Don't you hate forum posts where someone says they're going to do something and then they're never heard from again? Me too. So let's continue on with this! (Sorry for the delay)
I got my long sleeve jacket, my gloves, my safety glasses, a 5 gallon bucket, and proceeded to remove the old broken spring from the case while trying to avoid the explosive guillotine/machete/scythe of death, that these springs are rumored to be. And you know what, once the winding tension is off, these things aren't really under *that* much pressure. I simply started pulling from the center, using my hand to prevent any but one loop coming out at a time, and it just slowly slithered out into the bucket and was all floppy and tame. To be sure, it was trying to come out more rapidly, with multiple loops at once, which would have been a little more frenzied...but it took hardly any pressure to prevent that, and I don't think it would have required any more pressure than that to put it back. It's pretty amazing what a difference in force a few turns in an enclosed space can make.
The old spring was RUSTY! The rust had thinned out the metal noticeably in places, and it was already half cracked through in a second spot. That spring was bad. Interestingly the rust perfectly corresponded with the stuck on clumps of old graphite. Had the graphite sucked moisture out of the air and held it against the spring, rusting it? Or had the rusty parts of the spring somehow attracted all the graphite and allowed it to stick instead of sliding off? Not sure. I flattened it all out in the back yard (easier because it was thin and weak) and measured it. Right at 217 inches, which is about 18 feet, not the 20 mentioned earlier in this thread. I'm no mainspring theory expert, but I believe extra length, at least until you got so much that it was too tightly packed in the case, would allow for more run time. As long as there is a slack gap spot inside the case, then winding it simply moves that gap from the center to the outside as you go, and more length would allow more turns before you reached the outside. I think. So if you're making up a spring from raw spring steel stock, I would definitely go with the 20 feet (as long as it fits) instead of the original 18. The 1 inch width and .015" thickness specs were dead-on of course.
The shape of the keyholes was different between the original spring and the NOS replacement, as seen here:
Seems like the old style would help keep the spring from popping off of the nubs during sudden unwinding...and yet my old spring had done exactly that, as you'll recall. The new style definitely helped when installing the replacement though, so yay for that.
Once I had the empty case sitting in my hand, it became blindingly obvious how I was supposed to install it. Just look at this:
Yep. Just set it into the case like that and give it a quick push. Poof, into the case. It's wound tightly enough, and the circular retaining ring is smooth enough, that it moved as a unit down into the case rather than even trying to go "sproing!" Then it was a simple matter of hooking the spring to the nub on the center post... well, simply with some small stiff pliers. The spring is a little curved at that point so you need some pressure down in the center there to convince it to get on the nub. But pretty straightforward. Then getting it on the case nub is merely a matter of spinning the spring around in the proper direction until the hole was lined up, then spinning it a little more until it popped on the nub and then slid into the slot. Here is where the centered keyhole comes in handy, because it will simply do that all by itself if you spin the spring, rather then needing to guide it a bit high with pliers or something to get it lined up. The simplest way to turn the spring is with the winding key, but you need something to hold the case while you do that. Well, the motor will do that for you! But then with the lid off the spring case, winding it is a REALLY REALLY BAD IDEA. I mean, even the slightest bit of winding turns it into that potential whirling dervish of decapitation. So I did this:
Perfect! I wound it until the spring was at the nub (you can see there that it has nearly a whole revolution to go to get lined up) and while I thought I might need to get in there with some pliers, I didn't. It popped on the nub and slid down the slot and all was well.
You might have noticed what's missing here: the step about coating the raw spring steel with graphite before winding it into the case. Yeah, well, OK, but seeing as how the keeper was supposed to allow you to pop it into the case so easily, I just couldn't bring myself to unfurl it into the bucket and rub graphite on it and then wind it back into the case. So I did it this way, and simply worked graphite down in between the winds by repeatedly winding and running the motor, which moves the gap from the center to the outside and back again, allowing me to squish graphite down in there. You'll be happy to know that this still accomplished the part noted earlier in this thread about coating me, and everything I was wearing, and all of my tools and surroundings with graphite.
Seems to have worked. You have to be careful, but until the spring is wound up it simply isn't quite as dangerous as it's made out to be. If you're handy with tools and careful I wouldn't even call this difficult (just messy!) I would have no qualms about making a spring from raw stock at this point. Other that cutting the keyhole slots in it, everything is really straightforward.
Duncan