Joule Lamp

[NickZ]

  Lynx:
  Before I forget.  After I saw the toroid that you just made, I got to thinking... what would that core do if it was all packed tight with ferrite powder???
  You can take the ferrite cores, like from the back of old tv tubes, or monitors, and bang them to a powder with a hammer, sift it, and fill your tube with it.  What do you think?
Big ferrite toroid. I'll bet that it would work better than the hollow air core.
  There seam to be no getting away from the bigger is better, when it comes to these cores.
  Any ways, just thought that I'd let you know about that thought that came to me... before I forget it.
 
 

[JouleSeeker]

Thanks to Lynxsteam once again for a great idea --

  Nice progress this morning -- but I had trouble posting here this morning;
see vid -- here's the text:

Using the Lynxsteam build of the Lasersaber 2.0 SJR again...  Lynx asked me to tap the primary at winding 75 (instead of going to the end) AND to lower the voltage to 12.8 V, so that we would be able to run this with a standard 12V battery.  I did so -- congratulations Lynxsteam -- this now surpasses 100 Lumens per watt!

The data recorded on this vid show:  385 mA @12.8V = 4.9W (input).  The output is 493 x 10 = 4930 Lux.
The calibration factor (to lumens) is 0.103, so we have 4930 X 0.103 = 508 Lumens.
Thus, 508 Lm/4.9W = 104 Lm/W -- excellent progress!

Note:  at 12.0 V, I found 102 Lm/W.   Note 2: at 12.8V, without the ferrite rod, I found 75 Lumens per watt -- 104 Lm/W with the ferrite rod. It makes a big difference!

vid:  104LmPerWatt.AVI - YouTube

Thanks for further ideas, Lynx and Nick.  Lots to try out as we press forward.

[Lynxsteam]

Thank you Joule Seeker for your continued efforts on this device.  I think we have shown now that it is possible to make this with simple found objects as well as optimizing the design with precision.  Still a person needs the transistor, specialty wire, and the right shaped ferrite.  What we have achieved here is a blocking oscillator DC converter that doesn't ring, doesn't saturate at high frequency, and is fairly efficient with a single or multiple bulbs.  Seems pretty flexible to use with varying voltages and bulb types.

A couple ideas, I don't know how plausible.  There is something called a magnetic reactance transformer that doesn't use a transistor to oscillate, but I don't know if it applies.  Common black ceramic hardware magnets can be heated in an oven to 850 F slowly to demagnetize.  And just about any insulated wire will work.

A couple of comments.  Filling a torroid wound aircore with ferrite would be tedious and would be better to just use a manufactured torroid.  Stacking torroids on an aircore probably is mixing two very different mechanisms.  Torroids contain the field tightly inside.  Aircores need to have the field intersect at 90 degrees to the wire.

I have been picturing High Frequency movement of electrons.  An analogy would be to picture two tubes with tin foil covering the ends.  One has four small ball bearings inside, the other has 100 of the same ball bearing.  Shake both back and forth simulating AC.  The one with just four ball bearings shake back and forth hard and fast simulating high voltage and high frequency.  The one with 100 just roll back and forth slowly simulating 60 Hz and lower voltage.  The tin foil will be broken through in a short time by both methods representing work being done.  If electrons have mass, then mass times velocity would equal force.   Work is force times distance/time. 

In the case of these light bulbs we are using very little current at high voltage and high frequency.  The bulbs will run much cooler than if they are run off grid power, and that translates to energy savings and higher efficiency.

[NickZ]

  Yes, it may be a bit tedious to fill the hose or pvc pipe with ferrite powder, but it would work.  It would only take a few minutes to do.  As there may be no way of having a 14 inch ferrite rod 1/ 1/2 inch wide to fit into the PVC pipe, or hose.  Nor are there any 8 or 10 inch torroids that I know of.  The metglass ones cost $660.
 
  I guess that I'm used to crushing things like quarts, carbon, and other things for my power cells.  The ferrite can be obtained from any tv repair place, probably even for free.
When I'm ready to make the Mango version, I'll fill it up with ferrite, in the way that I mentioned.  Yes, it would be easier to buy the ferrite, but, can you?  I'll check on that too.
Anyway, it's just an idea.

[NickZ]

  Congrats on the 104 lm/w guys.  That's just great!