Control Engineering

Charlie Masi

When is a piece of equipment or system obsolete?
March 30, 2007

Prof. Hal Mahon, with whom I did some graduate work at UMASS/Boston, once said: “Nothing is obsolete if it does the job.” He said it when I raised my eyebrows over using 30-year-old vacuum tubes to build an RF receiver for a state-of-the-art nuclear quadrupole resonance experiment. It was the late 1970s and you could no longer find a 12AT7 dual triode in any catalog. I believe you could still get them from Raytheon, but maybe they were just selling off old stock. Hal and I had to raid our caches of World War II surplus radio equipment (spelled “J-U-N-K”) to find working units. I was popular for being one of the few graduate students who knew how to design circuits with them.

We used the venerable old things because the receiver we were building was going to look directly into the output of a several-hundred-watt pulsed radio transmitter. The high-gain transistors of the day would have instantly turned to trembling globs of molten silicon when faced with such abuse. Those clanky old vacuum tubes, on the other hand, took a microsecond or so to shake off the shock before getting right back to work.

In an era of light-speed technology development, we’ve grown to expect that equipment will end up in the trash bin long before it stops working. I keep a couple of VHS tape players around because my wife and I had collected over 300 titles in our video library before DVDs became common. At $10 a pop, it would cost over $3,000 to convert all of that to disks—if you could. (Try finding a DVD copy of Those Magnificent Men in Their Flying Machines…!) Sometimes “obsolete” equipment is all that will do the job.

Sometimes a technology works so well that it’s impossible to improve. The hammer is likely the oldest of simple machines, having been invented the first time some paleolithic pre-human (the Geico cave man?) wanted to smack a mastodon femur really hard and picked up a rock to do it with. We’ve added a handle and changed the materials a bit, but the principle hasn’t changed.

That’s enough encouragement for the Luddites out there. Clearly, things do become obsolete. Try buying a new black-and-white TV. Folks generally don’t make ‘em ‘cause folks generally don’t want ‘em. Color TVs do the same job, only better. When that happens, we call the old thing “obsolete.”

In the consumer sector, obsolescence seems to be a good thing. Consumer technology buyers want the latest and greatest of whatever they have. To paraphrase comedian George Carlin: when something “new and improved” comes out, suddenly what we have is “old and yukky.” No matter that we bought the thing last month and it has a service life of 15 years, we’ve gotta dump it ‘cause it’s “old and yukky!”

Obsolescence is a problem for industrial equipment users, however. After spending several hundreds of thousands, or just as often millions, of dollars building up infrastructure based on earlier-technology equipment, dumping it as “old and yukky” just because something “new and improved” has come out is fiscally irresponsible. Not only would you have to write off the old equipment (which the accounting department would rather depreciate over another two or three years—or more) and lay out the cost of the “new and improved” stuff, everyone in the plant would have to drop what they’re doing (hopefully, productive work) to take training classes on the “new and improved” stuff (which is definitely unproductive). That sort of thing can put companies out of business realfast!

There are a limited number of things that can make it worthwhile to render something obsolete—that is to make you want to dump it in favor of something newer:

The need for it goes away: The need for buggy whips went away when people switched from buggys for regular transportation to automobiles.

The cost of maintaining it gets too high: As equipment reaches senescence, keeping it running becomes an increasingly expensive proposition. Eventually, it becomes less expensive to scrap it and replace it with new equipment.

Newer technology vastly improves the process: The new technology makes the process run so much faster and more efficiently that the money saved (or increased money made) is enough to pay for the cost of switching over.

We like the third situation. It has a positive spin. Technology developers always want you to think their “new and improved” stuff is so superior that it’s going to pay for itself in no time. In truth, however, technology doesn’t develop that way. It’s a process of continuous improvement with few breakthroughs significant enough to justify scrapping functioning equipment.

Established systems include a large number of interacting components that, at any particular time, are at different stages of their service lives. As they fail one by one, you should replace them with the most advanced equipment available. In this way, the system undergoes constant rejuvenation so that its service life vastly exceeds the service lives of its components.

To make this process work, it is necessary to have forward and backward compatibility between equipment generations. The cost of a “new and improved” dingus is a lot higher if it can’t work with the “old and yukky,” but serviceable, equipment surrounding it.

C.G. Masi, Control Engineering Senior Editor, charlie.masi@reedbusiness.com

Update on cyber security standards development

Bryan Singer, (CISM, CISSP, FluidIQs Inc.) talks to Peter Welander about the work of the ISA 99 committee as they develop the Industrial Automation and Control System Security Standard. Singer, chairman of the committee, explains where the standard is now and where they’re headed as this important document takes shape. He puts the discussion in the context of current security thinking and practice for control system operators. Recorded at the PCSF, Atlanta, 2007. (The timing is 20:35.)

Download this podcast at http://www.controleng.com/index.asp?layout=audio&element_id=2140041028

Posted by Charlie Masi on March 30, 2007 | Comments (0)