Scope probe article

Evaluation Engineering has a great article about scope probes from a guy at Agilent. Jae-Yong Chang gives a pretty good overview of the issues surrounding scope probes. I would add two stories of my own. He points out that making differential measurements can be done by inverting one channel and adding that to the first channel. He points out that this is bad but does not really elaborate. I chased a little bump on a fly back converter waveform for three days, trying to figure out what was wrong. This was bad on two levels. First off, the power supply met all the design requirements. Nothing was getting too hot and all the ripple and regulation specs were met by my design. I just didn’t like the look of the fly back waveform since it did not look like the one in the textbooks or app notes with that little bump in the decaying voltage. Since this was a line-operated supply I was using the differential probe trick using the invert function. Well the two channels are never the same and the little bump was just a problem with AC common-mode rejection. I should have put an isolation transformer on the line and grounded the fly back circuit common node. Then I could have measured it single-ended and I would have seen that it was fine. Maybe new scopes that invert the signal digitally can work better in this phony differential mode but just go buy a diff probe.

This leads to the second “tip”, if a way to kill yourself can be called a tip. The other way to make a differential measurement is to float the scope. You plug it into on of those three-prong to two prong converters and tape off the little ground wire. The more directly hacker barbaric can just cut off the ground lug of the power cord. Then you put the scope, in my case an old 100 MHz Hitachi that worked and triggered great, on a piece of cardboard. When you hook the scope probe ground to the fly back common, the entire scope is hopping up to 170 volts. Don’t touch the metal cover. Another tip is to put some masking tape on the “delayed B” time veneer since that is a metal knob and touching that will kill you as well. Then just be really careful touching the plastic knobs, wear tennis shoes and maybe keep one hand in your pocket. When I did this I saw that the flyback waveform was really perfect. Of course the real solution is to buy a differential probe for a grand or two and make sure it has at least 200 volts common-mode rating so you can hang it across the line. Some FET differential probes are only good for a few volts. Don’t hold me, my mentor that told me about the cardboard under the scope or EDN, its associates or assignees responsible if you kill yourself. Happy probing.