"On board applications" chip
With the maturity of on-board application platform, the output value of global automotive electronics will grow significantly. The compound annual growth rate of automotive IC before 2018 is 10.8%, which is the highest in more fields. Among them, automotive IC in the Asia Pacific region is as high as 20%. The IC industry is all ready to attack this new blue ocean market.
The market of vehicle IC is more closed than the largest difference between ICT and IC. The development and validation period of the vehicle market may be 3 years. The operation mode of Time to Market in Taiwan and Chinese mainland has been inconsistent with the value concept. This paper will explain the IC Verification specification of aec-q100 and analyze how chip manufacturers can meet the needs of customers in car factories / module factories. The discussion focus will be on the requirements of aec-q100 version h, the newly revised specification in September 2014.
The threshold for entering the vehicle power supply chain is AEC and ISO / TS 16949
To enter the vehicle field and enter the supply chain of tier 1 vehicle and power plants, you must obtain two tickets:
The first is the aec-q100 (IC), 101 (discrete components) and 200 (passive parts) reliability standards pushed by the North American automobile industry;
The second ticket shall comply with the zero defect supply chain quality management standard ISO / TS 16949 (quality management system), and its connection can be explained with reference to figure 1.
Fig. 1 basic requirements of vehicle components
Reliability and quality requirements of vehicle parts and components due to market differences
The auto parts and components market can be roughly divided into three parts, including OEM / ODM (factory parts) / oes (factory repair parts), DOP (dealer option) and am (after Marketing).
The failure rate estimation and spare parts preparation strategy of customers will change due to the decision to enter different markets. OEM / ODM / oes is guaranteed by the original manufacturer. Due to its long warranty period, each vehicle manufacturer needs to balance the cost of manufacturing and after-sales service, and the threshold for IC suppliers to enter is high. DOP is an optional item made by dealers due to the sales strategy needs of the local market. The entry threshold is similar to the above, and the after-sales market (AM) has nothing to do with the original warranty, so the relative entry threshold and cost are low. On the other hand, there are many types of products with am on the other side, belonging to the audio-visual periphery, which has nothing to do with the active and passive safety, and the required reliability is also lower than that of the original parts (Fig. 2).
Figure 2 market differences of vehicle components
Understand the verification process of vehicle IC specification aec-q100
So, how should IC designers enter the automotive IC supply chain? First, you should understand one of the tickets aec-q100. Figure 3 shows the verification process in aec-q100 specification. This figure is drawn based on the manufacturing process of die design → wafer fab. → PKG assembly → tesling. The relevance of each group needs to refer to the arrow symbols in the figure. Here, the verification process is divided into five parts for simple description, and the details of each test will not be described in detail.
Figure 3 aec-q100 verification process
Design House
Area 1 is the functional test before and after the reliability test. The implementation mode of this part needs to be discussed between the IC design company and the test factory. The main difference from the general IC Verification is the temperature setting of the functional test, which will be described later.
Wafer Foundry
Area 2 of group D is the wafer level reliability verification of the wafer factory. The IC industry without fables must obtain relevant data from the wafer factory entrusted to manufacture.
Reliability Test
Area 3 refers to the items that need to be executed according to product packaging / characteristics. AEC divides them into group A (accelerated environmental stress test), group B (accelerated working life simulation), group C (package integrity test), group E (electrical verification test) and group G (cavity / sealed package integrity test).
Design Verification
Area 4 of part group E is failure mode and effect analysis and evaluation in the design stage, characteristic verification in the finished product stage and fault coverage calculation.
Production Control
Quality control in the regional production stage of group F, including yield / bin, using statistical methods to control and formulate standard treatment processes.
Read and understand the new requirements of aec-q100 h version
Since the release of aec-q100 version h in May 2007, after a lapse of more than 7 years, version g was released in September 2014. This paragraph will discuss the main differences between aec-q100 version h and version g, as well as the specifications after revision.
Check the temperature level
The IC supplier must first understand how the end customer uses this IC and its installation position in the vehicle, and determine the appropriate temperature level according to the actual temperature range. This temperature level definition will be applied in two parts.
The first part is the condition selection of each reliability experiment when the test plan is launched, such as temperature cycling (TCT) experiment. The temperature variation range and the number of temperature difference cycles at different levels will be different.
The second part is the functional test temperature before and after the reliability test mentioned above. The final test (FT) test of functional application must be carried out according to the temperature range set by the user. The temperature grade set is grade 1 (- 40 ℃ ~ 125 ℃), which represents the use of low temperature - 40 ℃, normal temperature 25 ℃ and high temperature 125 ℃ in Ft, It should be noted that the test temperature is determined in order. For example, the definition sequence of high temperature operating life (HTOL) experiment in FT test is room → cold → hot.
The new version partially cancels the temperature grade of grade 40 ℃ ~ 70 ℃ (Table 1). If this temperature grade is compared with the description of vehicle modules in ISO 16750 / SAE j1211 and other specifications, it cannot correspond to appropriate products. Therefore, the author believes that canceling this temperature grade is closer to the requirements of actual modules.
Table 1 definition of temperature grade
Comparative experimental items
In the new version, the added and deleted items include the following two types.
Cancel: Machine Model in high temperature gate leakage test and electrostatic discharge
The gate leakage (GL) part is mainly used to simulate the environment in which high temperature and high electric field may occur at the same time when the vehicle module is applied. This environment will cause GL failure of the equivalent capacitance and resistance in the IC package. This failure phenomenon can be recovered by high temperature baking.
The reason for cancellation is not stated in the specification, but according to the accumulated verification and test experience for many years, this failure mode often occurs in the process of burn in and reflow. Although the specification has been cancelled, if the same failure occurs in the production process or practical application, this method can still be used for reproducibility test.
Machine mode (mm) is synchronized with JEDEC's jesd47 specification. First, human body model (HBM) can be equivalent to MM's experimental results, and second, charged device model (CDM), which is more important than mm. Therefore, more emphasis should be placed on CDM testing.
As for the correlation between HBM and mm mentioned in the literature, from practical experience, ESD protection circuits of some products are still different between HBM and mm. Although this project is cancelled in the specification, IC manufacturers still have to face the treatment when ESD customer return occurs. ESD is defined as design verification, so it is still listed as a standard test project by various manufacturers.
New: lead (PB) free experiment
The automotive power and medical industry is different from the information and communication technology industry. The automotive power and medical industry focus on technology maturity, reliability and zero failure, rather than the most advanced technology pursued by information and communication technology. Therefore, the conversion time of automotive electronic products in lead-free process is later than that of consumer products, The formal inclusion in the test project also represents that the conversion of lead-free process has been quite mature, but some products such as engine room high-temperature applications are still allowed to use lead-free process. Lead free verification items include solderability, solver head resistance and Whistler.
Change the experimental conditions
The main experimental conditions are changed in the two experiments of high temperature working life (HTOL) and temperature cycling (TCT). For the other related experiments of wire bonding, the calculation of PPK is cancelled, Cpk can be used, solderability shows that steam aging can be replaced by baking, and the number of experimental samples in group G is slightly reduced, Please refer to the adjustment part of the reference specifications for the test items, which will not be described in detail here.
HTOL
There are three parts: one is that the growth of experimental time is 1000 hrs, the other is that the temperature is clearly defined as junction temperature (TJ), and the third is the definition of experimental high-temperature alignment grade.
TCT
The low temperature of the lowest standard is adjusted from - 50 ℃ to - 55 ℃, and the cycle number is partially increased. The equivalent experimental conditions can still be replaced by referring to the specification of jesd22-a104. For the part of experimental conditions, please refer to the description in part V later to better understand the meaning of this specification change.
What is general information
First, figure 4 is used to illustrate the basic meaning of generic data. If products a and B use the same process or material, they can be preliminarily defined as products of the same family. If product a is fully AEC q-100 qualified, the test results produced by parts of the same process or material are called generic data, regardless of the verified quantity and procedure, As long as the definition of generic data complies with the description of aec-q100, product B can also be declared to have passed aec-q100 by adding generic data after verifying the remaining items.
Figure 4 Qualification family and generic data
The new version has a clear description of the definition and use principles of generic data and qualification family, and simplifies its certification procedures. At the same time, it uses situational simulation cases to explain which process changes should be carried out, which experimental items and the number of lots, and has a clear definition. Due to too many contents, you can refer to the specification again if necessary.
Draft test plan
The most important part in this article and the biggest change in this revision is the robustness validation emphasized by the society of automotive engineers in sae-j1879 / j1211. The validation plan should be formulated according to the service conditions faced by the product in the actual application environment, Instead of applying all products with a single test standard / condition, that is, test for application, not test for standard.
When formulating an appropriate verification plan, the first step is to formulate the purpose of the design / production of the component, which is called mission profile. In addition to meeting the functional tasks, reliability tasks should be added. Table 2 shows the example of vehicle mission profile parameters.
Table 2 example of vehicle mission profile
The IC supplier shall consider that the components with different application functions will correspond to different mission profiles. If the IC working behavior is in non operating time, such as the application of alarm, the working time (life time) condition shall meet the condition of 116400 hrs at room temperature.
If the IC only works when the engine is on, the mission profile must meet the life time of 12000 hrs and the service temperature of its working position. Assuming that the average working temperature of the IC when the engine is on is 87 ℃, the HTOL test temperature used is 125 ℃, and the activation energy is set to 0.7 ev, Next, the Arrhenius model is used to calculate the temperature acceleration rate during the experiment, as calculated by Formula 1:
Aft = exp (EA / k) * (1 / tuse-1 / tstress) formula 1
In this way, the temperature acceleration rate aft = 8.6232 can be calculated, and the target life set above is 12000 hrs, so the HTO test time should be 12000 hrs / 8.6232 = 1392 hrs.
In addition to the example of temperature acceleration, the acceleration formula including temperature cycle / humidity has been listed in the new specification, and the text of the specification can be referred to again.
The above example aims to illustrate how to design an appropriate test plan based on the mission profile of the practical application of end products. I believe many quality control units engaged in IC design are quite familiar with it. What this paper wants to express is that the specification will gradually abandon the single standard and be handed over to the end user (e
Our other product:
