Design Tips

This issue, we have a humorous look at “ground” from Dr. Howard Johnson. Dr. Johnson frequently conducts technical workshops for digital engineers at Oxford University and other locations worldwide. You can see many more of his technical articles if you visit his web site at www.sigcon.com.
Please send me your most useful design tip for consideration in this section. Ideas should not be limited by anything other than your imagination! Please send these submissions to bruce.arch@ieee.org. I’ll look forward to receiving many “Design Tips!” Please also let me know if you have any comments or suggestions for this section, or comments on the Design Tips articles.

In a dimly lit room, two engineers stand on opposite sides of a table looking at the artwork for a printed circuit board.
“It looks big, solid, and strong,” says Simplicio.
“You fool,” replies Magneto, “the ground plane is peppered with many holes, yet it carries great currents. I believe the plane is weak and cannot be trusted.”
“Then I shall test it,” says Simplicio, defiantly.
Simplicio trundles off to his lab. He returns with a DC ohm meter and a physical PCB. He connects the meter to two points at opposite sides of the PCB and exclaims, “The resistance of the ground plane is so low that I cannot measure it! This board indeed has a wonderful plane structure. I can depend on it to ground all manner of things.”
“A fine experiment,” says Magneto, “but what kind of meter did you use for the test?”
“Why, a Fluke meter, of course, the same brand we had at the university where I studied.”
“And how many digits of precision does this meter show?”
“It shows four digits, but only two below the decimal point. And when measuring small resistances the last digit does seem to wander around a bit.”

Magneto sneers at his friend and laughs, “So, the smallest resistance you can measure is 0.01 ohm, and you can not even do a very good job at that, can you?”
“I can do better than that,” says Simplicio. So he goes once more to his lab and returns this time with a huge power supply. He configures the power supply to force current through the ground plane from corner to corner. Even such a large current produces voltage differences across the plane of only a few tens of milivolts. Convinced that such tiny voltages could not possibly perturb his digital circuits, Simplicio proclaims, “You are insane, Magneto. There is no problem with this plane. I have used such planes many times before and they always work.”
Magneto cannot believe his ears. How could there ever be such an idiot? Has he never heard of high frequencies? Or inductance? Has he no idea the sensitivity of RF emissions testing?
The two men argue long into the night. Magneto, master of the RF domain, lacks hard data to back up his assertions. Simplicio, captain of the quick and dirty experiment that proves nothing, prattles on about schedules, costs, and a thousand excuses why he cannot change anything.
Simplicio and Magneto will never arrive at a satisfactory solution because neither comprehends the whole problem. Simplicio sees only the digital functions and the cost. He has no vocabulary for understanding low-level signal coupling issues. Any signal less than 2 mV is, from his perspective, invisible—that limit being dictated by the sensitivity of his oscilloscope. He does not realize that 2 mV, if coupled to the world outside his product chassis, radiates like the blinking beacon atop the rock of Gibraltar. Two milivolts, in the world of RF affairs, looms larger than an elephant.
Magneto, on the other hand, sees clearly the RF world, but forgets his responsibility to educate Simplicio and to show, through direct experimentation, the correctness of his assertions. Digital engineers respond well to direct evidence—if you have the time to collect and organize that evidence.
Do you have that time? Does your company provide opportunities for communication with your digital counterparts, or resources to facilitate experimentation with your products? If not, show this article to your manager, for you risk falling into the deadly cycle of Simplicio and Magneto1. EMC

1 Simplicio, who refused to change his layout, left the company and shortly afterwards won the California state lottery. Magneto is still working to fix the design. Moral: the race goes not always to the swift, nor the battle to the strong (Eccl. 9:11).