As for the switches... why resist...? I bought them cuz I needed 6 of the metal ones to fix that curb-score Samsuck TV; and we both know there's no way in hell the rest of the kit will go unused in any of our labs. For 10 bux, a total no-brainer. ![Confused :-//](https://www.eevblog.com/forum/Smileys/default/confused0024.gif.pagespeed.ce.5xOqKkq0Co.gif)
EDIT: I'm bringing this to a new post as it really just made a dog's breakfast of the "cheap switches" conversation, and is actually something that's ON-TOPIC for a change.
That reminds me... this TV had the weirdest failure mode; I actually meant to post aboot this when it happened, but IRL (and the fact I lost the pics from when I was working on it) got in the way.After the usual decontamination with shop rags and alcohol,
I put the TV in the LR to use as our daily driver, as I've found that usually brings Murphy a-runnin' and is the best way to catch consumer electronics actually fucking up. ![Laughing :-DD](https://www.eevblog.com/forum/Smileys/default/smiley_laughing.gif.pagespeed.ce._hfWAz_QHO.gif)
It worked fine for a few days, then started randomly flicking from input to tuner, then randomly changing channels, or input to input, and argued with me aboot which input I chose when I tried to put it back. Turned out there was crazy noise on the IR sensor signal line, but not on the 5V rail that fed it. Further investigation with my 54645A showed
the same noise on the signal line from the front keypad (which shared a ribbon cable with the IR sensor to the uPC), only at higher amplitude. Unplugging the keypad made it go away.
![](https://www.eevblog.com/forum/testgear/test-equipment-anonymous-(tea)-group-therapy-thread/?action=dlattach;attach=938592;image)
The weird bit was that there were no actives on the keypad; just the tactile switches and pulldown resistors of various values to let the uPC know which button was pressed per the voltage on the line.
![Shocked :o](https://www.eevblog.com/forum/Smileys/default/xshocked.gif.pagespeed.ic.CrwtrNUjLO.png)
I ran out of time at that point due to IRL, so popped the cover on the back and just used the TV for a week with the keypad removed. After confirming the problem gone, I got around to visually inspecting every solder joint and testing every switch/resistor on the board, and they all tested right on. Next diag session I plugged it back in and was getting crazy voltage readings; probing the keys with my scope showed noise which changed just from moving or tapping the board.
On a board with zero active components. And shouldn't have any closed circuits. ![WTF? :wtf:](https://www.eevblog.com/forum/Smileys/default/wtf2.gif.pagespeed.ce.pxLiY5pjs4.gif)
After staring at the fukkin' thing and scratching my head for aboot 10 minutes, I went to visit the porcelain throne (like many of us, I seem to do my best cogitatin' there
![Wink ;)](https://www.eevblog.com/forum/Smileys/default/xwink.gif.pagespeed.ic.cldandycH0.png)
); during my visit it occurred to me there might be a intermittent short/contact in one or more of the tactile switches.
Went back to it and looked at the switches under a magnifier, and whaddayooknow... the top tin part just pops off. So I opened 'em all up and inspected, expecting to find a bit of dirt or rust inside one.
What I found was actually much stranger.On three of them, the silvering on the contacts had migrated ( I assume from electrolysis) such that it formed a very visible spiderweb of metallic tendrils across the surface of the plastic body (these appear to be made of some filled nylon or resin plastic) of the switch between the contacts. This didn't actually appear to have continuity on the one I removed for testing,
but it did have measurable resistance and capacitance if I put the silver-plated dome cup back in place.Having figured this out, I carefully scraped/cleaned all the switches with alcohol to remove the silver deposits and reassembled the TV with the PCB installed; the noise was gone on both data/signal lines, and it worked flawlessly until aboot a month or two ago, when I bought those switches and just replaced them all.
My best surmise is that moisture in the air promoted the electrolysis/silver deposition due to the constant 5VDC applied; once it got to a certain point the physical structure of the switch became a LC and/or RC oscillator. Now have 2 or 3 of them doing that at the same time and... batshit crazy TV.
![Shocked :o](https://www.eevblog.com/forum/Smileys/default/xshocked.gif.pagespeed.ic.CrwtrNUjLO.png)
mnem
![Tweak :-/O](https://www.eevblog.com/forum/Smileys/default/smiliey_cal.gif.pagespeed.ce.yYMcAvxBJk.gif)