Software for OCR'ing paper tape?

Hi It is no secret. I've posted it to this group before. First, consider how useless an ohm meter is. Usually something like a shorted capacitor is maybe 20 times the resistance of the ground plane. Looking for differences in the resistance across the ground plane is usually a waste of time. Most just use the high current method of making the shorted part hot or smoke. This is not always successful and can damage the board. The trick I use puts no current through the short. There is no need. First lets describe the problem if the short is between two traces and then the more complicated case that it is between to planes. Take a bench supply that can produce a couple amperes in the current regulated mode. Pick one of the two trace and connect the supply from end to end on that one trace ( we'll call it trace1 to keep from getting confused ). Place one fixed lead of the voltmeter on trace2( location isn't important ). With the other lead, of the meter, probe along trace1. When you reach the location where the meter reads zero volts, you've found the short. In some cases, there may be a lot of shunting resistance In this case, gradually move the contact points that you connect the power supply towards the zero volt point and also move the fixed meter lead. As you move the supply leads in, you'll pass the shunting resistances and get a truer location of the short. It is rarely necessary to do this on most solid state circuits. Now for the slightly more complicated case of power planes. Place the supply leads at two diagonal corners of the board. Use more current on a power plane( 5-10 amperes is usually good ). With the current across plane1, place one fixed lead on plane2. Now probe along Plane1 until you find the zero volt location. In this case, it usually forms a curved line. I usually just lay pieces of tape along this zero volt line. The short is someplace along the line. Now, move the power supply to the other two corners of the boards plane1. Repeat the finding of the zero volt line. Where this line and the previous one cross is the short. I've used this method in a production environment and maybe found several hundred shorts. It does take a sensitive meter, that is why I go for a 5 digit. I suppose it would work with a galvanometer used on a bridge as well but never tried that. At one time, I'd considered making a high current sine/cosine current source. One could then use an oscilloscope and just probe for a minimum circle or elipse. This would speed up the power plane locating. Dwight