By Ragnar Vaga, Global Business Development Manager and Keith Bryant, Global Director, Electronics Sales Yxlon International GmbH
In the connected Smart Factory of the future, all machines and systems will become essentially smart sensors, collecting all possible data from the production line and from the boards themselves, it is obvious that the quality of the raw data is one of the core components.
Therefore, the accuracy and sensitivity of those sensors is very critical, especially in quality control systems including AOI and AXI. Without diminishing the strong characteristics of the in-line inspection equipment, these still need assistive technologies for the verification of some challenging faults that appear at the same pace as the miniaturization of electronic assemblies.
In the connected Smart Factory of the future, all machines and systems will become essentially smart sensors, collecting all possible data from the production line and from the boards themselves
With the drive towards 100% inspection, in line X-ray and 3D AOI are becoming more frequently used at the SMT lines. However, both these technologies have certain limitations that may need assistive technologies for the verification of possible false faults. The false fault rate of the inspection system, one of the key items of the fault coverage, affects the overall efficiency of defect detection. Simply put, the more false faults you pass on to a repair station for validation the more real defects get missed, false faults also generate costs. Each defect reported by in line AOI or AXI must be verified, if the reported defect turns out to be a false alarm, then its verification effort is a cost that could have been saved, because the verification did not create any real value. False alarm verification costs include labor, capital equipment and other related costs such as maintenance and panel handling.
It is impossible to have a functioning SPC system if the root cause of failure cannot be identified, so false fails become a grey area in the statistics where it is not known if the failure is real or what type of failure it is. This is a real issue for any company trying to monitor and improve its yield, as lack of accurate data makes any improvements very difficult.
The way forward
Smart manufacturing should be a proactive and independent decision-making environment where connected systems or machines will “talk” to each other. That “discussion” is, however, based on algorithms. By definition an algorithm can give us only one single answer. But does this provide us the right answer or one from three or four or five alternatives? Will it provide the same response every time or different ones for the same issue? Many MIS systems are not yet ready to manage big data, due to the lack of smart analytic tools. Without human intervention to make those critical decisions, based on knowledge and experience, we still cannot talk about completely effective and efficient smart production. Therefore the reliable solution combines human brains and algorithms together, allowing Process Managers to make informed decisions after seeing all the data from the various collection points.
Considering all this, an at-line x-ray system should become an additional high-end sensor that can compensate, for example, for the limitations of AXI or 3D AOI. An at-line system is essentially an off-line system that is linked to the production line and placed beside it, allowing real-time response to issues and making the x-ray system a tool of the production and QA teams and not simply a test machine in a lab, used by QA staff.
Collected data is relayed to a single screen where they also have historical information and everything needed to make a truly informed decision. This is definitely a Smart Solution, it is not a “Lights Out” fully automated Smart Factory but it is much better than many factories today. It allows real time monitoring of production flow and the changing of settings and limits to ensure that the product produced remains within the optimum process window, giving potentially 100% acceptable product coming off the line, at least in terms of assembly defects. It is also a big step towards “full automation” and therefore well worth embracing as a way of improving yields and reducing costs.
This solution is simple to implement and therefore reliable, the at-line x-ray can be connected to an inline 3D AOI or in-line AXI inspection system directly where it becomes a smart verification tool where only questionable and unsolved faults will be transferred for repeat inspection on a high-resolution at-line x-ray system. The inspection results can be stored on the system PC, but can also be exported to the customer’s MIS, where it is more useful as it can be shared and viewed by many engineers.
The connectivity of the at-line system is based on Internet Protocol (IP) via Transmission Control Protocol (TCP). Further use of data (collation and visualization of results from different inspection systems) depends on the end customer’s demands and capabilities. The collected data allows the operator to understand what happened and make predictive analyses about future trends. The general task is to provide a dashboard view to the Process Manager who can track down the possible problem areas and initiate proactive measures to improve the quality of production, before rejects or scrap units are produced.
The at-line x-ray can be connected to an inline 3D AOI or in-line AXI inspection system directly where it becomes a smart verification tool
Added value of the Smart Factory solution
The ultimate target for this solution is to increase production yield and reduce the manufacturing costs, including troubleshooting and rework.
In high-reliability products, including aerospace, military and automotive products, there is no room for uncertainties. All possible defects have to be verified and reworked if possible. Conventional methods for troubleshooting are the ICT, microscope, etc. these are usually quite expensive and should be avoided, if possible.
As a reference we can use 700 USD as the hourly cost of the ICT method (incl. technology, labor and overheads). 400 USD per hour will be an average cost of a PCB rework. Obviously these numbers vary from country to country but should reflect the industry standard.
The value of this technology to the user will be on demand locations scanned from in line technologies and at line X-Ray, which can be compared side-by-side so that the user can make an informed and conclusive decision on the defect calls.
At the moment, the best Smart Factory quality control solution is the combination of human brain and algorithms