Supercapacitors for decoders are pretty common in the DCC world, to improve reliability over turnouts, intermittent track, and so on. They typically power everything in the engine - motor, processor, lights, and sound. For example, I have an ESU Loksound and 5F cap installed in my Lionel docksider. It will run for four or five seconds after power interruption, and goes over everything on my layout without a hitch, including long curved 096 switches and my #8 double crossover.
Conversely, I think it's been shown that the supercaps on the PS3 boards only support the microprocessor and ancillary functions like lights and sound. My experience with my own PS3 engines certainly supports this - they come to an immediate, ungainly halt immediately after loss of power or ground. If you have the misfortune of having that happen while running a consist, it's particularly unpleasant. So....
Suppose one would like to add supercap storage to PS3 engines for motor power as well. On DCC decoders, this is accomplished by tying the supercap to the decoder power bus just downstream of the rectifier and filter/regulator. In this way, if the track power goes away, the decoder and all its functions, including the motor, are still powered. MTH takes a different approach, however, and does not power the motor with their supercaps. Here is a diagram showing my wild *** guess at how the PS3 system is laid out:
Continuing my speculation - Track power, be it analog, DCS, or DCC, gets rectified and filtered, and maybe regulated at a high level (red). This high voltage bus is controlled by the microprocessor via an H-bridge, FET, or whatever to drive the motor(s). A low voltage regulator provides the 5V for microprocessor, lights, speaker, etc. The existing supercaps (green) sit here, between the low voltage regulator and the processor. Signal input is taken from the track as well (orange), bypassing all the power circuitry, for the processor to act on.
A motor supercap (shown in blue) would need to sit on the high voltage bus, between the filter/regulator and the motor controller. The 'supercap' would actually consist of three components:
- Charge regulator: This could be a linear or switching regulator, or even as simple as a resistor and zener diode. Basically anything to limit the charge current and max capacitor voltage.
- Capacitor: The supercap itself would be a single 2.7V cap. These are available in some crazy capacitances for just a few bucks these days. For example, here is a 25F cap that is basically 1" long and 5/8" diameter, for under $4. Operating from 2.5V to 1.5V, that will provide, nominally, 1A for 25 seconds.
- Boost converter: A switching boost converter with its output diode protected and set slightly below that of the high voltage bus. Needs a low minimum input voltage to maximize the usable voltage range of the cap.
So now here a few of the potential gotchas that I can think of:
- The big one: Obviously, figuring out where to tie this in to the existing board is key. A schematic would help of course, but I doubt one is available. Without a schematic, it comes down to some measuring with a scope and some (more or less...) educated guesses, and there's no way to know it's going to work without actually trying it. Of course, there is some risk to the hardware during this process.
- It is possible that the sudden stop behavior is actually intentional, and that no signal is provided to the motor controller when there is no power at the track (maybe to accommodate analog users and the pushbutton reverse?). In this case, adding the supercap to the high voltage bus would be useless, and without any workaround that I can see.
I'm sure there are others.
Anyway, like I said, just some random thoughts. Comments, suggestions, or any PS3-specific tidbits welcome!