Battery Free LED Flashlight Driver - SJR Looper V4

[TinselKoala]

@slider: I'm a little confused, because your first circuit doesn't look like it complies with the LS V.4.0 schematic.  The schematic shows the Base of the transistor connected to the coil AND to the cathode end of the LED string, and the other side of the coil is connected to the anode of the LED string AND the anode of the 1n34a diode.  Whereas in your circuit it looks like you have just connected the Base to the coil, then the other side of the coil to the Cathode end of the LED string.

So ... Huh??

[TinselKoala]

Fooey! I just blew all three of my blue LEDs trying to wire my board like yours!

[Slider2732]

Eek!
Blown LEDs are only good on sub volt circuits !

Have just checked and you're right. The reason being that I recall it didn't want to 'spark up' with different diodes. I changed to the germanium and it ran, but, along the way will have omitted the Base wire. A big sorry on that one !
If it's any consolation, i've ruined a couple of diodes from the copper rail heat..that stuff gets hot quick and stays hot.

Another thing that threw me out, was scope work. Check out the attached lol. I wanted to see what was coming into the Base, so attached to the negative rail and Base leg of the transistor. It came up as 179kHz, fair enough, but the trace was ridiculous. On 'Auto' it had set the channel to 1.00mV and was showing the oddest of screen image.

[Slider2732]

One thing being looked at today, is the size of the core and whether it impacts the run frequency.
Looks like it does.
I've made another circuit, to the correct schematic, to demonstrate the flickering now shown by a large pot core.
Light output is also very impressive when body capacitance is added. Rather than adjust the size of a variable capacitor, the thought would be that the grip strength on a flashlight body would lower and raise the output brightness.

https://www.youtube.com/watch?v=SzAO2xy876U


Update
Changing over to a mini coil of finger nail size and 40AWG, the output becomes very unstable. It flashes, dips in stages, brightens on a whim and all sorts. A very handy size, but very random and finicky in output. Adding a plastic top to the var cap would help, but it sometimes goes from a stable output into a flash, or will dim then brighten anyway. I thought it may be a worn spot on the capacitor, but it does it throughout the travel.
In the vid, I say 1000uH, which should be 1000uF
2 minutes demo vid.

https://www.youtube.com/watch?v=fxCWJpHZTyQ

[Bob Smith]

Slider
What would happen to your lights if instead of touching the knob of the variable cap, you were to clip it to: 
a) a wire running to earth ground, or
b) a wire running to a virtual ground (e.g., coil of wire)?
Bob

[TinselKoala]

Well, that's certainly nothing like what I get when scoping the Base wrt Emitter (neg rail) of my board:

(I changed to White LEDs since I don't have 3 good Blue ones any more....     and there is some difference in the waves using the White ones: higher peak voltages mostly.)

(The Auto setting can be a good place to start but I wouldn't always accept its "guess" as to what you want to be looking at... but in this case it does give me a stable view of that inductive ringing, still different from yours. I'm no big fan of the "auto" settings button on DSOs and rarely use it. )


By the way, the answer to the "mystery" of touching the negative battery pole to the circuit and it lighting up, is due to the hand touching the metal battery case; it's another case of capacitive coupling to the body. At least that's what I have determined with my construction.

[Slider2732]

TK - Will do some more scope runs for sure. That trumpet wave is quite the sight and it would be good to compare waveform traces.

Bob - here's a vid for you, of what was quite the interesting experiment. I added a large virtual ground to the variable cap.
A clue to its behaviour is "this is not the oscillator you are looking for".
http://youtu.be/x7GXcpJL2bE
1 min 40 sec video, it's 'processing' at the moment.

[Bob Smith]

Hi Slider
Thanks for the demo and taking time to film and put it online.  There's something special going on there!   
Bob

[Slider2732]

Have enjoyed our chat on the video page itself Bob


Here's a quick video, though not mega quick (lol) of a run with a 10uF tantalum.
Lasts about 30 seconds on a slightly unconventional circuit.
The wiring is to the SJR 4.0, but added are 2x small heatsinks either side of the diode. In this configuration, the circuit runs fine for about 30 seconds on the 10uF. Touching both heatsinks after the run has finished, results in another couple of seconds of light,
Coil is a tiny choke, with 100+200 turns of 40AWG.
The diode here is a 0.5V drop heavy duty Schottky (I think), the legs look capable of passing a good few amps.

https://www.youtube.com/watch?v=lOgCYBsU3x8

[Bob Smith]

 

[Bob Smith]

Hi Slider
Just a thought that I've been considering for a while with this setup - that it may be developed or tweaked to become a kind of open system. My thinking was that attaching a physical or virtual ground might help this along.  A condition for this to take place, as I see it, is oscillation which approaches resonance (in windings).  Bearden has made some helpful comments about the interaction of voltage spikes with the electrostatic environment over the years on video.  I would argue that resonance optimizes this interaction. Some interesting comments have been made about resonance recently on this site as well.
FWIW
Bob

[Slider2732]

I agree Bob and that was a good 'FWIW'

A follow up video was uploaded last night, but sat as 'Processing' for over 2hrs...I went to bed.
The idea being to explore the heatsink interactions. To hopefully find ways of transmitting energies wirelessly, that feed back around and draw in feeder supplies such as 60Hz electrosmog.
Such as, it was noted that with a heatsink either side of the diode and at a particular distance apart, the circuit wouldn't fire up unless the table lamp was on ! When off it wouldn't stay running, on and it would.
The presumed virtual grounding of the body interactions actually work on the positive rail side, not the LED side (which is connected to the coil and therefore may be thought to be the side that would interact). That is in common with my CFL light powered circuits, where the virtual ground of a heatsink is connected to the positive input from the collection circuit and collector plate (a CD usually).  But, this circuit has trouble firing up if the ineffective heatsink isn't present. Therefore it is needed and is doing something.
Dr. Stiffler explored matched heatsinks and their effects on his SEC circuits and this is showing similar results. 2x same aluminium heatsinks from Pentium 4 processors were used in my own tests at that time, but have since been lost...these 2 i'm using now aren't matched, but are at least partially effective.
Next will be scope runs where the 2 heatsinks are brought nearer and further apart...to see if anything happens at different distances of separation.

Here's that vid from last night, Fun and non scientific, after it was found that moving my coffee cup around affected the running of the circuit.
Water is a great energy transmitter and soaker, there are plenty of demo's where LED's on A/V plugs are shown lighting in a glass of water some distance from Slayer Exciters. In this vid, the placement of the coffee on top of a heatsink allows the hand interactions to be much further away from the heatsink and keep the circuit going.
In all, although the output isn't very bright and made dimmer by the table light being on, the 10uF tantalum runs the circuit for over 3 minutes

https://www.youtube.com/watch?v=2OLJHbdc-j0

[LaserSaber]

Hi Slider,

I love the videos that you have been sharing.  The Coffee effect is great.  You are getting some great run time for 10uF.

I will be sharing here more in the future.  I have to get my work for the day finished first.

-LS

[Aking.21]

Welcome Lasersaber.
 

[Slider2732]

Hey welcome aboard  

Thanks. Your little gem is morphing somewhat in my own experiments, but is so much fun to find out about and explore the properties.
Latest, is to move toward exploring capacitive couplings.  
So, rather than extend run times...i felt it to be a good idea to have an always running circuit.
The circuit can now run without any battery, no conventional power input.
Instead, i'm using an ambient energy scavenger to harvest from the table lamp.
Output from such a system can reach 30V unloaded, but there's not much of a current throughput - ideal !

The video is still processing (has happened often in the last few days, used to be almost instant).
It shows it up and running, easily replicable if wished. I would actually like to know how a real SJR 4.0, with correct Litz wire would run in comparison.
Perhaps an idea is to allow a charge up of a flashlight while sitting under a table light, to then be taken outdoors and used. It could be viewed as 'free energy' for the power source, if the light is on anyway. A well designed charging dock for the power transfer makes a good minds-eye route.

http://youtu.be/qmMj0LGgFo4

[LaserSaber]

Great work slider.  I would like to try running this circuit off an Electret.

Here's my latest build: https://www.youtube.com/watch?v=IxAE1tdbHXc

What do you all think about this:

http://www.wired.co.uk/news/archive/2012-03/09/230-percent-efficient-leds
http://phys.org/news/2012-03-efficiency.html

[Slider2732]

Perhaps similar to the electret would be a piezo disc. They output high enough voltage spikes, enough to light an LED in any case. Flexing between the fingers works well, rather than the time honoured technique of hitting them with a blunt instrument. Mounted to the side of a flashlight could work, if indeed a disc charged a small cap.

Enjoyed the update video. Was entertaining to hear you run out of words while the circuit ran on and on.
The mode switches are an intriguing development, as is the move to the larger pot core.
Over here, i've found that the tiny 40AWG thing has worked best for my own latest development.. wireless electricity instead of running through the LEDs.
For some reason the circuit runs best with that tiny one than any other tried.

Here's the video: https://www.youtube.com/watch?v=whl-EV0tpBw
(2min 40sec)

The circuit is running from the CFL output, out through a transmitting coil rather than through the LEDs, which run at just about the same brightness output as when directly wired into the circuit.
That itself was very surprising that the efficiency looks to be very high for the air-core transfer. Easy to do, just replace the LED's with 1x 1N4148 and a 24 turn coil, to complete the connection to the Base of the transistor. The receiver coil should approximate the transmitting coil and benefits from an approx 0.001uF capacitor across it.  


The picowatt LED running points to frequency optimization, as well as heat characteristics. I do know that heating an LED makes photovoltaic properties increase, after having much fun with a lighter in the backyard one time !
Hitting exactly the right frequency is of great interest, to link with the reduction of input voltage on circuits such as the SJR L 4.

[TinselKoala]

Neat series of demonstrations! I like the tuned coil approach. You could probably get more exact receiver coil matching by calculating the L and C values you need to match the "transmitter" frequency.
That is, you can use the measured inductance value of the coil to calculate the capacitor value needed to make the circuit resonant at the transmitter's frequency. Pick a stock value cap closest to what the calculation says, then you can fine tune by removing or adding wire to the coil. Or go the other way, start with a known capacitor value then calculate the inductance needed, then wind your coil to that inductance.
http://www.1728.org/resfreq.htm

I wonder about that big diode. I can't tell if it's a big SOD-64 package device. About 4.5 mm diameter "sphere"? Then that would be SOD-64 package style. Does it have any number markings on it? I have several that I've also pulled from CRT TVs, they say "3A" only.
The smaller but similar-looking BYV26 series of ultrafast rectifier diodes can also sometimes be found in the old circuit boards.
The photo below shows the difference in size.

[Allcanadian]

@lasersaber

First I must say I enjoy watching your video's and your unbiased perspective as if to say...here it is, here is what it is doing and you can decide what you think is happening.

Not to get too far off topic, I was reading the link you posted to 230% efficient LED's and found this statement.

However, while MIT's diode puts out more than twice as much energy in photons as it's fed in electrons, it doesn't violate the conservation of energy because it appears to draw in heat energy from its surroundings instead. When it gets more than 100 percent electrically-efficient, it begins to cool down, stealing energy from its environment to convert into more photons.

I find it odd that Duncan said one of his battery charger circuits showed abnormal energy levels as well as cooling of the terminals. At which point he was criticized for making impossible claims and yet the article from MIT seems to suggest that it is very much possible under the right conditions. I just found it odd that two very similar claims by two different parties could be judged so differently.

In any case MIT has laid the foundation on which we may validate our progress and if they can extract extra energy from the environment then so can we. All the proof we need can be found in their words and I would be willing to bet there are many ways to produce similar results in any number of circuits.

"Experiments directly confirm for the first time that this behaviour continues beyond the conventional limit of unity electrical-to-optical power conversion efficiency."

I should also note that many years ago I mentioned the fact there must be an electrical equivalent to the heat pump, hindsight is always 20/20. It relies on the fact that "heat" is actually EM energy as oscillations in matter and if an engineered field coupled to this EM energy producing a phase differential then energy must be transferred into the system as in any other resonant system. In fact all of what we falsely call thermodynamics relates directly to electromagnetic interactions in matter, there is no "thermo" because thermo is electodynamic in it's nature at the most fundamental level. Particle-fields do not have heat and it is the rate of oscillation of the particle-fields which we describe as heat.

AC

[Slider2732]

@TK - Trial and error is my usual and does need addressing. But, there are a few ball park ways of doing such things, after such a method has been the usual. It tends to be a case of using a variable cap, then matching to the approximate position of travel. I appreciate the advice
Yes indeed, that diode was a surprise. It's a BYM36C, had a look after you raised the point on the video. The datasheet says 1.22V drop, but it measured 0.519V on the multimeter and so was tried. It worked so well that it stayed put and wasn't even investigated for any numbers.
Huge fat legs and a datasheet full of out of bounds numbers would have seen it put to one side usually.
Perhaps the diode ties in with LS's comment in his video, about more current sometimes means better running. Maybe it is dropping 1.22V, but fast and with pinache lol

@AC - There's the point isn't it, if MIT says it's possible then people don't breathe as many garlic laden flame throwers at us guys !
I've no trouble with Duncans observations. Even an SSG here showed a couple of degrees of temperature difference on a MOSFET one time. Sure, it could have been the air of the rotor cooling the metal and legs of the MOSFET, but it has always given pause for thought.  
It does stand to reason that if such measurements have never been taken, then presumption will rule.

[Allcanadian]

@slider2732

@AC - There's the point isn't it, if MIT says it's possible then people don't breathe as many garlic laden flame throwers at us guys !
I've no trouble with Duncans observations. Even an SSG here showed a couple of degrees of temperature difference on a MOSFET one time. Sure, it could have been the air of the rotor cooling the metal and legs of the MOSFET, but it has always given pause for thought. 
It does stand to reason that if such measurements have never been taken, then presumption will rule.

I would go further and suggest that when MIT said "Experiments directly confirm" they did not mean it is a possibility but a fact. We live in exciting times and I believe very soon we will finally acknowledge the fact we are swimming in a sea of energy present everywhere. A while back I came to a very simple understanding which had profound meaning in my life. Look up...every star not unlike our Sun radiates more energy than we can possibly imagine thus all of that empty space out there must be full of energy if energy is to be conserved. It simply goes on and on until it is absorbed in matter thus energy is the norm not the exception to the rule, it is everywhere in everything.

This led to an understanding of what the Primary Fields (Electric, Magnetic and Gravic) are in reality... not what they do but what they are fundamentally. When I finally connected all the dots between what happens out there and how it relates to what happens down here the big picture presented itself. Consider the true reality of what the MIT guy's are saying... their LED gives off more light due to the fact our planet Earth has received energy from a nearby star, the Sun, which this planet revolves around. Ultimately that is where the energy came from and if that simple reality doesn't knock your socks off then I'm not sure what will. I have found it is very easy to get caught up in the nonsense happening down here on this little backwater planet but the true reality of our existence is that we are on a planet revolving around a star which is moving towards the star Vega at 140 miles per second and we should get there in about 230 million years.

I believe T.H.Moray called this insight the universal mind when we discard the simplistic notions of those around us oblivious to the true reality of what we are, where we are in this universe and where we are going. Most have created a false sense of reality down here and it is only when we begin to see past this that we can begin to make progress the likes of which few can even imagine. All we really need to understand is that all that stuff out there may effect what we do down here because we are not separate from it we are immersed in it.

AC

[T-1000]

Here's my latest build: https://www.youtube.com/watch?v=IxAE1tdbHXc

Hi again and welcome to this forum

I've been following your SJR looper experiments and how far they deviated from initial Delamorto circuit. Also I wonder if you did any additional tests when there are foil capacitor plates in ferrite pot and when they get second BEMF signal source for influencing coils inside. The introduction of electrostatic component from capacitive surrounding into coils is somewhat barely explored frontier and might be a key for adding energy into circuit from ambient surroundings (aether)...

Cheers!

[Slider2732]

@AC - am in agreement with your thinking. I picture it rather simply, like the bird sat on the household mains feed from the transformer. It has no idea what it's sat on, but it would if it touched something connected to the ground !
'All' we need to do, is ground in a different way, to connect.

@T-1000 - That seems similar to the pot core experiments of last year, where copper foil separated windings. Yes, it needs more exploration, because that seemed to fade away like it was some kind of fad.


I've returned to the 'better made' version of TK's build method (which was a winner for Lidmotor too I noticed).
With the tiny coil running best of those that i've made so far, I attempted a wind on a core that was even smaller. Like some kind of RSI inducing wish for dexterity troubles, 100+300 turns of 40AWG were wound on it. The thinking was, that i'd better get this sort of thing done now, before arthritis or something strikes in slightly older age !
End result, a great success, thankfully.
Run cap, 470uF.
Attached is a pic at 26 minutes, which I think is the benchmark now for a lower light output run ?
It ran onward until 45 minutes, with flickering starting at 40 minutes. That result means that I can further reduce the run cap from a '472' and brighten it slightly. The odd trait it has at the moment, is that it begins its run with a bright output for a couple of seconds, then dips into its basically solid lower light level.
Am using a germanium diode so far on this one, but will try one of the surprisingly good BYV type series fast rectifiers too.
(Harpbloke mentioned on my last upload that he put one of those in sunlight and got 300mV. I just did the same, at 6.45pm and saw 200mV...quite surprising for something that looks like it's made of a ceramic material.

[Slider2732]

Some testing results:

The wireless version of the circuit was tried as a AAA Ni-Cd battery charger.
Really, just to see if anything happened, because the throughput is so low. An ex solar garden light 100mAh AAA was connected via a 1N4148 diode and '101' ceramic capacitor, to a 12 turn coil and left sitting on the transmitting coil overnight. We leave a light on overnight, so the thinking was to perhaps make some use out of that fact.
Starting at 0.685V at 11pm, it was at 0.808V by 8am. Not a lot, but worth the run to have a look. Without the battery in place, voltages between 1V and 1.3V were noted, so a better made system may show improvements and work as a trickle charger.

Next, was to see what happens in the form of charging a capacitor within the original schematic based circuit.
A well discharged 470uF (to match the run capacitor) was placed over an output LED.
After 1 minute, the cap showed 90mV.
Reducing to 22uF, after 1 minute it sat at 0.504V
Reduced again to 1uF, the reading was 3.5V
The thinking, was that perhaps a capacitor could charge, at no detriment to the running of the circuit, which could then have a useful charge be periodically thrown back into the run cap.
During the runs, the other 2 LEDs brightened...which was not altogether expected. The cap could easily have completely killed off the output as it charged from practically zero, or at least have caused no effect to the other 2 as it charged.

Here are 2 scope shots, from the 100:300 micro coil version of the circuit.
The first is across ground and MPSA18 Collector.
The second is across ground and Base.

[Slider2732]

Just a quick update, noting this is my third post in a row.

2 single winding coils were tried, 1 on top of the other and secured together with a spot of superglue.
One is larger than the other, approx twice as large, with a similar gauge of wire used in both.
It works. Not as well as other methods, but ran for 10 minutes on a 470uf cap.
Also, a 2 pin connector is now being used to try out different capacitor sizes. Makes things easier.