electrical deaths

We would say that the wires were crossed or swapped.

A GFCI breaker senses the difference in currents carried by hot and neutral wires. If no leakage, the difference is zero, so the difference is the leakage. If this leakage exceeds a few milliamps for maybe a second or two, pop.

The two GFCI breakers each see a very large unbalance, the full current in that straddling circuit.

Thank god that he didn't.

We would say pull the cables.

Embedding a tube versus the wires is the best future-proofing one can get. As is making the tube too big at first.

I have a non-electrical war story, also from 1970s Washington, different apartment: One Sunday evening I made myself dinner, and while cleaning up, the kitchen sink faucet, classic single-handle design of some kind, and the hot would not shut off, instead roaring at full volume. Far too loud to ignore and call the landlord in the morning, so I took it apart. Whereupon I found a 6" (150mm) long straight piece of steel coat hangar wire. Removed it and put the valve back together, which now worked correctly. But I could not imagine how such a long wire could get there, as it could not turn the corners in the pipes leading to the valve. So I took the other valve apart. And found another 6" length of steel wire. Removed it. This was carefully crafted sabotage, intended to generate repeat emergency repairs. My then landlord was himself a thief, so I never told him - they deserved one another.

Joe Gwinn

Ah.

Oh, the GFCI breakers triggered every day at 6:30, just when the station transformer went up a notch.

That house has had spurious triggers for years. Everything fine for months, then suddenly one trigger. In the end, I changed the GFCI with one that self arms itself 30 seconds later. Three times, after that it stays off.

It remains to be found if he did. There could still be spliced wires hidden inside the tubes somewhere.

Anyway, the place will be owned by somebody else soon, so no longer my problem.

Ah.

Absolutely. Make the tubes big.

Embedding the cables directly in the mortar has been forbidden for several decades, but the house is older than that. In fact, I had to redo the cables in some areas, with surface conduits. I intend to move house in a few years time, the new owner will probably tear down the walls and redo all that.

Gosh.

In Spain the entire house has to be protected under one such device. It is in the code. Even houses built before the code had to be retrofitted.

My neighbour has a new led fixture above the street door, flood type. When it is off, it still has some light. I guess the switch is on the neutral, and there is a small leak.

Here there is one mandatory inspection by a licensed electrician before the electrical company connects the house, then done. You do the barbarics when he is gone. When you sell the house, there may be a new inspection, at the electrical company request. I don't know the rules. On other countries, there is a periodical inspection.

We have periodical inspections for gas (kitchen and heating); but not portable stoves (with bottled butane or propane).

If it's mains stuff the arcing can still spark a fire. I've had numerous cases where I found signal or control wires were tinned and then clamped, and the connections failed or became erratic. Usually after many years.

If the live and neutral run directly to the fitting and there is a long branch of cable running to the switch, it could be caused by capacitive coupling between the wires in the switch branch.

In modern UK practice (probably harmonised across Europe) there has to be an earth conductor in all fixed wiring. This is usually located between the two power conductors and acts as a screen against capacitive coupling - but very old wiring doesn't have it and, even if it is present, it may not have actually been connected to earth.

Multi-core wiring, which would be used with a two-switch system, only has the earth between two of the three conductors, so there is still capacitive coupling between the other two.

This lamp is a month old. I can't say with certainty that there was no lamp before, so the cable might be old, but I think not.

I don't know about mandatory earth.

Capacitive coupling, could be.

I managed to get stuck on UK 240 Volts back in the eighties. Picked up a live mains piece of equipment forgetting I'd removed the access base. It locked my arms up to my chest and I had to be switched off. Fecking scary feeling when you realise you can't let go. Still have the scars forty years later where the live solder pins tuned my skin to liquid!

I'm pleased that you are still here to tell us about that.

I got across the secondary of an old TV power transformer when I was a teenager, something like 800 RMS. I shook for a couple of days.

I once rode a big LASH ship from San Francisco to San Pedro to find one loose screw on a terminal strip. It was the tach feedback on a

I had designed the control system about a decade before that trip, when I was still a Tulane student.

I used a nonlinear function function generator from the throttle lever to the main steam valve, with a limited range of PID RPM feedback. That limited range prevented what could have been serious if the tach failed.

The trip was fun, tugboat rides both ends, my own cabin, good food, treated by the crew as an honored guest, performing a miracle with a small screwdriver.

I guess the best answer is yet another war story:

For short-term lab stuff, I don't worry about such details, but with power it's far more important to have mechanically solid connections - no test probe and alligator-clip jumpers (like I used in my youth) here.

When in 2019 I was figuring out the wiring of a friend's 3-phase Dahlander two-speed 240 Vac (60 Hz) 5HP motor (that powered a Clausing Colchester Triumph 13" lathe he was restoring), it had to be all 4mm banana plugs and jacks.

My friend's shop had been a commercial shop, and so had very high current available - not quite arc flash territory, but can be dramatic.

The problem was that when the lathe was received the motor's wire markers were missing, and so had to be deduced from measurements. The motor had six leads (only the frame is grounded) and one connects three power wires, with or without the three shorting jumpers. One configuration is full speed, another is half speed, al all others are melt-down with drama and smoke.

Measuring winding DC resistances with probes did not work - contact resistances too large and unstable. Likewise capacitances and winding inductances.

For this effort, I made a six-bar terminal strip, brass bars on a wood base. Each bar had a screw-terminal (for the six motor wires), and two 4mm diameter reamed holes (for the banana plugs, power and jumper). The bars are 0.25" wide by 0.75" high by 2.0" long, made of alloy C360.

Long saga, but succeeded, and the lathe has been in regular use since then.

For operational stuff, the solder tinning is only for low-voltage (<

.

Joe Gwinn

Feed a constant current through the winding and then measure the volt-drop with a pair of probes on the wires themselves, so that the volt-drop of the terminal contact area is not included in the reading.

That is a standard four-terminal measuring technique used for low value resistance.

I do have a 6.5-digit bench top digital multimeter.

The problem was that I didn't have four hands, and clip leads were too flaky as well. All such problems were solved using the 6-bar terminal strip.

Joe Gwinn

How about using AC from a signal generator?

I do have the equipment to that, but it wouldn't make any difference. The contact points are tiny (very low capacitance and inductance), but with a stuttering series contact resistance.

This was the first time I had encountered such a black-box problem to solve. If I had to do it again, I'd do it differently. One thing would be to use a small transformer to excite the windings with 24 Vac

Joe Gwinn