain third harmonic or any other triplen harmonic. Also, the neutral current will be zero. In fact there is no neutral connection anywhere in a six-pulse diode rectifier. This configuration suffers from the disadvantage of voltage clamping at converter terminals.The solution is line reactor as in the case of single-phase converter of similar type and such line reactors will not be as effective on DG Sets as in the case of Utility Supply. The schematic diagram and waveforms are shown below.This type of front-end is in use in VFDs of old design and in some rectifier units and battery chargers.
4.4 Three-phase fully controlled SCR Converter with d.c side Inductor. Similar to 4.2, but three-phase input. The line current distortion will be the lowest in this case (around 27 to 33% depending on load and smoothing inductance value). This is the most popular front-end design in UPS units , Adjustable Speed Drives , DC Drives , High Power Battery Chargers , Telecom Power Plants etc. Three-phase fully controlled SCR Convereters (i.e six pulse continuous conduction mode converters) are free from the voltage clamping problem. However they bring in the "voltage notch" and "ringing transients" back.The reasons for voltage notching and ringing have already been explained in the single-phase context.The solution again is line reactors in series with converter input. And the effectiveness of such line reactors will be compromised when the converter is fed from DG Sets due to higher source impedance level of DG sets. The schematic diagram and relevant waveforms are shown below.The waveforms show clearly that there are six notches per cycle in the phase voltage at converter terminals. Also note that the notches and ringing can generate many spurious zero-crossing events in the terminal phase voltages.Notch width and depth during commutation period are dependent upon supply system impedances, SCR firing angle, and load current.
4.5 Single-phase Versus Three-phase Nonlinear Loads In order to clarify various aspects of impact of nonlinaer loads on the distribution system a comparative study of various types of front-ends is carried out here taking UPS units as example systems.Assume that 15kVA UPS load is to be supplied from a DB which receives supply via a 35 sq.mm 31/2 core cable of 100m length from the main switch board.The UPSs are located 25m from the DB.Three cases are considered here. Case 1 – Three no. Single Phase Input 5kVA UPS units with uncontrolled rectifier type front-end connected to R, Y and B phases at the DB using three 2 core cables of 6sq.mm size and 25m length.(There are some UPS manufacturers who commit this kind of design-crime even now in India !) Case 2 – A three-phase uncontrolled rectifier type front-end UPS of 15kVA rating connected through 25m of 6 sq.mm 3 core cable. (There are manufacturers everywhere in the world who commit this design-crime even now when it comes to Variable Frequency Drives !) Case 3 - A three-phase uncontrolled rectifier with d.c side choke type front-end UPS of 15kVA rating connected through 25m of 6 sq.mm 3 core cable. In all the cases, the UPSs were assumed to deliver rated kVA at 0.8p.f at their outputs and their inverter efficiency was taken as 90%.The behaviour of the front end converters for the three cases were simulated using Pspice Simulation package and the line current and neutral current waveforms in the 3 ½ core 35 sq.mm cable supplying the DB are shown in Fig.5a.Also note that the return conductor current in the three single phase cables supplying the individual UPS units will be the same as the line current. The r.m.s line current is 34.2A and r.m.s neutral current is 59.2A.The line current is rich in harmonics and has a THD of 128%.Its third harmonic content is 90%.The triplen harmonic currents drawn by the three single-phase UPSs add in the neutral to yield a high level of neutral current in the three-phase cable-neutral current is as much as double the phase current! The total power taken is 14.3kW and the apparent power is 23.6kVA-thus the power factor is ~0.6. A 6 sq.mm cable has a current carrying capacity of about 35Amps.Here the r.m.s current is 34.2 Amps and hence the cables supplying the single phase UPS units appear to be fully loaded. But the current flowing in these cables have a THD of 128% and with such a large THD of current the cable has to be considerably derated. The rating is likely to be around 20A at this THD level. Hence the cable is grossly overloaded. 6sq.mm is capable of handling 8kVA per conductor if the current is undistorted i.e 24kVA of three-phase power. But in the present context it is going to be overloaded. A 5 kVA load requires only a 32A Switch-Fuse unit. But not a 5kVA Single-phase UPS! Such a UPS takes around 8kVA of distorted power and it takes 34.2 Amp of r.m.s current. A 32 A SFU will burn out. In fact the actual current taken by such a 5kVA UPS at full load will be more than 34.2A due to various converter losses which were ignored in the simulation. A 35-sq.mm cable has a current carrying capacity of 92Amps.In a 3 ½ core cable, the neutral conductor will have 46Amps capacity. And in the present case the neutral in the cable has to carry 59.2 Amperes of third harmonic current. In fact, due to increased in conductor resistance at higher frequencies, the current rating of neutral at 150Hz will be less than 46Amps.Thus there is going to be considerable neutral overheating, burning and cable damage in this 100m,35 sq.mm cable supplying three 5kVA single phase UPS units.35 sq.mm cable has 66kVA kVA carrying capacity-but not if the neutral is harmonically loaded. In this case we will require a 70sq.mm 3 ½ core cable to supply this UPS load if neutral overheating in the cable is to be prevented. Fig.5b above shows the current waveforms in the three lines supplying 15kVA three-phase UPS with uncontrolled rectifier/capacitor front-end and delivering 15kVA at 0.8 p.f at output. There is no neutral connection and no neutral current. Also, the triplen harmonic content in the line is zero. The r.m.s value of line current is 28.7A and its THD is 90%. The dominant harmonic is the fifth at 72%. The power drawn is 14.3kW and the apparent power drawn is 19.8kVA yielding a power factor of 0.72.Though the r.m.s value of current in the 6sq.mm 4 core cable is below its rated value of 35Amps, this cable is likely to be fully loaded at this loading condition due to the distorted nature of current. But in this case a 32A three-phase SFU will be sufficient unlike in the single-phase case. Also, there is no overloading of any kind on the 35 sq.mm 3 ½ core feeder.
Fig.5c Line Current Waveforms of a 15kVA three-phase UPS with Uncontrolled Rectifier + LC Filter front-end. Fig.5c above shows the line currents in the case of a 15kVA three phase UPS with uncontrolled rectifier plus Choke-Capacitor Filter. The r.m.s value is 22Amps and the THD is 28.4%. The power drawn is 14.3kW and apparent power drawn is 15.3kVA indicating a power factor of 0.93.Both the cables are being operated well under their rating. The table below gives the details of harmonic content of the currents in the three cases.
4.6 Other more troublesome AC-DC Converters which no one should make anymore ! These are the so -called Half Controlled Converters of single-phase and three-phase types. The single-phase half controlled converter uses two thyristors and two diodes instead of four thyristors.The three-phase half controlled units use three thyristors and three diodes.This saves some cost for the manufacturer ; but results in very high level of even harmonics (particularly the second and fourth harmonic ) due to lack of half-wave symmetry in current waveforms.And even harmonics are real bad things.They are difficult to filter. Many UPS manufacturers in India and this state of Kerala use this kind of input side converter in order to cut cost.They shouldn't be doing that ! 5. Consequences of Harmonic Currents and Voltages Generated by Nonlinear Loads Voltage waveform distortion magnitude
caused by the sinusoidal current demand of the nonlinear loads is a function
of the source impedance. Source impedance is not an easily defined value
in the case of a DG Set because generator reactance varies with time following
a sudden load change. Generator subtransient reactance (X"d) and subtransient
short circuit time constant (T"d) are primary parameters influencing distortion
during the short SCR commutation periods.
A standby generator
is characteristically of higher impedance than transformers. Significant
differences in kVA ratings of the two sources often contributes to greater
impedance differences. Utility transformers are frequently rated to carry
the total plant load. DG Sets are often only sized to carry emergency or
critical loads. Thus DG Sets may have 5 to 100 times greater subtransient
reactance than normal source transformers. Consequently, nonlinear loads
may work fine on utility, but may react entirely different when powered
by a DG set. Using an oversized generator to reduce reactance may be of
some benefit. However, to obtain a significant reduction in reactance is
not economically feasible.
Section 4 of this article clarifies that there are two kinds of voltage harmonics created in the systems by non-linear loads - first we have the harmonic currents demanded by the non-linear loads flowing through various system impedances thereby distorting the bus voltages upstream and secondly we have the "voltage clamping" and "voltage notching and ringing" type of voltage-source type distortions at the converter input terminals.The closer a particular bus is to a coverter bus , the greater the effect of voltage clamping and notching. In any case the net effect of non-linear loads is harmonic current flow everywhere and harmonic voltages everywhere. Harmonic currents produce high frequency flux change and cause heating in stator cores. Rotor losses also occur because harmonic currents in the stator will induce currents in the pole faces and amortisseur windings. Higher magnetic core temperatures result in a higher winding temperature. Generator stator heating is also a function of I 2 R loss. Winding heating is proportional to effective or RMS current squared. RMS current for a sinusoid wave is 1.11 times the average value. The RMS value of the distorted SCR circuit input current waveform is typically greater than 1.11 times average current. (See further discussion of instrument readings.) Derating or using a low temperature rise generator is a means of compensating for increased heat losses . Of course harmonic currents cause increased resistive losses everywhere , not only in the Generator windings. 5.2 Effect on Capacitors and Motors Increased harmonic voltage level during DG Set operation exposes the usual load i.e motors and capacitors to harmonic voltages.Motors can generally stand the resultant overheating. But capacitors will get overheated since their impedance goes down with frequency.It is true that the major power factor correction capacitors will be de-energised under DG Set operation.But the capacitors which are directly connected across the motors will not be de-energised and they may burn out or at least lose life fast. The inductive reactance level in the system is different when the system is on DG Set. Hence the parallel resonant freqencies of the power factor correction capacitor - generator reactance system will shif to lower levels and the risk of harmonic resonance should be studied.Best solution is to avoid having power factor correction capacitors on when the system is on DG Set. 5.3 Generator Voltage Regulation Problems The Automatic Voltage Regulator (AVR) of a DG Set maintains the terminal voltage of the set at a constant value under steady state conditions.It senses the terminal voltage , compares it with the set value and corrects any error by suitably chaging the field excitation current. The sensing part is the most important part. Many AVR designs sense the voltage across one phase ; some sense across two lines ; yet others sense all the three phases and average the measurements. Some AVRs calculate the rms value of the sensed voltage by using rms to dc converter electronics ; but they are few. Most of the designs use full-wave rectified average value and assume a form factor of 1.11 to convert that into rms.Obviously all these will result in DG Set voltage getting regulated at wrong level if there is considerable amount of harmonic distortion in the terminal voltage.Modern designs use proper filtering on the sensed voltages , sensing on all the three phases and true rms calculation (either using analog electronics or by using digital techniques in micro processors) to avoid voltage regulation problems in the case of DG Set serving non-linear loads. 5.4 Generator Speed Governor Problems But there is a more serious problem that can emanate from the excitation control unit of a DG Set delivering power to SCR Converters. The Engine Governor system of the DG Set needs a speed feedback signal.And this signal is usually generated by measuring the frequency of the sensed output voltage inside the AVR unit. And this frequency calculation usually involves zero-crossing information from the waveform.Normally the zero crossing events are supposed to occur at about 10ms interval (assuming 50Hz operation).But in the presence of voltage notching and heavy ringing transients from Thyristor Converter units there will be multiple zer-crossings within one cycle of ac waveform.This leads to large magnitude random errors in the frequency signal prepared by the AVR unit for use by the governor unit. And Governor unit gets confused and there results instability of the speed governing system.The solution is good filtering on the sensed voltage or locking onto fundamental component of the sensed voltage by analog or digital phase locking techniques. In short, the AVR unit of a DG Set has to be a specially designed one if that DG Set is going to handle predominantly non-linear load. 5.5 Instability of Thyristor Firing Controls It is possible that the waveform distortion caused by a Thyristor Converter system may not be tolerable to control unit of that system as well as other thyristor converters. This is not unusual for equipment that did not include considerations for a "limited bus" or high impedance source in its original design. It is, therefore, very important to advise the converter supplier of the existence of a standby generator set as a potential power source. The SCR must be triggered on to begin conduction. Output control of a AC-DC Converter is achieved by phase angle control, where delay angle before triggering is timed from the start of each half cycle or zero crossover. When the terminal voltage waveform is distorted zero crossover may move or there may be several zero crossings, causing erratic firing from one cycle to the next. Timing circuits must be designed to respond only to the 50 Hz fundamental output of the DG Set to avoid this problem. Filtering of the trigger circuits is possible and should be included by the converter manufacturer in original design. 5.6 Some UPS Units may conclude that DG Set frequency is wrong ! Some UPS systems are also equipped with over/under frequency relays which may respond to a distorted waveform (due to multiple zero crossing and/or erratic zero crossings) and cause the UPS systems to reject the input source even though fundamental frequency is within tolerance. Filtering the sensing signal to these relays may correct this problem. 5.7 Some UPS-DG Set Specific Interaction Issues 5.7.1 UPS units do not like rapid and large frequency variations ... A Utility Power System can be thought of as a large Synchronous Generator with low reactance and huge inertia. A huge inerta machine does not change its speed easily. Therefore the Utility frequency does not vary mush and when it varies it does not vary fast. A DG Set , in comparison , is a small inertia machine and consequently its frequency changes over comparatively large range whenever some load change takes place. And DG Set frequency varies rapidly. Of course if the load is constant the Governor System will bring the frequency back to its normal value. But Governor systems are never fast enough (neither should they be ; otherwise there can be instability problems in the Governor loop) to suppress frequency variations due to rapid load changes. Hence , obviously , intermittent loads (like Air Conditioners , Pulsed Heater Controls etc) are bad loads on a DG Set. UPS units generally synchronise to the mains input (except in low kVA range ; even there synchronising is needed if a synchronised transfer to bypass is required). This synchronising is done by some form of Phase Locked Loop(PLL) system. And all PLL systems have two limitations - (i) they can remain in lock only if the input frequency remains within a narrow range around the nominal frequency value (ii) they can remain in lock while the input frequency is changing only if the input frequency is changing sufficiently slowly i.e the rate of change of frequency (so-called "frequency slew rate") is less than some critical value.When the PLL loses lock the UPS refuses to take the input supply and switches over to battery , reporting some under/over frequency condition. Now , if the UPS is the last load that comes on on the DG Set it may succeed in continuing on the DG Set.But if UPS is put on first and then some other heavy load (for e.g some large motor starting) comes on the DG Set , the DG Set voltage and frequency dip down and UPS moves over to battery either due to low voltage condition or due to low frequency or due to PLL losing lock due to fast rate of change of frequency. DG Sets can have their frequency slew at 3 to 5 Hz/sec and that will be too much for the UPS PLL systems unless the UPS is specifically designed for that.Even if a UPS is designed with a wider frequency window and larger tolerance band for frequency slew rate , it is better not to risk it - it is better to avoid large switching loads like A/C units with On/OFF control , Heater load with thermostatic ON/OFF control etc on the DG Set which supplies a large UPS load. 5.7.2 DG Sets dislike the Inrush Power from UPS If a UPS Unit is used only to tide over the first few minutes of mains power loss i.e till the AMF system of DG Sets bring them up , it is quite likely that the Battery Units have low AH capability. In that case the battery would have run down considerably by the time the DG Set comes on.Thus the UPS will suddenly transfer all the output load power and the full battery charging power to the DG Set when DG Set comes on.If the UPS load is only a small percentage of the DG Set rating it may be OK. But if the UPS load is the major load , sudden inrush power of this major load will cause the voltage to take a swing and the frequency to go down rapidly.And the UPS goes back to battery due to low voltage or due to high rate of change of frequency.It may have to be reset or it may attempt transfer automatically after a delay. But in both cases the same thing will happen again. In short , the DG Set will appear to be refusing to take load. The solution lies in arranging a "Power Walk-in Feature" in the UPS Design.This is a design technique where the UPS slowly increases its power demand from the input supply immediately after transfer. UPS slowly ramps up the input power (usually over a period of 20 to 40 seconds) from 0 to 100% thereby avoiding sudden voltage and frequency changes and fast slew rates of frequency.During the ramping-up period the output of UPS is met partly from the Battery and partly from the mains. If the total UPS kVA in a system is more than 50% of the DG Set Rating it may be desirable to have this "Power Walk-in" feature in the UPS Units. 5.7.3 DG Sets do not like those LC Filters at the input side of UPS Units UPS Units in the 50kVA to 500kVA range generally use 6-pulse SCR converter front-end and handle the harmonic currents generated by this stage by adding some kind of harmonic filtering inside.This filtering maybe of two types. In the first design (which is the popular one) a passive LC Filter structure is connected in parallel to mains input.This series LC combination is tuned to the dominant harmonic (the fifth) thereby providing a shorted path for the fifth harmonic current taken by the 6-pulse converter.Thus the mains line current will have lesser harmonic content (< 10% usually). Now this LC Filter is a short at fifth harmonic and will attract all fifth harmonic currents -even those generated by other loads - into it ! Also , if the supply voltage has fifth harmonic in it for some reason , this over-enthusiastic filter is going to short it out ! Both are bad. So we need to isolate this filter from rest of the world.This is done by a 5% inductance in series with the mains input line before the line gets to the input side converter. The SCR 6-pulse converter will also consume reactive power at fundamental frequency.It would be convenient to arrange the L and C value in the LC filter such taht the filter will behave like a Capacitance at 50Hz and will absorb the right value of leading Vars to make the entire unit behave like a UPF load.That is the way the LC filter is designed and the reactive power drawn by the LC Filter is matched to the Converter kVAr at full load. Now we have to distinguish between two cases. Let us assume that the UPS is the major load on the DG Set. And precisely because of that this UPS will have "Power walk-in" feature. So when the mains supply from DG Set comes on UPS input gets connected to it , but it starts at low power - now we have the classic problem of an unneeded capacitor across the DG Set terminals. the capacitor comes from UPS input side filter.And since UPS is taking only small power to begin with, the DG Set is effectively on no load and it has a large capacitor across it - over voltage is the result. And , more seriously , the large leading VARs taken by the LC filter in the UPS may cause the AVR to reduce the excitation current to minimum or zero leading to instability in the DG Set.In any case the UPS will sense over voltage and trip its input breaker and promptly go to the battery.Again the DG Set will appear to be refusing to accept load. If the UPS is not the major load and/or it is brought in only after some other reactive power consuming load is put on the DG Set , the LC Filter may not create any problem. The solution is to modify the UPS design such that the LC Filter is disconnected when the UPS is on the Battery and connect it again only after the "Power Walk-in" is completed successfully. UPS units of recent design from reputed manufacturers have this feature. Passive LC filtering is not the only possible filtering. Some recent designs use "Active Harmonic Filtering" instead of passive filtering.The required components may be built into the UPS unit itself or it maybe available as an external add-on.In Active Harmonic Filtering , a PWM Inverter running from self-generated DC Bus works as a current source to inject a current that is equal in amplitude and phase to the harmonic current taken by the UPS front-end converter.Thus , this unit will supply the harmonic current required by the UPS front-end and UPS will draw more or less pure sinusoidal current from the mains.With this kind of filtering there are no heavy reactive power components and hence the problem of over-voltage due to the Capacitor in LC filter does not occur here.UPS units with integral active harmonic filter or external active harmonic filter are available from manufacturers like MGE. 5.7.4 Automatic Transfer Switch may create problems... Most generator-UPS installations include automatic transfer switches that switch the UPS back to utility power once it becomes available again. There can be some problem if the transition is open transition i.e the DG Set is taken off the bus first and then the mains breaker is closed. Now the UPS will sense a mains failure , go to battery mode and will start a "power walk-in" when the bus is connected to utility power.Thus the UPS will behave like a large capacitor (from LC filter) and near zero active power during the "open" period of 'open transition' and during the first few seconds of power walk-in. During the small interval when there was no connection to the bus (i.e DG set breaker was off and utility breaker was not yet on) the UPS LC filter capacitor will provide excitation currents to the motors running from the same bus.That turns the motors into Induction Generators and over-voltages may result. Moreover ,now when the Utility breaker is closed there is 'synchronising problem' with the induction generator voltages. If the UPS Input Filter Capacitor were not there , the rotor circuit flux decay in Induction Motor would have ensured that the back emf at the terminals of motor is near zero when the utility breaker closes.This problem is solved by by providing a fast means of detecting the transfer in the UPS and disconnecting the filter.Only a few UPS designs are available with this feature. The subsequent sections of this article deal with the observed current and voltage waveforms at an Electronics Research Organization.This establishment received Electrical Power from K.S.E.B Grid (the Kerala State Electricity Board) through a 11 kV/415V, 250 kVA Transformer with a Contract Demand of 160kVA.The total Connected Load at the time of audit was 215 kW+19.5HP+58 kVA+50kVAr. Two 100 kVA DG Sets were installed as Standby Supply.The major energy consuming loads were: Flourescent Light fittings in the office area and Computer Rooms & Labs, Ceiling Fans, Pedestal Fans, Table Fans, Wall Fans, 32 Split A/C units of 2T each amounting to 64T of refrigeration, Water Coolers, two 5 HP pumps, 9 HP of workshop machinery Computers and terminals, Printers and 43 kVA of UPS load.Though the contract covered only Energy Audit a cursory Power Quality Audit was also performed as a part of Energy Audit since the system voltage showed considerable amount of distortion especially when the system was running on DG Set. This section presents the salient data and findings and contains extracts from the Audit Report. There is about 43kVA of Single Phase UPS Load in the form of 1kVA, 2kVA, 3kVA and 5kVA Single Phase UPSs used in various laboratories and Computer Centres in the System. All UPS units in the system are of Single-Phase input type and the front end of all the units are of phase controlled thyristor converter type.These UPS units collectively consume about 22kW of power from the mains and contribute to prominent harmonic distortion of line currents in the System. Heavy distortion of currents drawn result in harmonic distortion of Voltages too, and prominently so when the System is on the 100kVA DG Set. 6.1 Harmonics When the System is on Grid Supply Single Phase UPS units draw a current which is rich in odd harmonics-the predominant ones being third, fifth and seventh. The third and all other triple-n harmonic currents add in the neutral conductor irrespective of distribution of UPS loads among the three phases. If the current drawn from a transformer has more than 15% T.H.D , the transformer losses increase in a significant manner. However, the resultant harmonic distortion in voltages will be marginal due to small short circuit impedance of the transformer. The waveforms shown below in Figure 6.1 illustrate these points. These waveforms show the phase voltage and line currents at the transformer secondary at the Establishment recorded on 10-3-1999 at around 2:00PM. These waveforms were recorded one at a time using Tektronix THS 710A DSO and hence do not reflect the correct phase relationships between various quantities.
The line currents are heavily distorted due to the single-phase UPS loads. The THD in R Phase current is 14.5%, that of Y Phase is 16.2% and that of B Phase is 28%. The phase voltages are more or less clean and have THDs less than 5%. The R phase current of the 50kVAr capacitor shows a THD of 15.8% and shows a predominant component at 11th harmonic (See Figure 6.2 below).
A detailed analysis of current waveforms coming from various feeders were carried out and it was found that there is no load that contributes this much of 11th harmonic in the system. Hence this predominant 11th harmonic in capacitor current and transformer current was seen to be the result of partial harmonic resonance between the 50kVAr capacitor at the main bus and the leakage reactance of the 250kVA transformer. Calculations using 4.52% leakage reactance value (found from the nameplate of transformer) reveal that a 50kVAr capacitor can resonate partially at 11th harmonic with this transformer. This partial resonance stresses the capacitor and transformer in addition to the extra stress brought in by the harmonic currents demanded by the non-linear loads. 6.2 The Neutral Current - Grid Supply The neutral current at the transformer secondary is shown in the waveform below in Figure 6.3. It shows a fundamental component and a third harmonic component that is more than even the fundamental component. This third harmonic is exclusively the result of single-phase UPS loads. The r.m.s value of this neutral current is 68 Amps and is almost as much as the r.m.s value of phase currents indicating heavy neutral loading.
6.3 Harmonic Content When the System is on DG Set However, the harmonic distortion picture changes considerably when the system is run on one of the 100kVA DG Sets. Even small degrees of harmonic distortion in generator currents will result in large amount of harmonic distortion in generator voltages through the large sub-transient impedance of the generator. Also harmonic currents cause increased losses in stator and field windings of generator. The harmonic distortion in phase currents and phase voltages in the DG Set were measured during the audit on 10-3-1999 and 11-3-1999 and were found to be excessive (See Figure 6.4 below). Waveform observation confirmed sizeable distortion in currents and voltages. It is well known that a distorted voltage results in loss of capacity of induction motors and overheating of motors. Also distorted voltage has been identified as a major reason for frequent failures of power factor correction capacitors connected across the motors. [Note-These waveforms were recorded one at a time and hence do not reflect proper phase relationships. They can be used only for harmonic analysis and not for power analysis or power factor analysis.]
The generator output voltage shows prominent distortion and ringing due to large harmonic current flow through its winding impedance. The Generator neutral current has a large third harmonic component that is even more than the fundamental amplitude (See Figure 6.5).
Switching on the 50kVAr capacitor at the main bus was found to result in a relatively clean output voltage waveform from the generator. This happens due to the filtering action of the capacitor. All the harmonic demand of the load goes through the capacitor instead of through the large synchronous reactance of the generator. This will result in a clean voltage waveform at the expense of overheating and possible burnout of the capacitor. Unequal third harmonic currents in the three phases of the generator brought about by unbalance in UPS load among the three phases will result in third harmonic appearing in line to line voltages. Third harmonic content in line voltage will result in circulating currents in delta connected motors which will go undetected by the overload relay. Also, third harmonic in line voltages interferes with AVR of generator and voltage sensing circuits of servo stabilisers etc. |
|||||||||||||||||||||||||||||||||||||
|
|