I highly recommend the GML throttles.
If you want power, look at the below links. It sounds like you understand the potential dangers of working with mains power. You will need fuses or circuit breakers also. These supplies can supply an amazing amount of power. I hesitate to suggest them because some clueless try to mess with them.
http://2railoscale.blogspot.com/2011/09/how-to-convert-used-pc-power-supplies.html
http://www.ppdnmra.com/MRRPWR.pdf
You can Google computer power supply model railroading
For a bench supply, Google computer power supply bench.
There has been some discussion in this forum in the past. Search it.
Rich
Having come through the amateur radio many years ago, Working with voltage isn’t my problem. My problem is justifying have twenty to thirty times the necessary available current. I normally worked with something in the 600V range, and about a quarter amp.
However, a computer supply furnishes far more current than I’d ever need, and poses hazards that don’t need to be. If I was trying to furnish power to a large club layout, okay, but I’m not. I’m powering a small layout with at most three loco’s at any given time. I’m looking at having 4 amps available, which is probably more than twice my need, a reasonable margin.
And unless you’re going to run your trains at full speed all the time, there is additional circuitry necessary. Wirewound pots aren’t cheap, and cheap pots will soon give up to the currents demanded. A simple one transistor circuit can increase the current that even a cheap pot can control, as long as the pot doesn’t have to carry that current. Any semiconductor device has a fixed forward voltage drop, and as an average, I figure one and a quarter volts, or with a fixed twelve volt input, the voltage delivered to the track will be below eleven volts. Twelve volts regulated in, either you’ll have to change the voltage, and God knows how they set that in the computer power supply, and He ain’t saying, or put up with the lower voltage. This is assuming a straight DC setup, not DCC, although I’d expect the same conditions to exist, a forward voltage drop in the circuitry preventing getting the full voltage to the motor.
There is no perfect answer, only what works for the individual.
Rich