I feel like that first post might have been a bit of an unfocused jumble. I'm interested in whatever options I might have, since I really think Gamepark and GPH devices still hold up in terms of build quality.
But to give this thread a focus, I'm really interested in the feasibility of the "battery caddy" idea I mentioned earlier, since:
1. Most of the options I see involve modding the battery compartment or soldering contacts. I'd like something that gets the job done with minimum trauma to the device.
2. Mods like the Sakura "flat" sticks and some of those I've seen on these forums seem to indicate something within tolerances wouldn't be too expensive to produce.
I have the stock wiz battery and the BL-5C from my LDK in front of me, and here are the rough dimensions and specs for each. I can provide more detailed information and photos if people need them.
The stock GPH wiz battery is 3.7 volts and is designed with a capacity of 2000mAh. It has three contacts (positive, thermistor, and negative)
By my measurements the battery is 44 mm by 82mm by 7.4mm at its thickest points.
The length of the "cell" portion of the battery is about 75mm.
The contacts are on a "tab" that is the flush with the battery on one side. The tab is 4.1mm thick and extends about 7mm from the main battery.
The BL-5C has the same voltage and appears to have contacts in the same order. and the one that comes with the LDK is listed as having 1020mAh capacity.
The dimensions are 34mm by 5.5mm by 53mm.
So that seems like those dimensions would provide (more than) enough room for a sturdy housing, and the BL-5C would give you about 2.5 hours of play time on an overclocked wiz.
While that's far short of the ten hours that rebuilt stock batteries were rumored to provide, it seems like a lower-risk, more accessible, and much cheaper approach. The battery door on the wiz is quite secure even without the two screws that typically hold it in place, and it wouldn't be too much of a hassle to open up and swap batteries. It's at least more convenient than the GP2X F100 ever was.
And, of course, I'm only talking about the BL-5C in the first place because it seems to be cheap and in ready supply. I'm sure there are better candidate batteries that I'm not considering.
Speaking of things I'm not considering, do any reasons jump out why a kit like this wouldn't be feasible? The most challenging and expensive part I can think of would be pin-to-pin element.