您所在位置： &
&nbsp&&nbsp&nbsp&&nbsp
电冰箱维修案例.doc 24页
本文档一共被下载：
次 ,您可全文免费在线阅读后下载本文档。
下载提示
1.本站不保证该用户上传的文档完整性，不预览、不比对内容而直接下载产生的反悔问题本站不予受理。
2.该文档所得收入(下载+内容+预览三)归上传者、原创者。
3.登录后可充值，立即自动返金币，充值渠道很便利
需要金币：350 &&
电冰箱维修案例.doc
你可能关注的文档：
··········
··········
1:冰箱白天制冷时出现频繁“跳机”现象，到了晚上制冷效果正常
晚上制冷效果正常，说明制冷系统和电路系统工作正常，这说明是机外原因造成的，原因是冷凝器散热不良。 ?冷凝器散热不良原因有：
冷凝器表明太脏；
（2）系统内有空气；
（3）周围空气不流动；
（4）周围环境温度太高；
（5）冷凝器内表明结油膜。 ?
该冰箱晚上制冷效果正常，这就排除了上述原因中的（1）、（2）、（3）和（5）。将冰箱移到别处便不再“跳机”。由此判定原因是冰箱靠近热源，白天餐馆营业，冷凝器散热不良，造成冰箱“跳机”，晚上餐馆停止营业，冷凝器能正常散热，所以冰箱可以正常制冷。将冰箱摆放到远离热源且通风的地方，故障排除。
电冰箱压缩机的外部和内部检修方法：
1电冰箱压缩机的外部检修法：
冰箱压缩机发生内堵时，可先拆开管路，用空气压缩机的排气管将高压空气压入压缩机的高压腔，再由高压腔进入低压腔，使积累杂质脱离阀片，即可达到清除的目的，该法在内堵不严重时有效。
冰箱压缩机发生抱轴且确定绕组正常时：A；将高压管和一个低压管堵住，再由另一个低压管向各压缩机注满冷冻油，然后将管口封死，接着用橡皮锤沿压缩机的外壳用力敲击，直至抱轴微微松动，于是机油便可浸入抱点润滑，此法可重复进行直至故障排除。B；接上述方法给压缩机加油后，将压缩机放入烤箱内，温度在80左右，保持30min即可通电试机。C；将压缩机启动元件拆开，换入50uF左右的电容，再多次启动压缩机，使压缩机在大电流冲击下，抱轴发生松动。
2冰箱压缩机的内部检修法：
若压缩机绕组不良，应先从整机上拆下定子绕组，拆卸时应记录好绕组周长，匝数，跨距，线径，并用优质耐氟线绕制。
（2）若阀片，活塞/气缸/转轴/缸体发生机械故障，其中损坏或磨损的均应换新
3:如何判断毛细管“”：毛细管脏堵的原因主要是制冷系统不清洁所致。毛细管脏堵有两种情况：一种是微堵：冷凝器的下部分聚集大部分液态制冷剂，流入蒸发器的制冷剂流动声明显减少，蒸发器内部只能听到“”的过气声；有时也能听到一股一股的制冷剂流动声，蒸发器结霜时好时坏。另一种是全堵：蒸发器内听不到制冷剂的流动声，蒸发器不结霜，若将毛细管与干燥过滤器连接处剪断，制冷剂喷出，即可判断为毛细管“脏堵”。
4:防止压缩机阀板/阀片产生碳化物的方法：
压缩机试机时间过长时，高压腔内要除去积水。
久置于空气中的制冷系统抽真空的时间应相对延长，一般为8H以上。如果条件允许，最好一边抽真空，一边为蒸发器/加热器加热。
改变抽真空的方法，将低压加液管抽真空改为从高压端过滤器抽真空，效果更佳。
制冷系统检漏或清洗时，应采用氮气或R12，不要用空气。
清洗和干燥时，清洗剂最好用煤油，不要用汽油，可减缓锈蚀现象的发生。
禁止往制冷系统中加甲醇。
如何判断电冰箱压缩机阀片碳化；电冰箱阀片，阀板发生碳化的主要原因有：（1）零部件清洗，干燥和防锈不彻底或致冷剂中有杂质。（2）制冷系统真空干燥不彻底或制冷剂R12含水量太高，水与R12起化学反应，生成盐酸，进而在制冷系统中腐蚀漆包线和铝等金属，腐蚀物一旦粘在压缩机阀片和阀板上，即产生碳化现象。（3）有些维修从员为防止或消除冰堵故障，修理时向制冷系统加入少量甲醇，导致化学反应，产生固体醇铝
正确维修冰箱的制冷系统：用HFC134a制冷剂替代CFC12（R-12）的应用技术日趋成熟。1；压缩机不能互换，HFC134a又名四氟乙烷，热性能与CFC12十分相近，不可燃，无毒无味，使用安全。与R-12相比，HFC134a蒸发压力较低，冷凝压力较高，压比更大。低压比时制冷量小，制冷效率较低。为获得相同的制冷量，故需对压缩机的构造，材料做部分改动。包括漆包线绝缘材料的选用，以及内装的润滑油，以便获得良好高效的实际工作效果。因此不能与使用传统CFC12制冷剂的电冰箱压缩机互换。2；采用特定冷冻油：压缩机冷冻油必须与制冷剂相溶，并且要有良好的润滑性，密封性，低温流动性及化学稳定性，由于HFC134a压缩机一般采用与之相溶性好的脂类或聚二醇（PAG）油，与CFC12压缩机所用的矿物油不同，不能互相代替，否则不仅不能满足压缩机的润滑要求，而且还有可能凝固堵塞制冷系统，故切忌混用。维修抽空时，真空泵必须更换为脂类冷冻油的真空泵。3；干燥过滤器不能替代；对CFC12系统而言，选择XH-5型干燥过滤器便能吸收水分。而对HFC134a系统而言则应选用吸水性更强，体积更大的XH-7型干燥过滤器，因二者所选用材料不同，维修中不能互相代用。4；维修操作时间应尽量缩短：HFC134a电冰箱制冷管路要求超净度高，制冷系统的抽空时间比CFC12长，以确保真空度不高于60Pa。维修实践中常采用二次抽空的方法，即第一次抽空5分钟，充注10Ghfc134a制冷剂，然后再抽空二十分钟，这样即可缩短抽真空时间，效果也
正在加载中，请稍后...欢迎光临智德教育文库，如需获取更多资料请使用搜索功能。
【图文】Conducted EMI Issues in a 600W
578IEEE TRANSACTIONS ON INDUSTRY APPLICATION, VOL. 36, NO. 2, MARCH/APRIL 2000Conducted EMI Issues in a 600-W Single-Phase Boost PFC DesignLeopoldo Rossetto, Member, IEEE, Simone Buso, Member, IEEE, and Giorgio Spiazzi, Member, IEEEAbstract—This paper presents the results of experimental activity concerned with the development of a 600-W boost powerfactor corrector (PFC) complying with the EMC standards for conducted EMI in the 150-kHz–30-MHz range. In order to accomplish this task, different circuit design and layout solutions are taken into account and their effect on the conducted EMI behavior of the converter is experimentally evaluated. Common-mode and differential-mode switching noise, together with input filters' design and topology and with the printed circuit board layout (in terms of track length and spacing, ground and shielding planes, etc.) are the key aspects which have been considered. In particular, the paper reports the conducted EMI measurements for different filter capacitor placements and values, for different power switch drive circuits, together with several other provisions which have turned out to be decisive in the reduction of the generated EMI. Index Terms—Conducted EMI, power-factor-corrector rectifier, printed circuit board layout.I. INTRODUCTION HE employment of boost power-factor correctors (PFC's) in order to comply with the IEC
low-frequency EMC standard [1]–[3] is becoming more and more ordinary in a large variety of industrial applications of switch-mode power supplies (SMPS's). This solution, however, increases the conducted interference generation of the power supply in the high-frequency range. As a consequence, while the low-frequency harmonic content of the current driven from the utility grid is normally well controlled and compliant with the aforementioned IEC standard, the high-frequency currents generated by the converter on the grid may be beyond the corresponding standard limits [4]–[7]. To avoid this, it is very important to properly design the EMI filters and the circuit layout so as to minimize the effects of the switching converter on the line pollution. This paper discusses the design of a 600-W boost PFC complying with the EMC standards for conducted EMI in the 150-kHz–30-MHz range [8]. Different circuit design and layout solutions are taken into account and their effect on the conducted EMI behavior of the converter is experimentally evaluated. Common-mode and differential-mode switching noise, together with input filters' design and topology and withPaper IPCSD 99–79, presented at the 1998 IEEE International Telecommunications Energy Conference, San Francisco, CA, October 4–8, and approved for publication in the IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS by the Industrial Power Converter Committee of the IEEE Industry Applications Society. Manuscript submitted for review November 3, 1998 and released for publication November 5, 1999. The authors are with the Department of Electronics and Informatics, University of Padova, 35131 Padova, Italy (e-mail: pel@dei.unipd. simone@dei.unipd. spiazzi@dei.unipd.it). Publisher Item Identifier S )02418-X.the printed circuit board (PCB) layout (in terms of track length and spacing, ground and shielding planes, etc.) are the key aspects which have been considered. In particular, the paper reports the conducted EMI measurements for different filter capacitor placements and values, different power switch drive circuits, together with several other provisions, which have turned out to be decisive in reducing the conducted EMI level of the converter [9]. By means of this design example, which employes two-layer PCB technology, the paper also shows that the application of the theoretically derivable EMC basic design rules, which, in principle, should guarantee the limitation of the EMI in a switching power converter, may, in some cases, become partially ineffective because of second-order effects (e.g., resonances, component parasitics, connections). The experimental results illustrate these unexpected outcomings and the validity of the adopted provisions which allow one to design a fully compliant power supply. II. BASIC SCHEME OF THE CONVERTER Fig. 1 shows the basic scheme of the considered boost PFC. The ratings of the converter are reported in Table I. These represent the typical characteristics of a PFC designed for a large variety of applications (e.g., telecom applications). A conventional and simple design procedure can be adopted to derive the necessary passive components' values, required to guarantee the continuous conduction mode of operation for the converter practically during the whole line period and a suitable output voltage ripple level. Also, the selection of the required switch and diode is almost straightforward, given the current and voltage stresses, which are easily determined analyzing the converter's typical waveforms. The resulting list of adopted components is reported in Table II. III. CONSIDERATIONS ON THE POWER STAGE DESIGN The considered topology is simple and well known [10]–[13]. However, when it comes to EMI control, it is necessary to adopt particular care in the definition of the layout of the power stage [14]–[18]. Of course, the main sources of EM noise can be easily identified in the power switch and diode. The reduction of the wire circuit lengths for the current return paths and for the high branches, together with the reduction of the areas embraced by loops, as shown in Fig. 2, appear to be fundamental high provisions. It is, therefore, fundamental that the area between the power switch and the two high-frequency bypass capacitors, which are used to drain the current pulses generated during theT/00$10.00 (C) 2000 IEEEROSSETTO et al.: CONDUCTED EMI ISSUES IN A 600-W SINGLE-PHASE BOOST PFC DESIGN579Fig. 1. Basic scheme of the boost PFC.TABLE I CONVERTER RATINGSFig. 2. Critical points of the circuit for EMI generation.TABLE II MAIN CONVERTER COMPONENTScommutations, is made as small as possible. Indeed, if the emitting area is small, the efficiency of the equivalent loop antenna, indicated in Fig. 2, is small, too. As a consequence, the amount of radiated noise which can couple with the circuit conductors, thus becoming conducted noise, also will be minimized. This is the reason why, even if the major concern of this design is the minimization of conducted EM noise, it is important to reduce the efficiency of the radiating sources as much as possible. In order to do that, another important practical provision is to twist, if possible, all the critical tracks of the PCB, and wind the wire of the toroidal inductor used in the converter power stage as shown in Fig. 3. This keeps the emitting areas as small as possible (introducing also a mutual cancellation of the fluxes), without compelling one to excessively shrink the magnetics and the size of the PCB, which is normally difficult and sometimes impossible. All of these considerations have a quite relevant effect on the design of the PCB of the converter. As shown in the upper part of Fig. 4, which represents the solder layer of the PCB, the tracks between the high frequency bypass capacitors and the switching components are kept as short as possible and twisted according to what has been sketched in Fig. 3. Moreover, this provision helps to extend the effectiveness of the capacitive filter to higher frequencies, by allowing one to increase the frequency of the resonance between the capacitors and the track stray inductance as much as possible. In order to evaluate this frequency, the total stray inductance of the switching loop can be estimated to be about 1 nH/mm (diode and capacitor lead distance must be included in the computation).Fig. 3. Reduction of radiating areas by means of twisted current paths (a) in the toroidal inductor and (b) in the tracks of the PCB.Observing the PCB layout given in Fig. 4, it is possible to see that the track connected to the power switch drain (the central lead of the MOSFET), which exhibits very fast voltage variations during the switch commutations, is kept very short and shielded by means of two constant voltage tracks and and the ground-connected converter heat sink on the solder layer, and a shield plane connected to ground on the component layer, so as to minimize the generation of radiated EMI. Also, the gate circuit, which exhibits an high peak current, must be accurately designed according to the previously discussed guidelines. As already mentioned, the key factor is the area embraced by the circuit which must be as small as possible. This is particularly critical when it comes to the insertion of a suitably designed RC snubber circuit. The snubber is highly recommended mainly to slow down the MOSFET turn-on, thus reducing the power diode recovery current. As a consequence, the turn-on peak current in the MOSFET is also reduced and, even if the switching time is increased, the converter efficiency is not heavily affected. Moreover, any ringing in the gate circuit can be greatly am therefore, the adoption of the snubber is advantageous also to damp such oscillations. The effects of this provision will be illustrated by experimental waveforms. Finally, the control circuit must be effectively protected against the disturbances by reducing the length of the sensitive tracks as much as possible. This is particularly important for580IEEE TRANSACTIONS ON INDUSTRY APPLICATION, VOL. 36, NO. 2, MARCH/APRIL 2000TABLE III MAXIMUM EXPECTED VOLTAGE NOISE DERIVED FROM CURRENT RIPPLE HARMONIC COMPONENTSFig. 6. Line voltage (100 V/div) and current (2 A/div). Horizontal scale: 2 ms/div. Fig. 4. PCB of the converter: solder layer (top) and component layer (bottom). The size of the board is 150 mm 100 mm.2point of a single star connection of the ground tracks, needed to avoid ground loops involving sensitive parts of the circuit (control circuit, gate circuit). The design of the power stage can be completed by determining the common mode and differential mode input filters [15], depicted in Fig. 5, which represents the complete converter circuit. The input filter for the differential mode noise, which is a third-order filter, can be designed calculating the maximum value of each harmonic component of the current ripple. The current ripple maximum amplitude is given by A (1)Fig. 5.Schematic of the converter including input and output EMI filters.the analog inputs of the control implementing the feedback loops for the regulation of the output voltage and input current. As shown in Fig. 4, the corresponding tracks must be short and shielded by means of tracks exhibiting a stable voltage level and a low impedance to ground. In the component layer of the converter's PCB, which is shown in the bottom part of Fig. 4, a shield plane is implemented which extends under the power switching components and is connected to ground in a single point which is the MOSFET source. It is important to notice that this is the centerAssuming the ripple waveform to be triangular and varying the duty cycle, it is possible to find the maximum amplitudes of the harmonic ripple components which, of course, take place at different duty-cycle values. In order to design the filter, we assume that the input current ripple flows through the line impedance (100 ), which is actually the equivalent differential mode impedance of the line impedance stabilizing network (LISN) requested by the standards. The expected noise amplitude, calculated at different harmonic frequencies without the input filter, is reported in the third column of Table III.ROSSETTO et al.: CONDUCTED EMI ISSUES IN A 600-W SINGLE-PHASE BOOST PFC DESIGN581Fig. 9. Conducted noise for the boost PFC (peak measurement).Fig. 7. Gate circuit snubber effect at turn-on. Top: turn-on without snubber (v 100 V/ v = 5 V/ horizontal scale: 50 ns/div). Bottom: turn-on with R–C snubber (v = 100 V/ v = 5 V/ horizontal scale: 100 ns/div).=Fig. 10. Effects of additional common-mode capacitive input filter insertion (peak measurement).Fig. 8. Switching frequency modulation effect on v Upper trace: modulation on. Lower trace: modulation off.voltage spectrum.Based on these values, it is possible to determine the required filter attenuation to comply with the standard. The attenuation must be suitably oversized (at least 10 dB V) to cope withpossible resonances due to circuit parasitics and with the additional contribution of the common mode component of the conducted noise. Considering again the 100- load for the filter, it is then possible to choose the values of the differential mode inductors and of the capacitors of the filter. A standard mH has been common-mode inductive filter employed, which presents a value of the differential mode H. The value of the line side filter inductance capacitor is consequentely selected to be equal to 1 F, while that of the rectifier side capacitor is 2 F split across the diode bridge rectifier. Expected filtered noise is then given in the fourth column of Table III. It is worth noting that the switching frequency has been selected so as to have the first and second harmonic component582IEEE TRANSACTIONS ON INDUSTRY APPLICATION, VOL. 36, NO. 2, MARCH/APRIL 2000Fig. 11. Effects of additional common-mode capacitive output filter insertion (peak measurement).Fig. 13. Effects of heat-sink connection to the negative voltage rail not close to the MOSFET source (peak measurement).possible to the major source of common-mode noise. For this reason, the filter capacitors are placed on the converter PCB (see Fig. 4). Their value, which must be lower than a given limit for safety reasons, depends on the parasitic capacitance of the circuit layout and, in this prototype, has been selected to be 3.3 nF. Another important factor in reducing the generation of EMI in a switching converter is the modulation of the switching frequency [19], [20]. In developing the prototype, the shifting of the switching frequency has been simply implemented by adding to the ramp generator circuit of the controller a periodic signal derived from the rectified input voltage. In this way the modulation frequency is no longer fixed, but is modulated at 100 Hz around its nominal value (70 kHz). It is worth noting that the disturbance signal is added to the ramp generator so as to get the minimum switching frequency when the commutated current is maximum, thus obtaining also a little improvement in the converter efficiency.Fig. 12. Effects of shield insertion between MOSFET and heat sink. The shield is connected to the MOSFET source (peak measurement).IV. EXPERIMENTAL TESTS ON THE PROTOTYPE The converter prototype has been extensively tested to reveal potential sources of EMI and to test possible solutions before the final measurements have been done. As expected, the low-frequency behavior of the PFC is very good, as can be seen in Fig. 6, where the good proportionality between line current and voltage can be appreciated. Fig. 7 describes the effect of theof the ripple below the lower frequency considered by the standards [4], [5], thus reducing the filter requirements. Lastly, as shown in Fig. 5, a common-mode capacitive filter is connected to the output side of the converter, that is as close asROSSETTO et al.: CONDUCTED EMI ISSUES IN A 600-W SINGLE-PHASE BOOST PFC DESIGN583snubber adopted for the gate circuit. As can be seen, the insertion of the snubber reduces the speed of the commutation and voltage ringing at turn-on (note that the time base is difthe ferent for the upper and lower parts of Fig. 7). Indeed, as shown voltage peak corresponding in the upper part of Fig. 7, the to the conduction of power diode recovery current (observe also voltage drop during this phase) summed to the inductor the current, is almost totally removed in the bottom figure, where the gate voltage sets to the level corresponding to the inductor current without appreciable ringing. This, as will be shown in the following, strongly improves the high-frequency behavior of the converter. The effect of the previously described switching frequency voltage spectrum modulation is shown in Fig. 8, where the is depicted. As can be seen, the effect of the switching frequency modulation is to modify the structure of the signal's spectrum. As expected, the spectrum is no longer composed of definite lines located at the multiples of the switching frequency, but is made up of large and flat bands, the biggest centered around the nominal switching frequency. The harmonics of the switching frequency are, indeed, almost totally eliminated. This turns out in a very relevant improvement of the high-frequency behavior of the converter.Fig. 14. Effects of gate snubber disconnection (diode reverse recovery occurs) with switching frequency modulation (peak measurement).V. EMI MEASUREMENTS The developed boost PFC, as schematically depicted in Fig. 5, has been tested according to the requirements of the EMC standards, which regulate the conducted EMI levels for these kinds of electronic devices [4]–[8]. As required, the converter has been connected to a stabilized voltage source through an LISN, and spectra of the input current have been measured in the range from 150 kHz to 30 MHz under different conditions. The result of the first test on the converter is shown in Fig. 9. As can be seen, while the converter's behavior at low frequencies is pretty good, there is an excessive high-frequency content in the current spectrum. As can be noted, the low-frequency part (up to 1 MHz) of the noise spectrum is made of multiples of the switching frequency and is mainly differential mode noise. Its amplitude decreases with frequency thanks to the adopted differential mode filter. Instead, in the remaining part of the spectrum, the noise, which tends to increase with the frequency, is mainly common mode. Adding a common-mode capacitive filter at the converter's input does not improve the situation as expected, because, as Fig. 10 shows, the differential noise increases. In fact, due to the circuit unsymmetry determined by the asymmetrical location of the boost inductor, the increased input current flows mainly in the input wire which does not include the inductor. This contributes to a differential mode current, thus worsening the low-frequency performance of the converter. Moreover, the high-frequency behavior is modified by the resonance between the common-mode capacitors at the converter input and output and the stray inductance of their connection (about 50 mm). Finally, the increased input current could possibly saturate the magnetic core of the input inductor, thus reducing the filter effectiveness.Fig. 15. Effects of gate series resistance increase with switching frequency modulation (peak measurement).To further confirm this interpretation of the measurements, the additional common-mode filter is moved to the converter's output, in parallel with the one already located there. This determines a better low-frequency behavior, since more common-mode current can now circulate on the converter output side without affecting the LISN, but also a further worsening of the high-frequency part of the spectrum because of the resonances of the newly inserted filter with the one