Each time I meet or speak with any of the T&M equipment houses, I hear one variation or another on the theme of getting design and test engineers to upgrade their bench instruments. Of course, these folks are in the business of selling you new scopes, spectrum analyzers, and DMMs. If they could, they’d sell you new ones every year but times are tough and budgets aren’t what they could be.
According to my contacts at T&M vendors, it’s not at all uncommon to walk in to a customer’s test lab and find scopes and other instruments that are 10 years old or more. If you’re designing, let’s say, garage-door openers or some other more-or-less static sort of system that doesn’t change a whole lot over time, that 10-year-old scope is probably good enough. That’s assuming, of course, that you do what’s necessary to keep it calibrated and running up to spec.
But what about those of you who are closer to, if not on, the cutting edge? Let’s say you’re debugging a PC board with a 3-Gb/s second-generation PCI Express link. We know that your 10-year-old scope just isn’t going to get the job done. How about a five-year-old scope? Does it have enough bandwidth, or enough built-in analysis capability? Is the noise floor low enough? Your aging scope is likely very much obsolete in terms of serial analysis capability, bandwidth, update rates, and other critical parameters that are going to get between you and a valid measurement.
And if that’s not hairy enough, consider the debug of a SERDES link at 10 Gb/s or more. Now you’re going to have trouble getting a good look at even the third or fourth harmonic unless you’ve got a whole lot of bandwidth, and that probably means that you’ll need an instrument of no more than two or three years’ vintage.
One way around the dilemma is for a test lab to have at least one high-end scope that basically lives on a cart and travels from workstation to workstation when someone has a measurement to make that's out of bounds for their bench unit. If this is the case for you, how frustrating is it to have to wait in line for that cart? Another option is to rent a high-end instrument for the debug of high-speed serial links, transmission lines, and other performance-critical aspects of your design.
I had a visit from some folks from Agilent recently who told me that test-equipment budgets are beginning to loosen up of late. There was a bit of a surge in 2009 and 2010 with the early wave of DDR3 and SERDES designs. Now, I’m told, hiring is on the rise in some systems houses and a new generation of young engineers is starting to find jobs. That means outfitting these young guns with test gear.
In addition, universities are gearing up to train those young engineers. I just learned that Tektronix has donated a pile of test gear to a new $43.5 million Engineering and Computer Science building on Washington State University’s campus in Vancouver, Wash. The WSU Vancouver expansion is part of an economic development project to provide additional workforce-ready engineering graduates for companies in the local high-tech cluster.
So where does this leave you? Are you struggling to get by with obsolete equipment? And if so, how are you managing to get the job done? How are budgets for equipment upgrades in your workplace? Let’s hear from you: Who’s got the oldest scope on their bench?