- This article is filed under:
- » View All
Reuse, Rebuild or Buy New?
by Winn Hardin, Contributing Editor - AIA Posted 09/18/2008
Remanufacturing is a multi-billion dollar industry in the United States. In many cases, the cost of remanufacturing a product is far less than new manufacturing, and even less than recycling a part into its constituent parts and then using those raw materials to make new product, particularly when energy costs are figured into the equation. The automotive industry, for example, has spawned billions in sales of reconditioned starters, alternators, and many other automotive parts.
When it comes to reusing, recycling or buying new machine vision equipment, however, total cost of ownership can cost a company a lot of money in the pursuit of ‘green’ alternatives. Sometimes – and the automotive industry is an excellent example – reusing vision systems makes sense. Many times it doesn’t. The best course lies in careful consideration of the details, your manufacturing needs, and a best guess at what the future holds.
Flexibility as a Selling Proposition
Flexibility is one of the great strengths of machine vision. A camera, optic, light and image processor can inspect tires or monitor steel smelting. However, as expert readers can attest, it’s highly unlikely that the same camera, optic, light and image processor would satisfy both applications.
Adding to the machine vision system’s flexibility is the length of time that vision vendors can support their products. “We get calls almost daily from customers that want to redeploy one of our legacy vision systems that we manufactured between 1992 and 2003,” explains Gary Kocken, Director of Sales for the Americas, PPT VISION, Inc. (Eden Prairie, Minnesota). “We have to make a decision on a case-by-case basis whether the system in question is redeployable per their new application, or if they would be better served by using our latest Impact Smart Camera System. When it’s a large installed base, perhaps in automotive or electronics, the answer is always ‘yes.’”
Despite the fact that machine vision systems are regularly deployed in dirty and dangerous industrial environments that one would think would limit the useable lifetime of the product, machine vision companies often feel that they are the victims of their own good engineering.
“It’s common for vision companies to support their product for 7 to 10 years,” adds Perry West, President of Automated Vision Systems Inc. (San Jose, California). “But some products simply refuse to die. Sometimes, the customer wants to redeploy a system used in one food inspection line to another line, but finding the original parts can be impossible, and new components won’t work in the old system. Those systems will need to be remanufactured, or maybe field retrofit with major components, to be able to be maintained…Obsolescence is a problem that people don’t talk much about in the machine vision industry.”
John Lewis, Market Development Manager at Cognex Corporation (Natick, Massachusetts) agrees that, in general, the pace of change is such that if you bought a vision system more than 5 years ago, it really makes more sense to replace it with current technology. However, he notes, “Cognex makes its vision systems upward compatible and offers a firmware upgrade program so that customers can use the new software tools we develop on old(er) systems.”
Wanted: Windows 98
When large customers with significant installed bases come to your company seeking support for discontinued vision systems, the answer is often a hesitant, “Yes.” Hesitant, because making old systems work is easier said than done.
Many older systems were designed before the advent of the Smart Camera with its embedded operating system. These PC host systems, along with their framegrabber and image processor boards, were designed around Operating Systems like Windows 3.0., 95, or 98, and for mother boards with ISA and PCI bus slots. “Our legacy products were built for 16-bit Operating Systems, whereas the newer Windows Operating Systems such as 2000, XP and Vista, are all 32-bit, so upgrading to the latest mother boards and OSs are not always an option. We’ve found it very difficult to source a motherboard that has the appropriate slots, with support for 16-bit drivers and a new copy of Windows 98 to upgrade to. It’s an expensive proposition, and many times customers don’t understand that upgrading older technology can cost as much as or more than new technology,” says PPT’s Kocken. “On top of that, PC host systems simply have more parts that can fail, may have compatibility issues with other machines that have newer operating systems, or are prone to viruses and occasional Windows crashes. A customer can spend $2500 to upgrade an older system, which is still using antiquated algorithms, and could be sidelined shortly after the upgrade due to other failed obsolete components, or $2995 on a new smart camera system that can do it all, is more deterministic, and can operate without the need for a PC host.”
John Petry, Marketing Manager for the Vision Software Business Unit at Cognex, contrasts, “it's often a good business decision for Cognex to support a customer request to upgrade a product to a new OS or PC bus. Particularly for our OEM semiconductor and electronic capital equipment suppliers that frequently have a "workhorse" product that they sell for years. In many cases, the engineer who developed the vision solution has moved on, so there's tremendous value to the OEM in extending the product's life. It's often good business for Cognex to help them, because the sales volume is high. As long as we can re-create the development environment, the work is usually manageable and the test efforts can be limited to the specific tools that the OEM uses.”
Reuse, Expansion, not Redeployment
In addition to large installed bases where economies of scale may tilt the equation in favor of reuse and upgrades rather than new technology, moving a vision-based vehicle ID system from one line to another, or retooling it for the latest model car, is one example of when reusing vision makes sense. Another case for reuse falls in the ‘mission creep’ category – but in a good way.
“After customers install their first vision system and become familiar with its power and flexibility, they tend to want to accomplish as many vision tasks as possible,” explains Cognex’s Lewis. “The hardware is already in place so adding that new inspection task to the vision system is primarily a software modification. Because this is more the rule than the exception, Cognex application engineers recommend that customers define the entire scope of the application to ensure that the vision system has enough performance headroom in terms of speed, accuracy, and acquisition requirements before specifying a vision system. It’s also good to consider any future requirements for increased throughput, ability to accommodate new product or changes to the existing product.”
When end users choose a vision system, they should be sure to pick one that can handle their needs with capability to spare, notes Cognex’s Petry. “Manufacturing environments are rarely static, so smart users will pick a software solution that can handle a range of lighting, shape and positioning changes over time, even if the up-front cost is slightly higher. The added payback is that vision systems that meet these needs are often the easiest to re-deploy for different applications, which means the return on investment may be substantially higher in the long run.”
As this discussion illustrates, there is no hard-and-fast rule about when to upgrade an older vision system, or buy a new system. West does, however, offer advice for those looking for a cheaper short-term solution.
“If someone has a 5 year old vision system and wants to redeploy it, they need to know they’re going to spend a lot of money re-engineering the system without any guarantees about how long the new system will be maintainable,” explains West. “It’s like a decision to buy a new car. You could take your old car to the shop and have the transmission rebuilt, but who’s to say the engine won’t die in 6 months only to discover that the parts needed to fix the engine are no longer available.”
There are currently no comments for this article.
Leave a Comment:
All fields are required, but only your name and comment will be visible (email addresses are kept confidential). Comments are moderated and will not appear immediately. Please no link dropping, no keywords or domains as names; do not spam, and please do not advertise.