Self Running Micro TPU, with closed loop.

[Bruce_TPU]

HEY 
MAY YOU TRAED TO PUT  RESITOR IN PARALEL WHIT THE CAP  AND TUNE WHIT  MORE PRICISLY  TO GET MORE TIME TO LIGHT THIS LED 

Hey Macedonia CD,

I tried resistors in parallel and nothing worked until I put a Mega Ohm resistor in parallel, then the LED's started up.  I am timing it now.  After, I will try the same test with them in series and will post the results.

Cheers,
Bruce

EDIT:
13 minutes, 46 seconds.  No improvement, but much worse.  Good idea any way MCD!   

[Bruce_TPU]

@ EM

I think that I may have proven that the coils are indeed involved in the extended lighting of the LED's.
I had the two LED's in parallel, and then changed two things.  This increased my best time by just over 20%.  My best time had been 19 minutes, twenty seconds.  With the following two changes, it stopped working at 24minutes, 27 seconds.

Change 1:  I noticed that my transistors did not have identical numbers on them.  I dug through my loose pile of transistors and used two with the number 130 on them, in parallel, wired according to EM's schematic.

Change 2:  I took my second ferrite toroid, (a blank, no windings yet!) identical to the first, and simply laid it on top of the other.  I used a small piece of 20 gauge  magnet wire to wrap one turn around the two of them to hold them in place, then ends wired to nothing.

And my time has increased by just over 20%!   Now, I am thinking, if it is just the slow discharge of a cap, neither of those changes would have affected it at all.  So, I am thinking this is proof that the ferrite toroid is indeed active.  This is really getting interesting, now.

I think that next I will build an exact duplicate of my first coil.  Making sure wire turn direction, number of wraps, etc. are identical.  I will then place this on top of my first one and wire it in exact parallel.  The next experiment will be to wire it in parallel but next to the first one, and see the time differences.  Lastly I am building an exact duplicate circuit, that will be powered from a "different" battery.  Each coil will then feed back into the other and will have a tap to light a large bank of LED's.  Lastly I will add an AM radio coil and "tune" the circuit.  But one thing at a time.

Cheers, and good night!
Bruce

EDIT: I have now ruled out the additional ferrite blank above the wound one as to making any difference.  It seems, that matching the transistors exactly gave an additional 20% operating time.  I need EM or Duff to come up with a replacement type transistor.  Something better, so that two in parallel are not needed.  They seem to play a big role in "length of operating time".

[wattsup]

@btentzer

Great stuff indeed.

I just put my trigger coil on my new big ferrite and wanted to see how sensitive it was. Placing the coil flat so the trigger is a 6 o'clock, I put a compass at 9 o'clock at 1 inch distance. I could move the compass with only .1 vdc. I tried the same thing on my small coil set-up and the trigger got active on the ferrite at .6 vdc. So I am very happy to see this ferrite has alot of capacity (depth to hold and/or transfer a charge).

A toroidal transformer has to have an extremely efficient ferrite if it is too replace the bulky and heavy mass of a regular laminated core transformer while converting the same amount of power. I'm sure this will be an advantage.

[4Tesla]

Have you tried a 3904 NPN transistor?

4Tesla

[pese]

In any way.
this will not give mor power aut
that you ahrge in the capacatior before.

this is nothing es than an smal inverter
(even if you think , to becomes back-EMK)

Pese