[R-390] Bad solder joints in other projects (long)

[Scott, Barry (Clyde B)]

Warning: Do not proceed if sand hurts your eyes...

I had an interesting experience this past weekend.  I built a Super CMOS III keyer a few years ago.  I originally housed it in a plastic enclosure and was having problems with it going bananas when I tried to use it with the transmitter at full output.  Apparently it was a problem with some small amount of stray RF getting to the keyer's components.  The designers suggested I move it to a metal enclosure which I did.  It solved the problem of the keyer losing its mind, but I created another problem.

The keyer has the capability to drive a small speaker for side-tone purposes.  It uses a 2N2222 in a configuration where the collector is at supply potential, a square wave is supplied directly to the base, and the emitter is connected through a 100-ohm resistor to the speaker to ground.

When I first powered it up, the internal side-tone did not work.  Hmmm.  It worked before.  What could have gone wrong?  If I applied a solid stream of dits or dahs, the speaker would sometimes start to work.  If I switched it off and back on again, it would not work again until it was good an ready.  Something was wrong.

I looked at the circuit on the scope.  The collector had a constand 4.6VDC which was good.  The base was seeing a 4.6VDC square wave.  Fine.  The signal on the emitter, though, looked really strange.  The positive half of the square wave was okay, but the "zero" half, did not look right.  It would drop off rapidly from 4.6VDC, but would then slowly drop all the way back to zero.  It would do this until the side-tone began working and then it was a nice square wave once again.

Thinking the 2N2222 was going bad, I replaced it.  The symptoms remained.  What could be wrong?  My wife suggested it was a bad connection.  Ha!  How could she possibly know what was wrong.  In my infinite wisdom, I discounted that theory and kept looking for the answer.  As I continued to work on it, suddenly I found that if I moved the circuit board a certain way, the speaker would work and if I moved it back, it would stop.  That was it!  A bad connection somewhere.  She was right!!

But where was the problem?  The connections all "looked" okay.  I finally found that the wire going to a switch I installed to cut the speaker off could be moved one way and back and the problem would appear and disappear.  I had found it.  Freshening that solder joint fixed the problem.

All that over a cold solder joint!  Oddly, I think the strange "square-ish" wave I was seeing was the result of the signal "discharging" through the capacitance of the cold solder joint.  Don't know.  At any rate, it was one of the few times I have ever seen the effect of a cold solder joint.  I never suspected it would manifest itself in such a way in a digital circuit.

Just thought I would toss this in the mix as Jim's problem seemed to stem from a bad solder joint as well.  Is this a fun hobby or what?!

73,

Barry(III) - N4BUQ