When I was in high school a long time ago, I was one of three students in charge of the school's video system. This consisted of an open reel VTR, some audio gear, an RF modulator and a B&W camera. The camera was a modified surveillance camera which had a string of 2N706A transistors operating as the video amps. These transistors were socketed, and the sockets had little if any plating on them so the camera would occasionally get intermittent.
One day I was pulled out of english class to fix an equipment problem in the drama class across the hall. When I arrived there I found the students gathered around the dead camera, waiting for someone to fix it so they could continue with their play. I knew about those intermittent transistor sockets so I approached the camera and gave it a good whack on it's right side. A few of the students gasped at my actions but the camera sprang back to life! I received a hearty round of applause from the class, took a bow, and went back to my english class.
Back long ago when Apple came out with the Apple ///, they used sockets for the RAM and EPROM chips. The sockets were not quite the right thing, and with shipping vibration and maybe thermal on-off stresses the chips would walk their way out of their sockets. Apple recommended that if the computer got flaky, you should try picking it up and dropping it like 4 inches. This was a kinda heavy box, like 30 pounds, with non-squishy rubber feet, so if you do the math, 4 inches of acceleration, followed by aprupt decceleration over a distance of maybe .04 inches, that's 100G's pressing down on the chips. Probably did the trick, but oh man, what a cheesy repair technique.
Many long years ago, prior to my engineering career, I was a service tech for Sony corp, servicing recorders, mostly. We had a very standard service procedure, and about the third step was to slap the tape deck while it was playing. The slap would reveal a number of problems, the most common being a bad record-play switch. If the slap test revealed a noise, the next step was a blast of contact cleaner, which, if that fixed the problem, indicated that the switch needed replacing because the silver plating on the contacts had worn off and the switch had become intermittant. The slap would also reveal a broken connection to the tape head, which that piece of #30 stranded wire would beak under constant flexing if the routing was not exactly correct. And a different sort of noise from one channel would indicate that a soldered board connection had broken and needed to be resoldered.
So that gentle slap would provide a diagnosis of many faults, with a minimum of tech time.
It's always a problem. Had two similar experiences just in the last week. My PC (an ancient XP machine I cobbled together with begged and scrounged bits) is a bit flaky and sometimes fails to boot. Percussive maintenance usually works. I have been running it with the case off and by a method similar to that in the story, determined that it's the video card that is the problem Since I reseated it in the slot, no more problems (cross fingers, grab nearest bit of wood :-)
And my label printer exhibits similar problems. Take battery cover off, rotate all the (AA) batteries half a turn or so, and it's good for another few labels.
If electronic equipment never failed, I'd be out of a job. So long may it continue!
Caleb Kraft- I certainly hope the moving pcb problems have gone away, no doubt aided by higher levels of intergration and surface mount. However, here are two similar that happened to me today. Went to use the scanner, up came the message your C4280 is not connected. I knew it (expletive deleted) well was, so I rebooted still the same then found that in moving a nest of cables that links this office together I had moved the USB slightly out of its socket. I know, go Wifi.
Well decided to watch my Smart TV and although it responds to my voice and various hand waving signals I decided to use the remote. Nothing happened, couple of sharp taps on the table and the connection between me and the world of TV was restored. (the old battery/spring problem)
Wise and no doubt experienced. I can remember visiting an instrument (scopes and the like) hire company in the UK. Their business model required that they kept detailed records of repair call-outs for the equipent out on hire. Those records showed that irrespective of original instrument manufacturer, most repair call-outs were in the first few days after delivery. While the hire company diligently serviced their customers requests one of the engineers at the hire company gave the following insight, "In most cases a good bang on the case would have solved the problem". The problem that is of boards and connectors moving in transit. It is to be hoped modern instruments have for the most part dealt with this problem.
Blog Make a Frequency Plan Tom Burke 17 comments When designing a printed circuit board, you should develop a frequency plan, something that can be easily overlooked. A frequency plan should be one of your first steps ...