Some time ago, I built the joule thief described in the attached link and pictured below. Since then, I have acquired an o'scope, and today decided to get the joule thief out and look at it with the scope. I used a new AA battery. Here's what I think I discovered.

The LED was being pulsed every 88 μS (which I believe is 11.363 kHz) at 5.6 Vpp. I am new to using the scope and wonder if these observations seem reasonable.

http://www.evilmadscientist.com/2007/weekend-projects-with-bre-pettis-make-a-joule-thief/

 

Yes. The basic principle is that an inductor that suddenly loses its current path (through the transistor) has to get rid of its energy (current) somewhere. The voltage will increase to the breakover point of the LED. You might find it interesting to change the color of LED so you can see that each color requires a different voltage to get it to allow the current through.

 

The numbers sound reasonable.

 

Some time ago, I built the joule thief described in the attached link and pictured below. Since then, I have acquired an o'scope, and today decided to get the joule thief out and look at it with the scope. I used a new AA battery. Here's what I think I discovered.

The LED was being pulsed every 88 μS (which I believe is 11.363 kHz) at 5.6 Vpp. I am new to using the scope and wonder if these observations seem reasonable.

http://www.evilmadscientist.com/2007/weekend-projects-with-bre-pettis-make-a-joule-thief/

Your photo looks like it's got a good toroid core with low resistance wire which gives you low resistance and low loss. 11 kHz is a low frequency for a JT; typical cores give frequencies around 50 to 100 kHz. All this means is that the core has very high permeability. All you need is a decent transistor such as the 2N4401, PN2222A or BC337. If you use a 2N3904, you can get about the same performance as the others by connecting three or four in parallel.

If you want to halve the current but keep the same brightness, try my 'Supercharged Joule Thief' circuit.