臺灣博碩士論文加值系統

由於半導體技術進步的關係，電腦的效能得以快速發展，但隨著CPU內部的核心越來越多，帶來的高熱量也越來越高。為了解決CPU的散熱問題，本篇論文要進行水冷系統的改良，並期待可以更有效的對CPU或其他電子元件降溫，以延長其壽命。
本論文總共有四個主題來做探討: (1)在水冷散熱中，探討冷排、冷卻水量和CPU本身功率對於CPU溫度的關係，以及探討冷排和冷卻水量對散熱效益的影響力比較；(2)水冷散中的循環中加入熱電致冷晶片以探討其對水溫和CPU溫度的影響；(3)以數值模擬的結果得到水溫對散熱效益的影響；(4)在實際裝機上，探討CPU不同功率時致冷晶片模組對CPU溫度的影響。
研究結果顯示，對於CPU溫度之影響，在不改變管路的幾何形狀下，改變冷排大小的散熱效益大於冷卻水量變化，因此在機殼允許的情況下，安裝較大的冷排有較大的散熱能力，而冷卻水量增加對CPU溫度的影響不大。由於冷卻水只是一個吸收熱量的介質，因此水溫和CPU的溫差越大，吸熱的效果會越好。在水冷散熱循環中加入致冷晶片模組，由實驗得到的結果可以觀察到當水溫降低時，CPU溫度同時會降低。最後以120mm的冷排加裝致冷晶片模組可以得到CPU的溫度明顯下降，且比無致冷晶片模組的240mm冷排散熱效率好，但會消耗掉更多的電能。

This thesis focuses on four topics: (1) find the relationship of radiator, water volume, controlling CPU power by C language and CPU temperature in water-cooling system; (2) find the difference of heat dissipation efficiency with radiator’s thickness; (3) obtain the effect of water temperature on heat dissipation by COMSOL. (4) Discuss the influence of thermoelectric cooling installed on real water-cooling system.
The research results show that for the influence of CPU temperature, the radiator benefit of changing the size of the radiator is greater than the change of the volume of water without changing the geometry of the pipe. Therefore, the bigger radiator, the better heat dissipation if allowed. Increase the volume of water has little effect on the CPU temperature because the cooling water is just a substance to absorb heat. It will not affect CPU temperature directly but it has an effect of heat dissipation. The temperature of water is obtained by using Arduino and Ds18b20. If the temperature of the water is higher, the heat dissipation of the water cooling system is better, but it is not good for the CPU temperature. Accordingly, the thermoelectric cooling module is added to the water-cooling system cycle to decrease water temperature. From the experimental results, it can be observed that when the water temperature is lowered, the CPU temperature is also lowered. Finally, the 120mm radiator is equipped on the thermoelectric cooling module to obtain a significant drop in CPU temperature, and the 240mm radiator radiating efficiency of the non-thermoelectric cooling module is better, but the module consumes more power.

摘要 I
SUMMARY II
第一章 緒論 1
1-1 前言 1
1-2 文獻回顧 2
1-3 研究動機與目的 3
第二章 實驗原理與分析方法 4
2-1水冷散熱系統理論 4
2-2 CPU功率控制與溫度記錄 4
2-3熱電致冷晶片原理 5
2-4 溫度感測 6
2-5 最小平方法 6
2-6 CPU熱損失計算 8
第三章 實驗設備與方法 9
3-1水冷散熱系統各項控制變因對散熱效益的影響 9
3-1-1 實驗設備 9
3-1-2實驗方法 14
3-2厚薄冷排的散熱效益比較 16
3-2-1實驗設備 17
3-2-2實驗方法 17
3-3 以數值模擬軟體計算水冷散熱效益 17
3-3-1 水冷頭數值模型設定 18
3-3-2 冷排的數值模型設定 19
3-4不同功率的熱電致冷晶片對CPU溫度的影響 19
3-3-1實驗設備 19
3-3-2實驗方法 21
第四章 結果與討論 22
4-1水冷散熱系統各自變數對散熱效益的影響 22
4-1-1 CPU的熱損失 22
4-1-2 CPU溫度和控制因子的關係 23
4-2厚薄冷排的散熱效益比較 26
4-3 數值模擬計算水冷散熱系統 27
4-4熱電致冷晶片對於水冷系統的散熱效益的影響 27
第五章 結論 29
參考資料 31
表格附件 34

1. 李俊賢. 半導體測試治具的水冷頭散熱性能研究. 碩士論
文, 國立成功大學, 2011.
2. 陳律安. 電腦水冷散熱系統之效益研究. 碩士論文, 國立成功大
學, 2005.
3. 楊宗穎, and 陳建燁. 迴歸直線方程式之參考推導過程.
4.Serial Ports. https://librambutan.readthedocs.io/en/latest/lang/api/serial.html
5. Bahiraei, Mehdi, and Saeed Heshmatian. Efficacy of a
Novel Liquid Block Working with a Nanofluid Containing
Graphene Nanoplatelets Decorated with Silver
Nanoparticles Compared with Conventional Cpu Coolers.
Applied Thermal Engineering 127 (2017): 1233-45.
https://doi.org/10.1016/j.applthermaleng.2017.08.136.
6. Chu, R. C., R. E. Simons, M. J. Ellsworth, R. R.
Schmidt, and V. Cozzolino. Review of Cooling
Technologies for Computer Products. IEEE Transactions
on Device and Materials Reliability 4, no. 4 (2004):
568-85. https://doi.org/10.1109/tdmr.2004.840855
7. COMSOL. Conjugate Heat Transfer.
8. DallasTemperature. Ds18b20說明書.
9. Elnaggar, Mohamed H. A. Numerical Investigation of
Characteristics of Wick Structure and Working Fluid of
U-Shape Heat Pipe for Cpu Cooling. Microelectronics
Reliability 54, no. 1 (2014): 297-302.
https://doi.org/10.1016/j.microrel.2013.08.019
10.Goth, Gary F. An Overview of the Ibm Power 775
Supercomputer Water Cooling System. Journal of
Electronic Packaging 134, no. 2 (2012).
https://doi.org/10.1115/1.4006140.
11.Haywood, Anna M., Jon Sherbeck, Patrick Phelan,
Georgios Varsamopoulos, and Sandeep K. S. Gupta. The
Relationship among Cpu Utilization, Temperature, and
Thermal Power for Waste Heat Utilization. Energy
Conversion and Management 95 (2015): 297-303.
https://doi.org/10.1016/j.enconman.2015.01.088.
12.intel. Intel 第4代處理器. 2013.
13.Kernighan, Brian W., and Dennis M. Ritchie. The C
Programming Language Second Eidition.
14.Oh, Hyoung Woo. Advanced Fluid Dynamics. 2012.
15. Arduino + Ds18b20 如何取得多顆 Ds18b20 溫度感測器序號
和溫度值. 2015.
http://ruten-proteus.blogspot.com/2015/07/arduino-
ds18b20-ds18b20.html
16.Saeed, Muhammad, and Man-Hoe Kim. Numerical Study on
Thermal Hydraulic Performance of Water Cooled Mini-
Channel Heat Sinks. International Journal of
Refrigeration 69 (2016): 147-64.
https://doi.org/10.1016/j.ijrefrig.2016.05.004
17.Singh, Randeep, Aliakbar Akbarzadeh, Chris Dixon,
Mastaka Mochizuki, and Roger R. Riehl. Miniature Loop
Heat Pipe with Flat Evaporator for Cooling Computer
Cpu. IEEE Transactions on Components and Packaging
Technologies 30, no. 1 (2007): 42-49.
https://doi.org/10.1109/tcapt.2007.892066
18.V, John H . Lienhard IV ; John H . Lienhard. A Heat
Transfer Textbook 3rd Edition. 2001.
19.Arduino Pratices : Ds18b20 Temperature Sensor. 2013.
http://yehnan.blogspot.com/2013/01/arduinods18b20.html
20.Zhou, Xiaotang, and James Braun. A Simplified Dynamic
Model for Chilled-Water Cooling and Dehumidifying
Coils—Part 1: Development (Rp-1194). HVAC＆R Research
13, no. 5 (2007): 785-804.
https://doi.org/10.1080/10789669.2007.10390986.