I'm certainly no electrician by any stretch of the imagination. But I know enough about the work involved to make a lamp or two. And that's pretty much what I did. And it's a lot more fun than some might think.
At the moment, with the frosted globes, and the Christmas bulbs, they're more decorative than anything. But I had enough brains to use sockets rated for 75 watt bulbs, so that can be changed if desired. But I can't do much about the globes.
Since the black lantern was the prototype of the two, the work is rougher than I'd like. But there's not much I can do about that. By the time the red lantern came around, I knew what I was doing, and what I wanted to do, so it was a lot easier.
The frosted glass provides greater flexibility since the colored bulb can easily be swapped out for something different. Unfortunately lighting isn't properly uniform, meaning I either need something stronger than a 7 watt Christmas bulb, or I need to paint the globe a single color, put a full 60 watt bulb in, and see if it can still light it up to a desirable degree.
I find myself in need of some assistance on this matter. I can work with AC just fine, but DC power gets to be more complicated for me.
The basic idea is to take one of these cheap lanterns, and a cheap LED flashlight, and combine them to make an LED lantern with old-skool aesthetics, but without costing a lot of money, and without running on 2D batteries.
I have a basic idea for how to do the wiring, the cutting, the grinding, and the fitting that will be necessary. But what I absolutely don't know, and can't find, is information on what swapping out the three AAA batteries at 4.5 volts, in favor of a single 9 volt battery, might do to the diodes if they're exposed to double the electricity. Any help would be much appreciated, as I don't want to order in parts I don't need.
Disclaimer: it is many, many, many years since I learned this and I could be horribly confused. But I'm pretty sure I'm not. Check everything with a test harness and a multimeter first.
OK, so first you need to establish something: were the AAA batteries in parallel, or in serial? If in parallel, then you had 4.5 volts , and triple the life of one 4.5 battery (or more likely, cel). If in serial, you had about 13.5 volts going through there. If you want to know, digital multimeters are cheap and as long as you do it right, boringly reliable.
If it was 4.5V, your LED might fry under 9V (unlikely) or burn bright (likely) but perhaps not efficiently or maybe not even very long. There are a lot of ifs, ands and buts involved.
If you had 4.5V, then if you put two LEDs in parallel over your 9V battery each one should get (assuming equally resistant wiring) about 4.5V of electrical tension across it.
If it was 13.5V, then your 9V would get you a dim light. If you then put three 9V batteries in serial, and two LEDs in parallel over it, you should work out at 13.5V per LED.
But first test everything with a multimeter. Then test it again the other way around.
The principle behind serially connected batteries:
the batteries are daisychained, with the positive terminal of one connected to the negative of the next. You get (minus internal resistance and a few other factor of relatively minor importance) an addition of the voltage across them. Thus 5 * 9V batteries in serial gives you 45V of tension. If you have 45V running over a circuit offering 1 ohm of resistance, it should give you 45A of current flowing, because V/R=I.
The principle behind batteries connected in parallel:
the batteries are all linked with like terminals together. Between the positive side and the negative side each battery can only push at 9V, thus they collectively push at 9V but each one takes a lot less of the load, producing a much smaller amount of current per battery. This means that each battery lasts a lot longer before being exhausted.
The easy way to tell whether they're in serial or parallel, if the connection doesn't make it obvious, is a multimeter.