[Vivtek]

When Hurricane Maria hit us in 2017, we lost power for two weeks (we were extremely lucky in that - we were among the first 4% of the island to regain power due to our proximity to a hospital).

During those two weeks, my entire lifestyle changed, obviously, so it's difficult to know what changed because of that, and what was just the fact that e.g. I went from sedentary tech work to nailing salvaged roof scrap on top of the house on our farm.

But - my blood pressure dropped. I lost weight. I did a lot of reading of books. My notes are more interesting for those two weeks because they were ink on paper and I could insert drawings and diagrams wherever I wanted. They were more thoughtful and less reactive.

I fixed a mosquito zapper by sheer force of will (the capacitor was shorting across a resistor - poor design, but once I figured it out I could fix it. I'm still using that zapper today, and only shorted the capacitor out across my finger three times. Ow.) That day, I was become death, destroyer of mosquitoes.

I guess I quasi-fixed the roof on the farm house by sheer force of will, too. Had to dig the nails out of the scrap - no hardware stores open. My son considers it a formative experience.

So I didn't choose to go cold turkey, and it was only a short time, and its lifechanging nature is impossible to tease away from concurrent events, but still - anecdotally, it's probably worthwhile to try it voluntarily.

[throwanem]

FYI, that resistor was probably there to bleed charge off the cap when power was removed from the circuit, to keep it from biting. Rather than bad design, it's a standard feature; as you've discovered, big caps without it can be mean, not only to careless fingers, but to the circuit they're in, besides. Granted, it sounds like the resistor had failed short as you found it, and without access to a replacement, cutting it out to get the bug zapper working again was a solid play. But now might be a good time to replace it.

Chesterton's fence is a useful principle for reverse engineering, and this is one example. Another is the snubber diode you find across switched inductive loads; the naïve assumption is that a reverse-biased diode can't possibly be doing anything there, and ideally you don't have to learn from expensive experience that, without it, the switch contacts will at best be eroded by arcing from the stored energy in the load finding a path to ground, and at worst that inductive kick will spot-weld the contactor and send the motor running away until switched out of circuit with a hammer.

[dralley]

> But now might be a good time to replace it.

But be careful about that capacitor, because it might be holding a charge that can't go anywhere but your fingers.

[throwanem]

Ideally, you drain the capacitor first with the same resistor you're about to install as a bleeder.

Much less ideally, you can short a cap with a screwdriver or something, but the degree to which that's a bad idea scales at least linearly with the value of the cap.

I did that once with a photoflash cap, while trying to troubleshoot a failed Nikon SB-R200 ring-mount flash head - expecting the failure to be in the control circuitry, I wasn't sure whether it had died with charge on the cap. It had! Luckily I had the good sense to point the thing away from me, because it melted a chunk out of the screwdriver tip and distributed it as slag across my worktop. Even as it was, my ears took most of an hour to stop ringing.

A fair question at this point is: with a capacitor big enough to be that dangerous, why wasn't there the kind of bleeder resistor we're talking about? In this case, it's also a design feature, because flashes run on batteries and you don't want to waste charge, or have to wait all over again for the battery to charge the cap every time you switch on the unit. Too, these flash heads have no externally accessible contacts through which the cap might discharge into the user, and the charge circuit uses a MOSFET to switch battery power to the cap, so even if you go poking fingers into the battery compartment, it still won't light you up.

Nonetheless, it serves as a good example of why you want to be very careful with high-power capacitors. The one I'm talking about is only about the size of the second joint of your thumb, small enough to fit into a flash head that itself fits into an adult's palm. Even so, at full charge it had enough juice to blow up a screwdriver and injure my hearing - and if I'd been even more careless and discharged it through my actual hand, I don't doubt I'd have ended up with a permanent scar.

Be smarter than I was! Discharge your big caps through a high-value resistor before you do anything else with them.

[rcfox]

I inherited a stereo amp from the 70s, and I'm pretty sure one of the capacitors has gone bad. I've managed to find a schematic and PCB layout for it (remember when service manuals were a thing?) but I've been too worried about discharging capacitors to poke at it.

For now, I just deal with a mains hum that's only audible when nothing is playing.

[throwanem]

Amp power caps aren't usually bitey enough to be dangerous, although they will certainly hurt if you're careless with them. They also usually have bleeders, since a line-powered unit doesn't need to economize on power the way something like one of my flash heads does. I usually just poke a screwdriver across them when I'm doing a recap job, and I've never had them so much as spark. But for maximum caution, I'd use a 100K or so 1/2W resistor, connected with clip leads across the capacitor leads, and left for a minute or two prior to desoldering.

For an amplifier of that vintage, I'd probably be more worried about the fact that it'll likely be neither grounded nor double-insulated, meaning it's possible for an internal isolation failure to present a potentially lethal line potential on any metal parts of the case.

[Vivtek]

Yes, that lesson took very firmly.

[Vivtek]

If the spark literally jumps across between the leads of the resistor before the capacitor charges fully, it is bad design, and not a standard feature.

Once I insulated around it so it could do what it was supposed to do, the zapper worked perfectly, and still works perfectly.

[Vivtek]

It was pretty cool in its broken state, too, with a rhythmic spark occurring inside the handle (which houses the electronics). But I wanted to kill mosquitoes. Our house in Ponce was built for air conditioning - without power it was miserable, as the air circulation through the small windows was effectively nil. We were forced to keep all the doors open to survive.

That first day I killed over forty mosquitoes. Under the dining room table it was just bzzt-zzt-zzt-ztt. Seriously one of the more satisfying episodes of my life.