Joule Thief

[TinselKoala]

You might have missed the picture of the two Fedex big rig trailers in the mess of a 100 car pileup they had in Texas just recently.   Maybe that's why TK's stuff and the other guys stuff is missing.   Fedex is probably scrambling to replace stuff or sort out the mess...

Oh, man, do you have a link for more information?  According to what they told me, one of the missing boxes was last scanned in Ohio and the other one in Fort Worth.... so if that crash was in the last week it's certainly possible. But of course they didn't mention any crash to me on the phone.

Oh... man..... I just found the link. Thanksgiving Day. That certainly puts it in the right time frame..... My delivered box came on the Monday after TG weekend.... But the location of the crash is weird. Eastbound lane of I-10 east of Houston, about 300 miles East of me, so the trucks were moving away, I think, and FedEx did tell me that one of my boxes had been scanned in Fort Worth, which is 350 miles north of me and does make sense as en-route.... so who knows.
I feel badly for the people who were killed. The photos really look like it was a huge mess.

[TinselKoala]

Meanwhile... back in the DeepBunker..... OK, I built the circuit and could not get it to oscillate at all with a TIP3055 or 2n3055 in the Q2 position, no matter what transformer setup I tried. I finally got the thing oscillating well with that little choke I showed earlier: I wound about 300 turns of #30 on it, potted it in epoxy, then wound about 10 turns primary on top of that. I tried a bunch of different transistors in both locations. My favorite setup right now is to use a germanium 2n1304 for Q1 and a 2n2222 for Q2. The 2n2222 gives me the highest voltage spikes (I tried the germanium, a 2n2219a, and 2n3053 and TIP31A, all oscillate, and 3055s don't work for me).

I see the green LED on the output glowing and I see some very interesting interplay between the AA battery and the 2 Farad reservoir cap. I have seen the voltage increase... but nothing like what gadgetmall has achieved. I just started using a NiMH battery to see if there's any better result for me.
The system is really sensitive.

[totoalas]

http://www.youtube.com/watch?v=Lwm0r0bk8dM
Another variant  ac dc current reducer

[scratchrobot]

Set up a camera and video it, explosion and all!

Yes that would be nice 

Nice job on your variant Gadget! If I get the time I will try to replicate.
Maybe you can put a load over your battery to dump the excess energy 

[NickZ]

   Using my best ferrite rod transformer from the 3.0 circuit, a TIP3055, and a 2n4017 transistors, and adding a 1n4148 diode from the secondary to the primary coils, and another 1n4148 diode with a cap, the other way around, I ran the circuit all night long. I tried different caps, they all made the light get brighter than just with the diode alone.
This morning the 1.5v D cell alkaline battery was reading 1/2 v, down from the one volt reading it was at when it started. This was a dead battery basically, as I wanted to see if it would self charge, but it did NOT.
  This circuit is strong enough to blow the white led, if left on without the diodes, and cap. Even on a dead 1 volt battery. I also added in series with the led, a 10 led 4volt led bulb which was barely lighting, although the white led was bright. Maybe this is what caused the circuit to not be able self charge. Once the circuit is turned off, the voltage returned to 1 volt.
  Anyway, so far I see no real self charging, yet.  But, I'll make some more changes, and remove the bigger 10 led bulb from the system.  This is still a very interesting circuit, and this was my first try.
  More to come...
                                                       
   NickZ