In early November ralphjy told us that his NVR had died. In the course of the discussion and Ralph's experiments it was speculated that there was a short on the board that was pulling down the power supply. Over the years as I have serviced equipment this has always been a challenge to find the point or component that is the source of a short on a circuit board. In the past I have followed traces and cut them as needed to step by step find my way to the source of the problem. I have often thought how nice it would be to just have a magic viewer that would highlight the source of the short. The challenge of finding a short on a board has only gotten more difficult as the boards have gotten more complex.
In early November, just as Ralph was having the NVR failure, I had pre-ordered an inexpensive Thermal Imager from a Chinese source. I had no idea what to expect but the price was right and I wanted to see what it could do. The Imager arrived today and so I read the manual and familiarized myself enough that I could do some simple experiments. Here is the Thermal Imager.
My plan for this experiment was to find a big simple circuit board with a fairly long trace that I could put some power into and see if the Thermal Imager could see it. The board that I selected was originally from a TV and I had already pretty much salvaged all the parts from it. After spending several minutes trying to find a long trace with good connection points I gave up and tried a different approach.
My next approach was to put one lead of the ohmmeter on a jumper wire and then drag the other probe across the board hitting the other jumpers and watching for a meter deflection. After a short time I was rewarded with a needle jump from my analog meter. When I explored a little further I was delighted to find that not only were the two contact points across the board from each other but there was actually a diode between them. Since there were a few scatter components still on the board I had luckily found one that had a component in the circuit.
The next step was to see if my new Imager could find the diode that tied my two contact points together. At this point I had no idea where it was. I hooked up my power supply as you can see on the picture above so that my current from the power supply was forward biased through the diode. With the power supply on constant current I brought the current up to 250 mA. The voltage was about 1 volt. I then used the Imager to scan the component side of the board for hot spots and I was not disappointed.
At this point you can see that I got what I paid for with respect to the resolution of the Imager. Between the two images I was however able to identify the diode in question. Good old D155 was where the heat in my image was coming from. I looked at the trace side of the board with the Imager but the 250 mW of power that the power supply was dumping into the circuit was pretty much all showing up in the diode.
The next step was to take a jumper wire and short across the diode. At 250 mA I was pretty certain that I would not be able to put any more current into the diode without frying it. With the jumper in place across the diode I brought the current up to 1.5 amps and about 0.7 volts. This put about a Watt of power into the circuit and I knew that a lot of it would be going into the traces connecting my two power attachment points. Now when I looked at the trace side of the board there was a definite green glow that came down the left side, across the bottom and then snaked its way to the vicinity of the second contact point.
I have to admit that I was hoping for a little better resolution than I am getting but it is a whole lot better than I had before. At least now I have a general direction and area to deal with. If it was important to actually identify the trace in question I would have to get out a fine point marker and use the old method of marking it as I followed it with my eyes. Over all I am pleased with this experiment and the new Thermal Imager. I am retired now and the volume that comes across my bench is a trickle compared to when I was working. This Imager is perfect for me at this stage and it will give me an additional diagnostic tool when I am occasionally looking for a short on a circuit board. If I was back doing the volume that I used to do it would probably make sense to spend a little more money so that components and traces could be more easily resolved in the images.