Joule Ringer!

[NickZ]

  Slider has even used a regular AC wall adapter as a Joule Ringer transformer, and has made them work, as they come. So, I thought that used E-cores can also be made to work,  as they come,  or possibly just winding the additional primary on the outside, as is sometimes done with axial inductors.  It may be better to remove the original windings, but not easier.
  Some E-cores are only inductors, and not transformers. I have several bigger E-cores also that are transformers.  More and more transformers are been made from the ferrite E-cores now.
   Testing the primary turn ratios can be done by leaving longer wire ends not wound, and adding or removing winds as needed, a couple at a time. That's where the "light box"  would come in handy.
  Slider, just uses a toilet paper tube (a paper towel tube would also work) as a light box, to put the smaller 1 to 2.5 watt light bulbs into, and a small solar cell (like the ones that come on calculators) on the other end of the tube, to measure the lights voltage given off by the different bulbs, or circuits being tested.  A regular volt meter would also work to see the output. A bigger version of that idea would also work, for the larger bulbs.
Not as scientific, but No costs are involved.   
In my case, just my eyes work well enough for me. If I can't tell the difference in bulb intensity, I don't really care.
                                                                             
  Nick

[conradelektro]

The transistor matters, use 2N3055

Finally I gave up the MJE13007 and used the 2N3055 (yes, I am stubborn). This was an important step, because the 2N3055 works without any additional components. A 100K resistor between the base of the transistor and the positive rail helps sometimes to start up oscillation.

I am still using my E-Core from FERROXCUBE ETD49/25/16-3C90 with 20 turn primary (2 parallel wires wound in 10 turns and then connected in series) and a 100 turn secondary, which results in a 2:5 step up ratio.

This set up (see the attached picture) produces reasonable results with a 5W and 2.5W LED lamp. My 9W LED lamp and various unmodified CFLs do not work.

When I use a 1:10 step up ratio with the 2N3055, all lamps I have work (LEDs, unmodified CFLs) but the power draw is very high, about double as should be (double as specified by the manufacturer of the lamp).

What I conclude from this:

1) Use the 2N3055 transistor (very important, no other components needed; in case you want to be super clever, use a 100K resistor between base of transistor and positive rail)

2) The windings on my E-Core are not right. I will try 20 turns primary and a tapped secondary with 120/160/200 turns.

I still have to get more LED lamps in order to try more than one of the same type in parallel.

Greetings, Conrad

[JouleSeeker]

Conrad -- agreed on the importance of the transistor, 2n3055 works well.  I did burn one out, accidental short on the output. 

Nick -- the cardboard box is cheap; the light meter recommended is only $13.30 on Amazon (prime) right now, and easy to use.  And repeatable.  And allows experimenters to calibrate easily and compare Lumen/Watt -- quantitatively.    Anyway, we may disagree on this point but I appreciate your input.

Here is a new vid by Peanutbutter, showing how straightforward it is to build a light-box.
http://www.youtube.com/watch?v=x3xuXKqMnjM&feature=em-uploademail

@all -- don't underestimate Lynxsteam's air-core, near-Tesla coil.  I like his approach!  I'm amazed frankly at how easy it is to start oscillating, and how well it works overall.

-Steve

[NickZ]

 There is no disagreement Steven, the light box is of course better. But, some of us are half way around the world from Amazon, and import duty, international shipping charges and delivery fees, as well as other issues are also involved. Otherwise the $13 cost is very reasonable.  If you see what I mean...

  What I am wondering is what happened with the Lasersabers 10 led bulbs lighting on 200 mA???   If resonance is not involved with that, was is. Some guys are actually drawing more amps than when using a regular AC 110v grid source. So, we really need to get this down, and the use of the "light box" may be very helpful with that.
 
  Lighting a small house or shop, from a 10 watt solar panel 24/7 like totoalas is doing now is also great news.  I'm on that one like a cat following a mouse.

  Stefan does have a point though, concerning the damaging effect of super bright daylight spectrum 5000 or higher rated Led bulbs. The warm whites are more user friendly, and safer. Also, as no mercury is involved, as with the CFLs. I had forgotten about that as CFLs are so cheap and obtainable everywhere now. But, CFLs don't seam to last very long in my climate though, nor do the cheaper Leds, as I have a pile of dead CFLs ready to be converted over and gutted for further use with the Exciter, or Ringer circuits, if they'll work for that also.
  In any case,  this is all very exciting stuff...
                                                                     Nick

[Lynxsteam]

I don't get good results using a 3w Phillips LED bulb.  It draws more off this circuit than its rated for.  Lidmotor speculates that there is something in the circuit of the dimmable LED bulbs that is working with this simple JR circuit.  I agree with that.  So don't stop at those low watt LEDs and assume your circuit isn't working.