数据加载中.....
|
jsp.display-item.identifier=請使用永久網址來引用或連結此文件:
https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/52534
|
| 题名: | 方形同軸吸波材料測試製具的結構與性能探討 |
| 其它题名: | Investigation of a novel square coaxial test fixture for the EMC absorber material |
| 作者: | 饒家豪;Jao, Chia-hao |
| 贡献者: | 淡江大學電機工程學系碩士班
李慶烈;Li, Ching-lieh |
| 关键词: | 同軸吸波材料量測;EMC absorber measurement |
| 日期: | 2010 |
| 上传时间: | 2010-09-23 17:53:04 (UTC+8) |
| 摘要: | 本論文的主要目的在設計一個可以用來量測一種吸波磚型態的吸波材料的製具，並探討其校正的程序與效果。首先，藉由探討電磁波於同軸波導內的傳遞情形，檢視與確認模擬結果的合理無誤之後（使用模擬軟體SEMCAD X），進一步藉由模擬設計與驗證適當的準具尺寸與所需的校正kit的長度。
在研究初期，對於製具結構的適合尺寸尚無法精確掌握，使用軟體模擬來進行輔助設計，可以減少實際製具製做的金錢與時間的浪費，並提供吾人對其特性的初步瞭解，相當便利。
對於量測製具結構的探討，吾人首先模擬一單純的同軸線結構，其內半徑為1.52mm，外半徑為3.5mm，目的在證模擬軟體的可信度。接著，為了因應真正製具結構的阻抗不連續的效應，吾人修改單純的同軸線結構，刻意在同軸線的內導體引入一金屬細環（將內導體加粗即可），使它成為一阻抗不連續的結構，反射量約-30~-15dB左右，其目的在確認，針對此種阻抗不連續之結構，吾人可以利用校正的程序，將阻抗不連續的效應於反射參數的量測值中予以校正排除（deembedded），進一步獲得單純來自於吸波磚材料的反射貢獻。
在藉由模擬軟體確認針對一具金屬細環的阻抗不連續之結構，校正的程序的適用性之後，接下來，吾人就完整的同軸正方形製具進行模擬，其前面連接以同軸線激發，中間則是喇叭型的過渡區域，正方形製具的結構及尺寸雖然和激發同軸線相差很大，但中間的喇叭型過渡區域則可適度的減少反射，其反射量約-15dB左右，針對此種同軸正方形製具的阻抗不連續之結構，吾人可以利用上述的校正程序，將正方形同軸段之前的阻抗不連續效應於反射參數的量測值中予以校正排除（deembedded），進一步獲得置於其內部的吸波磚材料對反射的貢獻。
在使用軟體模擬的過程中，吾人發現其在低頻(約500MHz以下)的結果不盡理想，為此，嘗試改使用外部的高斯函數當作激發訊號的source，目的在增加其直流與低頻的頻譜成分，以改善低頻的模擬結果。另外，也優化了校正的方法，從原先使用三種不同距離的短路片，變成多種（至多六種即可）不同距離的短路片，如此一來，便可有效解決校正所帶來的反射係數尖峰（peak）產生的問題。
The main purpose of this thesis is to design a test fixture for the absorbing materials (in brick form with square shape) used in the EMI chamber, and to explore the calibration procedures. Initially, by examining the propagation of a electromagnetic wave in an uniform coaxial line, the characteristics of the simulation results are correctly verified after several tests (using the simulation software SEMCAD X). Furthermore, the design and verification of the calibration kits with appropriate length are conducted by simulation.
At the first stage of test fixture design, the appropriate size of the structure usually could not be well known precisely, where software simulation can be employed to aid the design. By this, we can control and reduce the money and time for making and re-making the actual test fixture setup. It also provide us a preliminary understanding of its features in a very convenient way.
For the deep study of the measurement test fixture , this thesis first simulate an uniform 50 Ohms coaxial line structure, within an inner radius 1.52mm and outer radius 3.5mm. The purpose is to verify the simulation software in the credibility and consistency. Then, in order to test the effect of impedance discontinuities existing in a real measurement system, I modified the simple coaxial line structure, for which a thin metal ring is deliberately introduced on the coaxial line inner conductor. Certain part of the inner conductor is widen to introduce the impedance discontinuous, of which the reflection is about -30~-15dB. The purpose is to confirm that for this kind of impedance discontinuity the calibration procedure can be employed to calibrate the effect of impedance discontinuities such that the measured reflection parameters can be correctly de-embedded. Thus, the contribution to the reflection simply due to the brick of absorbing material can be determined.
After the confirmation for the small ring of the coaxial line structure and the applicability of the calibration procedure by software simulation, this thesis conducted another simulation for a coaxial square line measurement system, in front of which a coaxial line is connected to excite the measurement system, while in between there exists a flaring horn (the transition region). Although the size of the square coaxial line system is greatly different from the excited coaxial line, but the middle horn-like transition region is appropriately designed to reduce the reflection, of which the reflection is about-15dB or so. For such a coaxial square system with impedance discontinuity, the above-mentioned calibration procedure can be employed again to calibrate the effect of impedance discontinuities such that the measured reflection parameters can be correctly de-embedded. Thus, the contribution to the reflection simply due to the brick of absorbing material placed inside the coaxial square system can be determined.
During the course of software simulation process, It is found that at low frequencies (around 500MHz and less) the de-embedded results are not satisfactory. To improve this, the Gaussian function is used to serve as an external excitation signal source, of which the aim is to increase the dc and low frequency spectral components. Finally, it is found that the characteristics of the calibration results show some unreasonable peaks in the reflection coefficients. For correction of the calibration procedure, this thesis examines the effects of utilization of more calibration kits：instead of three shorting plates with different distances, more (up to six) shorting plates with different distances are tested, and the results show that the problems of unreasonable peaks in the reflection coefficients are successfully resolved. |
| 显示于类别: | [電機工程學系暨研究所] 學位論文
|
文件中的档案:
| 档案 |
大小 | 格式 | 浏览次数 |
|---|
| index.html | 0Kb | HTML | 151 | 检视/开启 |
|
在機構典藏中所有的数据项都受到原著作权保护.
|
