Yes one should analyse the actual problem other than just solving it and becomming happy as in we have done a lot of work .I remembered once when i was in university and was making one of my unmanned ground vehicle after a particular circuitry was done i tested the circuit and it wasnt running i digged out that the problem was with one of my IC which got burnt i removed the IC and replaced it again the circuit was not working after testing same issue was comming with the circuit i again removed the IC and replaced and thrice circuit wasnt running after 3 hours i came to know that i was attaching the Same burnt IC again and again thats why my circuit was not running and my time got wasted .It was a very bad plus a funny experience too .
Moral of the story is always throhe burned componenets instead of using them again and again.
This is a lesson for anyone who fixes anything -- all the way to replacing a light bulb. We always assume that if we replace a part and it still doesn't work, there must be another cause. I suppose the lesson is that where's there's smoke, there's fire. If one part is bad, there's a chance that other parts in the same lot could be bad, too.
Yes, Nancy, I can see how this would become a one-time problem. Once you've discovered that the replacement part for a faulty part may also be faulty, it's bound to stay with you from then on. This is likely a one-time incident.
Rob, the problem of component quality and test is one that has been around for a long time. It is a numbers game. Many years ago, my father was in charge of component procurement at a government electronics lab. Parts had a shelf life, and after that they were discarded. Employees could pick them out of the bin, but they could not be sold or given intact. Nedless to say, I got a lot of them to "play" with implementing and designing circuits. Of course there were bad components, but it was not a large percentage. Since I was just doing this to learn and for hobbist purposes, it had little effect. On the other hand, the engineers who depended on them would waste lots of valuable time if they had use them.
I would worry about what was shorting the diodes. Parts don't usually fail without some sort of mechanism, especially a diode shorting. If a replacement blew right away I would be suspicious that something was making it blow.
I had an audio power amplifier that lost its finals. You can always assume there was some sort of abuse at the output, so I simply replaced the transistors. Everything worked well, but then I cranked the output to clipping with the output lightly loaded, the output transistors instantly shorted. I did a little digging and found that the VCE breakdown on the transistors was being exceeded when the amplifier wasn't loaded. As long as there was a load the brute force power supply would sag enough to prevent breakdown, but in absence of a load, the higher voltage swing would cause the transistors to exceed their breakdown. Fortunately there were transistors in the same family with a higher breakdown voltage. I replaced the finals with the higher breakdown transistors and the amplifier served the band as a monitor for another 20 years.
I agree, Rob - maddening. The only good thing to come out of that type of situation is the realization that it is actually a possibility, no matter how remote. I've been there, done that, got that t-shirt of spending a lot of time troubleshooting a board after replacing a component - only to find the replacement component was also bad. But afterwards I never wasted half a day again before verifying that the replacement component was good...that falls in the same bucket as the time I spent thirty minutes troubleshooting a circuit without verifying first that the power supply was working to the board...oops, did I admit that online?!!!! Of course that happened when dinosaurs still roamed the earth, but both incidents have stayed with me to this day...
This is quite a difficult problem. We've seen this before in a Gadget Freak. A component is identified as faulty. It's replaced and that solves the problem. Never in a million years would guess that the new component was also faulty. Maddening.
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.