Technical Library | 2016-05-30 22:24:00.0
As a part of series of studies on X-Ray inspection technology to quantify solder defects in BGA balls, we have conducted inspection of 3 level POP package by using a new AXI that capable of 3D-CT imaging. The new results are compared with the results of earlier AXI measurements. It is found that 3D measurements offer better defect inspection quality, lower false call and escapes.
Technical Library | 2023-11-20 18:10:20.0
The electronics production is prone to a multitude of possible failures along the production process. Therefore, the manufacturing process of surface-mounted electronics devices (SMD) includes visual quality inspection processes for defect detection. The detection of certain error patterns like solder voids and head in pillow defects require radioscopic inspection. These high-end inspection machines, like the X-ray inspection, rely on static checking routines, programmed manually by the expert user of the machine, to verify the quality. The utilization of the implicit knowledge of domain expert(s), based on soldering guidelines, allows the evaluation of the quality. The distinctive dependence on the individual qualification significantly influences false call rates of the inbuilt computer vision routines. In this contribution, we present a novel framework for the automatic solder joint classification based on Convolutional Neural Networks (CNN), flexibly reclassifying insufficient X-ray inspection results. We utilize existing deep learning network architectures for a region of interest detection on 2D grayscale images. The comparison with product-related meta-data ensures the presence of relevant areas and results in a subsequent classification based on a CNN. Subsequent data augmentation ensures sufficient input features. The results indicate a significant reduction of the false call rate compared to commercial X-ray machines, combined with reduced product-related optimization iterations.
Technical Library | 2016-04-14 13:49:44.0
A system level test, usually built-in test (BIT), determines that one or more subsystems are faulty. These subsystems sent to the depot or factory repair facility, called units under test (UUTs) often pass that test, an event we call No-Fault-Found (NFF). With more-and more electronics monitored by BIT, it is more likely that an intermittent glitch will trigger a call for a maintenance action resulting in NFF. NFFs are often confused with false alarm (FA), cannot duplicate (CNDs)or retest OK (RTOK) events. NFFs at the depot are caused by FAs, CNDs, RTOKs as well as a number of other complications. Attempting to repair NFF scan waste precious resources, compromise confidence in the product, create customer dissatisfaction, and the repair quality remains a mystery. The problem is compounded by previous work showing that most failure indications calling for repair action at the system level are invalid. NFFs can be caused by real failures or may be a result of system level false alarms. Understanding the cause of the problem may help us distinguish between units under test (UUTs) that we can repair and those that we cannot. In calculating the true cost of repair we must account for wasted effort in attempting to repair unrepairable UUTs.This paper will shed some light on this trade-off. Finally, we will explore approaches for dealing with the NFF issue in a cost effective manner.
Technical Library | 2020-07-02 13:29:37.0
Industry standards such as J-STD-005 and JIS Z 3284-1994 call for the use of viscosity measurement(s) as a quality assurance test method for solder paste. Almost all solder paste produced and sold use a viscosity range at a single shear rate as part of the pass-fail criteria for shipment and customer acceptance respectively. As had been reported many times, an estimated 80% of the defects associated with the surface mount technology process involve defects created during the printing process. Viscosity at a single shear rate could predict a fatal flaw in the printability of a solder paste sample. However, false positive single shear rate viscosity readings are not unknown.
| 1 |
