Self Running Micro TPU, with closed loop.

[sanmankl]

@all,

Just got back this morning from a long flight and right now, I'm back tinkering this micro-tpu.

This evening, I'd noticed something odd.

I had my "all-copper" tpu running with just 1 blue LED and at around .8V, the blinking stops and when I put my fingers to the LED legs, it starts blinking again right down to .4V...I haven't tried a separate grounding for the circuit yet....but I think I will....Don't know what's going on but certainly the human body aids the LED blinking....

Secondly, I rewound my iron wire (recycled from previous "iron coil" to another torroid. It's my biggest I have. OD is 72mm and I "suspect" it's a "pure" iron one. It's definitely not a composite. Coil turns are as follows: Collector: 22 turns (iron wire) , Recapture: 14 turns (iron wire) and Trigger at 12 turns. All turns comes to fill up about 1/3 of the torroid and my run time is 26 minutes. Did the test twice. Verified correct. Modified EM's circuit in two areas. My LED + goes to coil and the negative goes to GND. Changed cap to 220nF instead of 1uF. Run time longer with 220nF. Played around with 470nF. Result is almost the same with 220nF.

EDIT: Winding for the above is straight (no overlay) Collector, Recapture and Trigger. Equally spaced between these three coils.

I've been examining my stash of torroids and after going through the torroids with many coillings; I can only deduce that the composite ones performs poorly when compared to the iron ones.

By the way, I picked up 2 rolls of iron wire at my local tesco. It's called "garden wire" Comes in 60 feet roll. It's not the same quality as my white-coated  iron wire but I'm not complaining paying about US$0.80 per roll.

Managed to kill another 2 Blue LEDs and a 2N2904 in the process.....

Next stop: wound the big torroid and test.

Cheers, cp

[wattsup]

@guys

When you mention cap values, please include the voltage rating also since this can change dramatically the device performance.

Also, I think each builder should chose one of their former posts and keep a build spec updated there and include the location of the build in your slogan section like @Acerzw has under each of his posts. This way we, you can still add new posts of changes, etc.,  but have a link to your standard build somewhere on the thread.

Just a suggestion.

I have a lot of radio ferrite antennas (rod and flat types) that already have one two or three coils wound on them as standard design. Tried a first one having three coils and the led blinked right away, but only lasted 12 seconds with a 4 volts charge. I'm only charging to 4 volts cause I'm tired of blowing my leds and putting a resistor before the led just eats more juice.

[tak22]

here's a new run time target for you .... 1000 years 

from Everyday Practical Electronics, December 2006

"this circuit will brightly flash an ultrabright LED at 0.5Hz for more than twenty years ââ,¬â€œ drawing 12μA off six high capacity AA batteries. If the component values are used, it will flash, although more dimly, for close
to one thousand years ââ,¬â€œ drawing just 0.3μA"

just a little humour, or maybe something to learn from ...

tak

[HopeForHumanity]

So does the length of the wire around the torroid have any relation to the frequency of the led flashing?

[Bruce_TPU]

@btenzer

If you reduce your cap it will be interesting to see if the run-time drops in exact proportion or differs...

Have you tried placing your cap in your coil?

Acerzw

@ Acer

I haven't tried the cap inside my toroid yet.  I will get to that experiment soon though. 

And I too was wondering if the run time would drop in direct proportion to the uf reduction of the main cap.  I will stop and Radio Shack tomorrow and buy an electrolytic 450uf or 500uf depending on what's available.

@all,

Just got back this morning from a long flight and right now, I'm back tinkering this micro-tpu.

This evening, I'd noticed something odd.

I had my "all-copper" tpu running with just 1 blue LED and at around .8V, the blinking stops and when I put my fingers to the LED legs, it starts blinking again right down to .4V...I haven't tried a separate grounding for the circuit yet....but I think I will....Don't know what's going on but certainly the human body aids the LED blinking....

Secondly, I rewound my iron wire (recycled from previous "iron coil" to another torroid. It's my biggest I have. OD is 72mm and I "suspect" it's a "pure" iron one. It's definitely not a composite. Coil turns are as follows: Collector: 22 turns (iron wire) , Recapture: 14 turns (iron wire) and Trigger at 12 turns. All turns comes to fill up about 1/3 of the torroid and my run time is 26 minutes. Did the test twice. Verified correct. Modified EM's circuit in two areas. My LED + goes to coil and the negative goes to GND. Changed cap to 220nF instead of 1uF. Run time longer with 220nF. Played around with 470nF. Result is almost the same with 220nF.

EDIT: Winding for the above is straight (no overlay) Collector, Recapture and Trigger. Equally spaced between these three coils.

I've been examining my stash of torroids and after going through the torroids with many coillings; I can only deduce that the composite ones performs poorly when compared to the iron ones.

By the way, I picked up 2 rolls of iron wire at my local tesco. It's called "garden wire" Comes in 60 feet roll. It's not the same quality as my white-coated  iron wire but I'm not complaining paying about US$0.80 per roll.

Managed to kill another 2 Blue LEDs and a 2N2904 in the process.....

Next stop: wound the big torroid and test.

@ sanmankl
If you could post a picture of your newly wound coil, that would be appreciated.  I must try to locate pure iron insulated wire.  I suspect the wire I used the other day was an alloy.  It was made in china, and does not say.  It was probably lead wire!  LOL   

You may want to try higher voltage rated LED's and Transistor if you keep frying them.  That may turn out to be a good thing!

Touching the LED and it comes on, well, that is just strange.  I will try that tonight when mine dies.

Welcome back to the hunt.   

@ All
This afternoon I spent a great deal of time carefully reducing all of my circuit wires, by at least half, and sometimes eliminating them altogether.  I am now testing to see if this has increased my time.  I will let you know if it does.

Holiday Cheers,
Bruce