What was your best failure when you powered it up? Everyone with any experience has that DOH! experience at one time or another. I'll start it off.

Back when Commodore PETs were new (pre Commodore 64's) I built an analog joystick adapter for mine using 4 555 timers, to convert the variable pots to digital signals. It had 4 miniature audio jacks, mono instead of the now more common stereo. Looked pretty good, except for one little thing. In my anxiousness to finish the project I wired the transformer (120AC/6.3AC) backwards, and put an overrated slow blow fuse in the circuit. Again, this was before wall warts, you pretty much had to build a power supply if you needed one. So upon power up all 4 jacks blew thick streams of smoke, very impressive. The project was a total bust, the reversed transformer did what transformers do, boosted the output to 2.3KV before frying, and in the process frying anything that had a PN junction in the rest of the case. Opps.

 

I had a 6" x 18" board with about 15 tantalum caps all on the bottom. Seemed we had a problem with them being in backwards, or being overstressed.

First one burst into flames, which were coming around the edge of the board.
Powered it down, then blew it out like a birthday candle.

Next day another one exploded. Over my shoulder, and bounced off the wall across the room.

It took us about a week before we felt safe to have a close look at the bottom of the board.

 

Tantalum Caps sometimes get very noisy before they go bad. I was looking for such a case on a test fixture with cold spray, and the sucker caught fire. Since I would have to fix whatever it damaged, and I had the spray in my hand, I used it like a fire extinguisher. Those suckers burn like match heads when they go up.

 

When I was building a regulator for my solder pen, it because of some reason didn´t work. No heat or flames, but it just didn´t power the pen.

So I left the circuit on and started measuring at various points. After a while of trying to find out the problem, I accidentaly shorted a 7910 regulator with the tip of the probe. It didn´t go open, but to make it funnier it became a short.
On the way out it also took a rectifier diode on input, so suddenly there was cca 35VAC from a 120W transformer on a 2200uF/15v cap.
The cap was a LITTLE upset, so it blew all its inwards over the bench with a loud BANG, and I jumped up in shock away from the table.


Ever since that I am really cautious when measuring power supplies and other high-current things.

 

Never hook up a 12VDC motor with blocked rotor to a 120VAC source. And does anyone know why the smoke was yellow?

 

About 30 years ago I decided to replace the wimpy little radiant heater in my bathroom with a fan-forced heater, controlled by a separate 24 hour timer. After installing the timer and the heater, I turned the breaker back on. The timer made this loud growling noise, the fan put out a gale-force wind, and after a few seconds, wind-driven flames came out of the heater! In a panic, I turned off the switch, and after a little head-scratching, it dawned on me that the old heater was powered by 220V. DUH! Why I had assumed it was 110V, I don't know.
Of course, the new heater was wired for 110V, and the timer was 110V. The timer and the fan motor were destroyed. I managed to salvage the heater by moving the resistance-wire terminals around so that they were in series instead of in parallel, I found a 220V fan motor, but it had a smaller diameter shaft, which I fixed to accept the fan hub by making a shim out of some Mylar sheet stock. I gave up on the timer, because I couldn't find one which ran on 220V. I was tempted to get another new 110V unit and run it off one phase of the 220, but the cable in the wall was 2 wires plus a bare ground. I would have had to use the ground as the neutral for 110V, which I was not willing to do.
I'm a pretty good EE, but a lousy electrician.

 

I was working on the stamp perforator for the university. It had a zillion tiny pins used to make the university logo on each Post Office stamp, so employees wouldn't use them for personal mail. It was sticking and not completing each stroke.

The monster was not much more than a big solenoid and frame holding the pins. The base was a heavy plate with the matching holes. There was no obvious allignment adjustment, so I decided to try lubrication.

The stuff was part teflon plus oil in a spray can. I spritzed and hit the switch. No better, so did it again several times. Then there was a large yellow flash and the whole 25 pound perforator jumped a couple of inches off the workbench. Shouda looked for the label on the can - it used propane as the solvent/propellant.

The perforator didn't work any differently, but the chad was knocked out all over the benchtop.

 

I was pretty distracted the other day, and managed to connect Vcc and Vee to an LM324 backwards. When the power was turned on, I received a loud "Bang!", and about a 1/2" long blue flame squirting from the side of the LM324. Well, plenty more where that one came from. LM324's are so antique that I might consider doing that to the rest of those I still have kicking around.

Fortunately, my mishaps have been rather few and far between. I've seen/heard some rather spectacular failures by co-workers that I have no wish to duplicate.

Sometime in the late summer of 1976, myself and fellow radar "tweets" had been working on an aircraft in the hangar, along with Hydraulics shop. Hydraulics managed to spill a good deal of hydraulic fluid on the hangar deck (floor), so we retrieved some ground clay (think cheap cat litter) to absorb the mess and sweep it up. Meanwhile, one of my fellow techs continued to work on the radar, which seemed to be having problems with the transmitter section. While I continued to clean up the mess, he went to get a high-voltage probe to test the transmitter power supply; 18.5kv, 2KW.

Unbeknownst to him, someone had previously dropped the HV probe, and it was cracked longitudinally. When the transmitter power supply voltage came up, he put the probe on it, and the HV used him as a conductor. There was a loud snap, and I turned around just in time to see him perform a backwards half-gainer head first into the trash can we'd been using to deposit the swept-up ground clay in. It happened so fast that he had no clue where he was or how he got there. Fortunately, he wasn't injured and there was no damage to the radar system. We had a good laugh at his expense

Then there was the story about a Gunny (Gunnery Sergeant) in an Ordinance shop who decided to build his own stray voltage checker. Before you connect live ordinance (particularly missiles) to aircraft weapons systems, one must check to ensure there are no stray voltages present, which might accidently cause the ordinance to fire. The standard piece of equipment for stray voltage checking was rather large and awkward to deal with, which is why the Gunny decided to make his own more compact version.

So one day he was out on the flight line, testing a missile umbilical for an AIM-9 Sidewinder heat-seeking missile while there was someone up in the cockpit re-setting circuit breakers. For some reason, the Gunny had the umbilical connected to both the wing rack and the missile at the same time, and as Murphy would have it, the missile "saw" a stray voltage and the rocket motor ignited, causing it to launch and streak along the flight line, straight through a hangar, at near-Mach - leaving a very stunned Gunny wondering what in the heck happened.

The doors on both ends of the hangar were open, and even though the hangar had aircraft in it with personnel working on them, the missile bounced right through without damaging anything. Fortunately, this was a "practice" Sidewinder; it had a live rocket motor, but no warhead. The incident reportedly took place sometime in the early 1970's.