A while back a cheapie digital multimeter I have started to become intermittent so I thought I might get a decent bench top multimeter. I looked around and decided on an HP, no Agilent, no Keysight as they are now called model 34450A 5½ digit one.
I have used it since then doing the normal type of measurements.Yesterday however I was testing a board that I had assembled and used a hot air rework station to solder all the components down, as my reflow oven is currently inaccessible due to all the stuff in my workshop as we are moving home early next year and our house is about to go under the auctioneers hammer. When the board was powered up, it drew copious amounts of current, so I checked the power to ground resistance, and bingo, there was a “dead short” between the 3.3 volt rail and ground.
Shorts on power rails are never much fun to find, and while thinking about the problem, I realised that my new DMM would measure milli-ohms so I thought I might be able to find the short by measuring the resistance at various places on the board. I started out measuring in a few places, and sure enough there were differences in the values, but none really close to zero. I thought what is the resistance of the test probes, and found that there was about 40 milli-ohms in the leads, so I thought ok I can subtract this from the reading.
Then I had an “Aha” moment, the DMM has a null function, so I pressed the tips of the leads together and hit the “Null” button, and after a couple of seconds the display changed to show not only the nulled resistance, but also the total resistance – beauty. I started probing the board, and although I was getting fairly low readings, the readings were staying around 80 milli-ohms. Then I probed on the other side of a small inductor that was feeding a 3g/4g modem and the resistance dropped to almost zero. I knew I was onto something here, and because there were few components on this side of the inductor, I was able to remove the 3 or 4 small components and prove that the short was either under the modem or in the modem. Once the modem was off the board, I could see that when I had reflowed the modem, I hadn’t heated it enough and there was still some unmelted paste there. I cleaned the paste up and checked again and the board now showed open circuit. I had found and removed the short without having to cut tracks to isolate and find the area the short was in, which made me really happy, as I really dislike cutting tracks on boards.