ICP-SM-782简介

IPC-SM-785《表面安装焊接件加速可靠性试验导则》

美国电子电路与电子互连行业协会（IPC）在1992年发布了IPC-SM-785 Guidelines for Accelerated

Reliability Testing of Surface Mount Attachments《表面安装焊接件加速可靠性试验导则》，该标准对于如何进行表面贴装焊点的可靠性试验提供了指导意见，提出了应该如何评估可靠性试验的结果、应该如何从可靠性试验的结果外推到焊点在实际使用环境中的可靠性，并且为更好地理解加速试验提供了相关的背景知识和设计思路。

IPC/EIA J-STD-029《倒装焊、芯片尺寸封装、球栅阵列和其它表面安装阵列封装的性能和可靠性试验方法》 在2000年IPC与美国电子工业协会EIA联合发布了IPC/EIA J-STD-029 Performance and Reliability Test Methods for Flip Chip, Chip Scale, BGA and other Surface Mount Array Package Applications《倒装焊、芯片尺寸封装、球栅阵列和其它表面安装阵列封装的性能和可靠性试验方法》，该标准专门针对倒装焊、芯片尺寸封装、球栅阵列和其它表面安装阵列封装的质量和可靠性，提供了详细的测试方法，并且为供应商和用户提供了根据试验数据确立的可接受标准。该标准可以用来确认原材料的选择、优化生产过程、坚固老化产品、预测产品的长期可靠性。试验成功的关键是需要仔细的计划、设计和制作适当的测试设备，才能得到有意义的结果。其中，焊点的可靠性测试是其中重要的一部分。 IPC-9701《表面安装锡焊件性能试验方法与鉴定要求》

在2002年IPC又发布了最新的IPC-9701 Performance Test Methods and Qualification Requirements for Surface Mount Solder Attachments 《表面安装锡焊件性能试验方法与鉴定要求》，该标准建立了明确的试验方法来评估电子组装中表面贴装焊点的性能和可靠性，通过测试可以对刚性电路板、柔性电路板、半刚性电路板上的表面贴装焊点的性能和可靠性划分为不同的级别，同时提供了一种近似方法把可靠性试验结果与焊点在实际使用环境下的可靠性联系起来。

该标准可以用来确认产品的设计和生产组装过程可以生产出符合要求的产品、可以根据通用的数据库和分析技术来预测焊点的可靠性、同时提供了标准化的试验方法和报告方法。 IPC-SM-785和IPC-9701的区别：

IPC-SM-785是一个指导性文件，不是标准。由于没有适当的标准，出现了高度加速的实验方法，不符合IPC-SM-785指引的方法；还有一些过分的主张，比如试验结果即意味着产品的可靠性。不断缩小的元件尺寸现在要求将焊接点的可靠性设计到元件中去，需要一个客观的手段来提供一个在竞争的产品中比较可靠性的方法。基于这个理由开发出的IPC-9701，主要目的是研究那些发生在元件与PCB之间由于热膨胀不匹配而引起的焊接点可靠性问题。

另IPC-9701明确规定了焊点连续电性能测试的方法─Data logger等测试方法及相关专用测试设备，如Espec AMR系列。可以连续动态捕捉各种条件下开路情况的失效数据，不会受电干扰影响；电阻值、温度值等可以进行自动连续测试采集，可以准确地分析失效产生机理；同时可以控制试验设备，更真实地再现环境与失效之关系；通过计算机储存，利用丰富的统计功能可以进行包括威布尔分布等各种统计分析。

IPC-SM-785

This standard was developed by the SMT Accelerated Reliability Test Task Group of the Product Reliability Committee of IPC. This document provides an excellent introduction to issues of solder joint reliability testing and analysis. Some highlights particularly relevant to Event Detectors are quoted below：

Section 4.3.1 Failure Definition [para 3]

A solder joint that fails by fully fracturing typically does not exhibit an electrical open or even a very noticeable increase in electrical resistance. A failed solder joint is normally surrounded by solder joints that have not yet failed and therefore the solder joint fracture surfaces make compressively loaded contact. Electrically, the solder joint failure manifests itself only during thermal or mechanical transients or disturbances in the form of short duration (~ 1 usec) high resistance spikes (>300 ohms). During thermal changes the solder joints are subject to shear, not tensile, loading; therefore, fracture surfaces of fractured solder joints slide relative to each

other producing the characteristic short duration intermittents. Therefore, in this context, the practical definition of solder joint failure is the interruption of electrical continuity (>300 ohms) for periods greater than 1 microsecond. [16] ...

Section 4.5 Statistical Considerations

[para 2] Solder joints are not uniformly subject to the same thermal expansion mismatch, and thus the same amount of fatigue damage, because of their distances from the neutral point of the component or connector. Therefore, the total number of solder joints cannot be counted as the sample size. For test purposes and depending on size and symmetry, a given component or connector can be partitioned into two or four samples, if the continuity nets are correspondingly designed. However, treating each solder joint as an individual sample assumes incorrectly that they all have the same failure probability.....

[para 4] Therefore, the partitioning of the test samples needs to include essentially all solder joints into the continuity daisy-chain with each partition having an equal likelihood of failure. ... Section 7.8 Failure Criteria for Solder Joint Fatigue Tests

The complete document is available from www.ipc.org Other sections of particular interest for the design of accelerated test programs include:

4.0 Surface Mount Solder Attachment Fatigue Behavior and Reliability Prediction

5.0 Design for Solder Attachment Reliability

6.0 Manufacture/ Processes

7.0 Accelerated Reliability Testing 7.1 Reliability Program/Strategy

7.3-7.5 Thermal/Mechanical Cycling /Vibration 7.7 Mechanical Shock Testing

Appendices: References and Bibliography

Solder Joint Failure (technical article)

Werner Engelmaier, Engelmaier Associates, L.C. 7 Jasmine Run, Ormond Beach, FL 32174 phone: (386) 437-8747 fax: (386) 437-8737

email: Engelmaier@aol.com website: www.engelmaier.com

The monitoring of electronic assemblies for fractured solder joints is not a straight-forward task. Fully fractured solder joints, regardless of whether they occur during accelerated reliability testing of prototype test vehicles or operational use of product in the field, do NOT result in steady-state electrical opens. Typically, solder joint fracture occurs initially on only one solder joint in the whole electronic assembly; while the other solder joints have been subjected to the same cyclic environment and have accumulated fatigue damage, they will fail following a

statistical failure distribution. The fractured solder joint will cause intermittent failures or circuit malfunctions. Also typically, the failures occur in a corner-most solder joint of a component

having the highest, or near highest, probability of failure, perhaps due to a large component size, a large CTE-mismatch between component and substrate, etc.

The electrical indications of failure are characteristically intermittent because the fractured solder joint surfaces never physically separate as long as the component is still attached to the substrate by some unfractured solder joints. The solder joints neighboring a fractured solder joint keep the solder joint fracture surfaces in compressively loaded contact. The displacements during thermal cycling, and even during board bending are predominantly in the x- and y- (shear) directions, NOT the z- (tension) direction. Thus, electrical indications of solder joint failures occur only during dynamic loading of the solder joints. To get a measurable manifestation of failure requires that the solder joints be monitored either during the transient periods in thermal cycling or during a mechanical excitation (vibration or mechanical shock.)

In practical terms this means that electrical continuity daisy-chained nets of test vehicles for accelerated reliability tests need to be monitored continuously during cycling. Multiplexed monitoring looks at given continuity test net always during the same time slot, which may or, more likely, may not, fail during the critical transient transition periods. It clearly is not adequate to look for failure during a halt in the cyclic test program. Neither is it adequate to look at equipment with a reported failure indication on some test fixture that does not in some form dynamically excite the assembly under test; the likely result of such an investigation is \No Trouble Found.\

Given the need to test a statistically adequate sample size under a variety of parametric variations, the monitoring equipment requirements can become formidable rather quickly. For pure research purposes, it might be of interest to determine when a partial crack in a solder joint results in a specific increase in electrical resistance or a specific decrease in solder joint strength. However, from a product reliability point of view, these criteria are not particularly relevant. Partial

fractures need to be almost complete with the fracture surfaces physically separated before even a 10% increase in resistance can be measured. Similarly, the reduction in joint strength due to micro cracks or partial fractures has no significant reliability impact, except for uses with

significant vibrational loading. Solder joints in operating electronic equipment are not subjected to stresses large enough to cause overstress failure in solder joints even with substantially

reduced strength. Further, equipment to record high-speed, transient, electrical resistance data is expensive; a fact that becomes very important with increasing numbers of test nets that need to be monitored.

From pure functional considerations, monitoring for short duration high resistance spikes is most meaningful, since this closely resembles the actual failure indication in product. From practical considerations, this allows the most cost-effective use of resources, since monitoring equipment to meet this test requirement is relatively inexpensive.

Engelmaier Associates, L.C. is a consulting company that provides consulting services on reliability, manufacturing and processing aspects of electronic packaging.

Engelmaier Associates, L.C. provides services to OEMs, printed circuit board manufacturers, PWB subcontract assembly companies and end users in applications such as telecommunications,

computers, military, aerospace, and automotive. These services can be in the areas of design for reliability, accelerated reliability testing, reliability prediction, design for manufacturability and quality, processing effects, environmental stress screening, and failure cause analysis, for solder joints, plated-through holes/vias, and flexible circuitry. Workshops in theses areas are offered as well.

Werner Engelmaier, President of Engelmaier Associates, L.C., has been instrumental in the development of many industry documents, such as IPC-SM-785 \Reliability Testing of Surface Mount Solder Attachments\

Methods and Qualification Requirements for Surface Mount Solder Attachments\\ANSI/IPC J-STD-012 \ANSI/IPC J-STD-013 \Technology\

表面贴装焊接点试验标准

By Jack Crawford and Werner Engelmaier

本文介绍，由于有问题的测试方法和过分的主张，IPC开发了一个标准来保证正确的焊接点可靠性试验。理想的焊点形成一个可靠的、电气上连续的、机械上稳固的联接。适当的可靠性设计(DFR, design for reliability)是需要的，以保证适当的性能。使用DFR设计的焊点，当以良好的品质制造时，可以在产品的设计运行环境中工作到整个设计寿命。

加速试验问题

在DFR方面，请参阅IPC-D-279《可靠的表面贴装技术印制板装配设计指南》。可是，在许多情况中，足够的可靠性应该通过加速试验来证实。IPC-SM-785《表面贴装焊接的加速试验指南》给出了适当的加速试验指引。IPC-SM-785是一个指导性文件，不是标准，适当的加速试验要求相当的资源与时间。由于没有适当的标准，出现了高度加速的实验方法 - 不符合IPC-SM-785指引的方法 - 还有一些过分的主张，比如试验结果即意味着产品的可靠性。

不断缩小的元件尺寸现在要求将焊接点的可靠性设计到元件中去。需要一个客观的手段来提供一个在竞争的产品中比较可靠性的方法。基于这个理由，开发出IPC-9701《表面贴装焊接的性能实验方法和技术指标要求》。

可靠性试验要求

虽然JEDEC的试验单独地涉及到元件，但是2002年1月发布的IPC-9701的主要目的是试验那些受到发生在元件与PCB之间的热膨胀不匹配所威胁的焊接点的可靠性。因此，应该考虑完全不同的物理参数和损坏机制。由于PCB在多数情况下是一个常数(考虑FR-4，厚度足够防止由于PCB弯曲的应力释放)，因此试验要设计来显示适宜性，或一个给定元件因此而对各种运作环境缺乏。为了试验的目的，PCB与表面涂层应该标准化，使得它不影响试验结果。

这些方面不应该妨碍在IPC-9701中描述的方法的使用，以比较性地评估不同的表面涂层，或任何

其他变量，只要清楚地叙述了与IPC-9701的不同之处。对产品运行环境的任何推断都是无效的，例如该文件附录A中所注释的条件。

表一和二提供了IPC-9701的试验条件和合格要求，一起有结果试验温度范围(ΔT)和平均循环温度(Tsj)。也包括了相关的注释，有关推荐的试验条件和合格要求，以及超出IPC-SM-785警告的温度循环范围。

表一、IPC-9701中描述的试验条件 试验条件 TC1 TC2 TC3 TC4 TC5

T(min) 0 oC -25 oC -40 oC -55 oC -55 oC

T(max) +100 oC +100 oC +125 oC +125 oC +100 oC

ΔT 100 oC 125 oC 165 oC 180 oC 155 oC

Tsj 50 oC 37.5 oC 42.5 oC 35 oC 22.5 oC

注释 推荐参考

违反IPC-SM-785 违反IPC-SM-785 违反IPC-SM-785

表二、IPC-9701中描述的合格要求 合格要求 循环数 NTC-A 200 NTC-B 500 NTC-C 1000 NTC-D 3000 NTC-E 6000

说明

推荐参考TC2,TC3,TC4 推荐参考TC1

IPC-9701标准化了五种试验条件下的性能实验方法，从良性的0<->100oC的TC1参考循环条件到恶劣的-55<->125oC的TC4条件。符合合格要求的热循环数(NTC)从NTC-A变化到NTC-E，NTC-A等于200次循环(在任何试验条件下容易达到，基本上只保证适当的焊锡湿润)，NTC-E等于6000次循环。在T(min)和T(max)温度极限的驻留时间对于所有试验条件都是10分钟。

相对的慢加速TC1试验条件是基准试验目的的首选参考试验条件，因为该试验最接近地模仿实际使用条件，外部损伤机制支配的可能性最小。对实验条件TC1的首选合格要求是NTC-E，即等于6000次循环。

IPC-9701也包括附带条件，对于试验条件TC3, TC4和TC5的温度循环范围可能具有不止一种损伤机制。这个事实会破坏IPC-SM-785所述的焊接可靠性适当加速试验的警告条件 - 焊锡的时间、温度、应力有关的材料性能的结果。由于多种损伤机制，这些结果的混杂会造成对产品可靠性评估的加速试验结果的推断出问题。无任如何，要为这些实验条件提供加速因子，以提供相对的指引。

加速因子与方程

存在两个加速因子(AF, acceleration factor)：1) AF(循环) - 与焊点的循环疲劳寿命有关，该寿命是在有关给定使用环境中产品寿命的试验中获得的，2) AF(时间) - 与焊点失效的时间有关， 它是在有关在给定的使用环境中产品寿命的试验中获得的。

循环失效的加速因子是：AF(循环) = N(现场) / N(试验)。

时间的加速因子是：AF(时间) = AF(循环) x f(试验) / f(现场)

IPC-9701使用Engelmaier-Wild焊点失效模型来评估加速因子，该模型在IPC-D-279的附录A中作了叙述。当然，也存在其他模型可用于这个目的，但是由于大多数这些模型是基于来自焊点加速可靠性试验的经验数据，所得到的加速因子不应该与来源于焊锡疲劳数据的模型有太大区别。IPC委员会预计该文件的未来版本将包括来自其他模型的加速因子，当这些模型可以得到的时候。 Jack Crawford is director of assembly, standards and technology with IPC, Northbrook, IL; (847) 790-5393; e-mail: jackcrawford@ipc.org.

Werner Engelmaier is president of Engelmaier Associates, L.C., Ormond Beach, FL; (386) 437-8747; e-mail: engelmaier@aol.com.

(Presented by Aaron 08/20/2002)

循环失效的加速因子是：AF(循环) = N(现场) / N(试验)。

时间的加速因子是：AF(时间) = AF(循环) x f(试验) / f(现场)

IPC-9701使用Engelmaier-Wild焊点失效模型来评估加速因子，该模型在IPC-D-279的附录A中作了叙述。当然，也存在其他模型可用于这个目的，但是由于大多数这些模型是基于来自焊点加速可靠性试验的经验数据，所得到的加速因子不应该与来源于焊锡疲劳数据的模型有太大区别。IPC委员会预计该文件的未来版本将包括来自其他模型的加速因子，当这些模型可以得到的时候。 Jack Crawford is director of assembly, standards and technology with IPC, Northbrook, IL; (847) 790-5393; e-mail: jackcrawford@ipc.org.

Werner Engelmaier is president of Engelmaier Associates, L.C., Ormond Beach, FL; (386) 437-8747; e-mail: engelmaier@aol.com.

(Presented by Aaron 08/20/2002)

本文来源：https://www.bwwdw.com/article/d6bw.html