the output being hf was rectified by a bridge and filtered by a low esr cap.
current limiting was a .47 resistor.
it worked fine till a few cycles of a window ac and fridge where nasty spikes appeared on the output.one led failed short and the other 3 overheated and melted the thermal epoxy holding them to the heatsink.in excess of 400f!
i measured 80v spikes on the output of this thing!
a bulb would survive but not an led.
have not tested any recently since i dont want the rfi here or at my neighbors.they get tossed on sight.
had a string of those puck lights given to me a while back that the electronic transformer had been causing tv and fm interference.
am was totally unusable.i went there to check the tv as they thought it needed a better antenna.it was fine till their kids turned on the lights on the reading table.
remove lights and tv works fine on all channels.only reason these didnt go to the trash is they are nice aluminum and glass ones.easy to led mod.

If that lamp was well isolated around the 12v output, they migh have used something nasty like a choke or electronic "dimmer" to make the "12V"

You mean in place of the "electronic transformer"? I don't think that would be the case at all...

Thats what i mean. Why not ? (other than very spectacular lamp EOL in case of hte choke)

Thats what i mean. Why not ? (other than very spectacular lamp EOL in case of hte choke)

Because the 12V lamp would flash over even without the EOL when subjested to mains voltage, even short time, pulses (the triac dimmer based "transformer")...
And even the series choke ballast won't provide the required insulation...

The problem lies somewhere else, but I do not want to bias "toomanybulbs" before he really describe, what was there (some of the small details count) - because my idea could well be completely wrong...