Self Running Micro TPU, with closed loop.

[Bruce_TPU]

@ Samankl
A very interesting post.  If you can tweak your frequency up to about 7.8 Hz for me and try that, also 7.3, with the same set up and see if the times are the same.  It does indeed like a load.  Are your LED's in series, parallel or both?

Also, looking at EM's original circuit, the bottom green wire going to the transistor.  Do you have that going to your collector or emitter?  Mine goes to my emitter.  It is the only way it will work for my circuit.

Holiday Cheers,
Bruce

[sanmankl]

@ Samankl
A very interesting post.  If you can tweak your frequency up to about 7.8 Hz for me and try that, also 7.3, with the same set up and see if the times are the same.  It does indeed like a load.  Are your LED's in series, parallel or both?

Also, looking at EM's original circuit, the bottom green wire going to the transistor.  Do you have that going to your collector or emitter?  Mine goes to my emitter.  It is the only way it will work for my circuit.

Holiday Cheers,
Bruce

@Bruce,

I have tweaked my frequency to 7.8 and then to 7.5 with my copper wire coil. 7.5 seemed to work better. Difference of about 3 minutes. LEDs are in parallel.
My LED positive goes to the 15T wire and the other end goes to gnd. The bottom green wire that goes to my Transistor collector. If yours goes to the emitter, then what have you hook up to your collector?

What transistor are you using? I'm using BC548 and the pin-outs are CBE (looking at the transistor, pin pointing downwards with the marking facing you). The 2N3904 are EBC (same orientation).

Funny thing is I 'can't' tune my iron wire coil to the frequency. Tried just now. With even a 4M7 pot (set to the highest), I'm reading 27Hz with my DVM...... I don't know why. The LED look's to be blinking even slower than I have it with my copper wire coil. Strange.

sanmankl

[Bruce_TPU]

@sanmankl

To answer a couple of your questions.  I am now using NPN Silicon MPS3904 transistor, after having tried out about 6 different ones.  It gives me 29 minutes, 30 seconds, with eight LED's.

The bottom green wire of EM's circuit goes to my emitter.  My collector goes to the negative side of 1000 uf cap.  It only works this way on my circuit. 

     / |  Emitter           1
    |  |  Base              2
     \ |  Collector        3

Right now I am running several experiments with a second coil in series.  I will let you know if anything interesting happens.  I also want to try ALL of my LED's in parallel.

Holiday Cheers,
Bruce

EDIT:
An interesting side note, is that if I touch ground of my kick start to the positive of the battery and the positive of my kick start to the ground, the LED's light up so bright as to hurt my eyes.  But then releasing the wires from the battery, they will not blink.

Also, on my board, hooking them up, ground to the ground and positive to the negative 12volt 100ma supply it will also light very brightly and then when the wires are removed, it will not blink.

Any ideas why from you more experienced electronic guys? 

[acerzw]

<removed>

[4Tesla]

@btentzer
It is the same circuit.. just using different values. replace the 1k resistor with a 62k resistor and replace the 100k pot with a 10k pot.  This is just an idea to be able to have more control over the frequency.  The original circuit that I posted should work fine.  Also as you can see by what sanmankl used, you can use a different RC values.

@sanmankl
Cool!  Glad you're trying the circuit in the 7hz range.  Did the circuit run longer than when running at 2hz?  Have you tried other frequencies?.. what happens at 20hz?

@acerzw
The cap in the coil sounds like something worth trying!

4Tesla