"Egad, a solid state switching device is the least trusted."
EGAD (not to be confused with Steve Gadd), it looks like I wrote too little. I can fix that.
The miniature switching power supplies I'm referring to in new generation AC adapters ARE FULLY ISOLATED. They have a real live transformer between the input and the output, with galvanic isolation, high voltage insulation, all of the international safety certifications, etc. In fact, the transformers are harder to certify (more reliable) because of the higher frequency. The switch in "switching power supply" is in series with the input, after the input rectifier and filter. DC equal to the peak AC value (140V to 312V, depending on the country) is chopped at around 100KHz into a high frequency transformer primary. The secondary is rectified and filtered as in a normal wall wart. Isolated feedback to the primary control circuit (something not in a normal wall wart) stabilizes the output - a very good thing.
If the input switch either shorts or opens, there is no flux change in the primary, and all energy transfer stops. If the feedback loop opens, the controller shuts down. If the input voltage exceeds the design limit, the controller shuts down. If the input current exceeds the design limit, the controller shuts down. If the controller dies, there is no pulse drive to the switch, and everything shuts down. These things are way smarter than the old style.
"As for the 1 k output resistor offering protection or relying on skin resistance??? Not me. The feedback circuit can also fail producing a higher than normal voltage."
I was referring to a "normal" zapper circuit, which has no feedback. If you meant feedback in the AC adapter, see above.
In fact, if you think about the "stabilized wave" or any other non-555 zapper circuit or zapper output stage, the same applies. Unless you have one running on 18V or 24V or some other high DC value, and are relying on a feedback resistor in an output amplifier circuit to keep the output down to 9-to-12V peak, I don't see how any of the zapper circuits I've seen described can produce a higher than normal voltage. From what I've seen, the norm is to run the output pretty much at saturation up against the rail.
Of course, I'VE NOT SEEN YOUR SCHEMATICS, so I'm just spitballing here. Subtle, huh?
As for the 1K resistor, scope photos on several websites indicate that it is a significant current limiter/output voltage reducer.
"A failed regulator can output up to 100 volts DC in certain circumstances."
I don't see how. To squeeze out cost (a 1-cent price difference in a part equals $80,000 *per day* savings across the industry), the transformers are designed for a very narrow range of pulse widths and current amplitudes. The modulator will max out or the core will saturate (primary pulses are so wide that there is not enough time for the core to reset between them) before the output reaches 3 times the design limit.
"Use only a power adapter rated for use with medical equipment designed to attach to the human body."
The medical rating on power supplies does not address any of the issues you raised. There is nothing in UL544 or its successors which address circuit reliability, monitoring, fail-safes, etc. There used to be a difference in the insulation voltage rating (4KV medical vs. 2.2KV industrial, or something like that), but I think the non-medical regs have pretty much caught up. Some of our vendors have dropped the non-medical products from their lines, and sell only medical rated supplies to all markets, because it was more expensive to maintain two separate product lines than to "give away" the better performance where it wasn't required.
The only difference between medical and non-medical power supplies is the allowed leakage current, the amount of current which can flow through a sneak path from the power line to you to an earth ground connection, such as the metal rail on the side of a hospital bed.
"At least with a split bobbin transformer, you are isolated from the line."
True, but a bit misleading. All transformers (isolation types, not auto-transformers) isolate you from the power line. That's their job. Some designs trade off efficiency for cost, but there is nothing magical about a split-bobbin construction. A cheap split-bobbin is just as dangerous as a cheap bifilar-wound. Also, split-bobbin construction is *not required* for medical certification of a power supply.