In the late 1970s, the first room-sized liquid-cooled Cray-1 was built, clocking in at 136 MFlops of sustained arithmetic computation. The term supercomputer was coined to described the unprecedented computational capabilities of the machine, and a new generation of computing technology was born. Barely three decades hence, garden-variety handheld smart phones are approaching 50 Mflops, and a high-end desktop computers are typically measured in gigaflops. Such massive increases in power and equally massive decreases in size have given rise to enormous complexity in integrated circuit design, with it’s attendant challenges of manufacturing complexity, heat dissipation, and problem diagnosis. Modern mobile computing exacerbates theses challenges with environmental unpredictability, and the need for consumer ergonomics is frequently at odds with design for manufacturability. The end result: the exploding power and sophistication in consumer-grade electronics has given rise to the need for frequent repairs which can be expensive to correct, and time-consuming to complete.
A few recent trends will serve to illustrate the problem in detail.
Processor Speed and Integration
Moore’s Law has been in effect since the invention of the transistor; it states that transistor density doubles every two years or faster. Half-pitch feature size is halving every 5.4 years; processor performance is doubling every 1.8 years; the average transistor price is halving every 1.6 years. These and several other similar metrics indicate that every form of communications and computing technology is doubling in price-performance, bandwidth, component density, and capacity every year. The practical implications for consumer electronics are enormous.
For example, modern notebooks and netbooks designs now feature highly-sophisticated logic boards that integrate variegated ports and peripherals, random access memories, WiFi controllers, video and HD sound chipsets, HDMI, USB and Fire Wire. These feature-rich logic boards typically utilize extremely dense, high-pin-count devices for the processor, video chip, Northbridge and Southbridge components. Such density increases the probability of failure somewhere on the board, and with it, the frequency of replacement. Similarly, the complexity of repairing these logic boards at the individual component level is dramatically heightened.
Density and complexity have attendant heat dissipation and airflow considerations as well. Consider the Northbridge chip as a specific component example. The Northbridge chip is responsible for communications between the CPU, RAM, and video cards, and also plays an important role in determining how far a computer can be “over-clocked.” Modern Northbridge designs can approach temperatures of 55 degrees C (130 degrees F). End users who may inadvertently restrict air flow can exacerbate heat build-up inside the unit, causing these components to fail prematurely. Mobile devices are particularly susceptible to this kind of mistreatment, but even non-mobile computing equipment is susceptible to solder joint stress fractures when exposed to extreme temperature swings. These can be particularly troublesome, because the failures may be intermittent, and therefore difficult to diagnose.
The following profile demonstrates the frequency of logic board replacement contrasted with the remaining primary components. Note that this profile is inclusive of in-warranty, extended warranty and out of warranty repairs, all customers supported, and spans more than 15 brands. The study is based on a random sample of 10,000 notebooks and 1500 desktops repaired in 2010. The customer mix and sample volume are regarded as sufficient to discount any risk that may be introduced by the nature of our client’s contracts with their end users. See figures 1 and 2.
Software and Configuration
Today’s multi-platform operating systems such as Windows Vista and Windows 7 running on 32 and 64 bit platforms are highly-configurable, particularly regarding wireless and networking capabilities. Technicians are therefore required to ensure that the drivers and controllers are appropriate to the operating system under scrutiny. Coupled with the proliferation of installed applications and the possibility of virus infection, this has the effect of obfuscating the boundaries between configuration, driver, software and hardware issues. The result is the frequent misidentification of software or configuration issues as a hardware problem when performance suffers or fails entirely.
Many newer consumer electronics products are now configured using LED screens instead of LCD screens. Both technologies are backlit, but LEDs offer greater contrast ratios, better color representation, wider viewing angles, and lower power consumption than their LCD counterparts. They are higher cost components, however, in part because of the complexity of manufacture. The presence of integrated power inverters, for example, has made the diagnostic and part-swapping process easier, as technicians have now have fewer individual elements to diagnose, but accessing and repairing the inverter board has become more complex. It’s a classic case of moving the challenge to the back end of the product cycle.
The Case For Third Party Repair
Technology is evolving, and evolving rapidly. In order to remain current and competitive, so must the skills, equipment, and knowledge of the repair providers that support it. The more sophisticated the design and the more advanced the technology, the more sophisticated the overall repair is likely to be. Some of these advances have made it easier to diagnose a defective part, but this is more than offset by cost and complexity, not only of the part in question, but of the technology needed to undertake the repair. The need for expensive equipment, highly-trained technicians, and tightly-controlled processes has become critical, and the business case for utilizing third party repair providers for cost-effective in-warranty, extended warranty and out-of-warranty repairs is now self-evident. OEM/ODMs specialize in design and manufacturing; retail organizations specialize in sales; warranty companies specialize in risk and contract management, but none are likely to nurture the core competencies necessary for post-sales, repair and aftermarket customer service - exactly what third-party repair providers excel at.
There are multiple technical and operational factors to consider when choosing to outsource your computer repair.
Source of Supply
A typical OEM/ODM will only manufacture a limited number of replacement logic boards to support the in-warranty cycle. When these parts are required outside of the manufacturer’s warranty period, the supply may not exist until it can be harvested from these same damaged or obsolete computers at a later time. During the extended warranty period or out-of-warranty period, proprietary components such as logic boards can be difficult to source. Generic parts such as hard drives, optical drives, memory, batteries, etc. can be sourced more easily. The challenge is to locate quality parts that are readily available at an economic price.
Internal Component-Level Repair
Component-level repair should be regarded as a key competency for any repair provider. It is, however, an exceptionally challenging service to offer, due both to the cost of equipment and engineering staff needed to support it. Board-level repair allows a repair facility to offer the ability to replace individual on-board components, rather than simply replace entire logic boards, video cards, etc that reside in the defective device. This has multiple benefits; including reduced turn-around time, reduced cost, and total control over the quality of the repaired assembly. The tradeoff is the expensive equipment and the intensive training required to be able to do this.
As BGA’s (Ball-Grid-Array’s) become more and more dense (packages can now have as many as 1400 contacts, or pads) the science of successfully removing and reapplying a replacement device is exceedingly complex. The objective is to apply enough heat to remove the desired device within the proper time-temperature continuum, without also removing any of the surrounding parts, or damaging the pads or circuit traces. Too much heat applied too quickly can risk damaging the device or the circuit board; too little heat for too long can have similarly deleterious consequences. This difficulty is exacerbated by a simple fact: when accounting for factors such as the number of pads on the device, the thickness of the board, the surrounding components, and the type and content of the solder, it is immediately evident that each and every board is unique. See figure 3.
It is also important to note that the ability to repair defective logic boards not only reduces costs but also reduces pressure on the OEM/ODM to provide a steady supply of spares during the warranty period. This results in reduced manufacturing and inventory costs.
Clean Room Capability
Clean rooms are a requirement for repolarizing LCD and LED panels and for tab bonding of LCD panels. A defective LCD panel on a computer or television can often be repaired by replacing the polarizing film and eliminating scratches or imperfections on the screen. Replacing tabs by bonding on a replacement tab circuit will often remedy horizontal or vertical lines of pixels that are not functional. See figure 4. Neither of these technologies, however, are capable of repairing isolated pixels or broken glass. Again, the ability to repair an LCD screen makes the service provider less reliant on outside sources, reduces costs, and enables greater control over the repair operation.
Tightly-controlled processes and accounting for customer assets as well as the service provider’s inventory are critical activities for managing any kind of volume repair. Capabilities such as automated order entry via EDI or FTP feed, barcoded status’ and tracking, customizable reporting at both a detailed and summary level, and electronic invoicing will streamline operations and help make repairs both efficient and cost-effective.
Local Mom ‘n Pop repair shops are predominately designed for addressing individual single repairs. Larger Contract Manufacturers, by contrast, are designed to manage thousands of identical devices. The ability to scale from low volume customers to high volume customers with a high product mix, therefore, will enable the service provider to offer flexibility but growth in tandem with your business.
The ability to repairing multiple brands and multiple vintages of out-of-warranty computers is dependent upon possession of both technical data and an accurate bill-of-materials. Repair companies with market longevity, particularly in a high-brand-mix environment are more likely to have these resources for older models requiring repair.
Whether you are an OEM looking for an in-warranty repair provider, a warranty company looking to have a 3rd party administer and repair your extended warranty claims, or a major retailer looking to outsource out-of-warranty claims, a 3rd party repair provider with the above capabilities can provide a cost-effective repair solution, and enable you to focus on your core business.