Hey guys,
A buddy just let me know about your little discussion, so I thought I'd enter the thread and help you along with some of your doubts, guesses and questions.
OK, let's do 'em in order.
- "I'm guessing some kind of joule thief."
It's not a Joule Thief, differing in a few ways; since it has no coil/step up method of modifying the voltage/taking the voltage up from an average of 1.1 V DC up to something higher in order to meet the demands of the LEDs.
The switching is also different from the typical basic Joule Thief circuit in that it uses a PNP as well as an NPN. The PNP is used because it has a nice 'n' low gate voltage of .7 to .9 V DC, which means that the AA can handle the job until it drops below that value, as seen in the video when I take readings near the end of the test.
As well, the Joule Thief hits the light bulb end of the circuit with a fairly A/C shaped wave form, all of which is typically directed (undioded) at a light type that makes use of A/C power.
This is an LED setup for a reason.
And that reason is that the LEDs are connected in opposite polarity to the battery.
So the battery simply isn't the thing that is powering the LEDs.
- “No magic there - LEDs will light with much less than 20mA. An AA battery will give at last 3 watt-hours. Or 62mw over 2 days. A low-current LED will light as brightly as those shown on 1mA, =approx 2mW. So lighting 23 LEDs on an AA battery is entirely plausible.”
These LEDs are clearly wired in parallel.
A shitkicked year old alkaline Costco issue Kirkland will NOT 'give at least 3 watt hours'.
Check the independent quotes from others online who have published their testing of many kinds/brands of AA battery. You will find that this grade of Alkaline give out something much closer to the value I suggest in the video - around 2,184mAh, times the average of 1.1 V, gives you a total output of 2,310 mWh. That is exactly how you would expect this thing to drain if there were no LEDs connected at all.
What I don't show in the video, and I kinda wish I had in hindsight, is the mA draw without and then with load. At the top end, when we're talking fairly fresh battery, around 64mA, the load drives the draw up by around 4 or 5mA.
And no, you can't get an LED to near full brightness on 1.4 V and 1 mA. That's just crap talk. Sorry.
- “The guy's doing it without understanding what he's doing.”
Dude. I didn't ask the question at the end because I don't know the answer.
And I didn't end up with those 5 particular cores, with the correct permeability/saturation value required to exhibit this phenomenon, by happy accident.
I've been an avid student of the quantum vacuum for 16 years.
Pretty sure I understand what I'm doing, thx.
-”Back of envelope time:
An alkaline AA has a capacity of about 2500mAh at low currents
high efficiency LED's can easily operate at that level of apparent brightness @ 1mA or less, even 0.1mA I've done before (brightness is pretty linear).
23 red LEDs in series at maybe 1.6V drop at that current is say 37V, so @ 1mA that's 37mW total.
Say 1.2V average for the cell is 3Wh capacity, divide that by 37mW and you get 81 hours.
Add some loss in efficiency for whatever step-up converter you use, and Bob's Your Uncle, 52 hours is no problem at all.”
OK. They're not in series. And you can clearly see that in the video. If you own a breadboard, you know the two rails on each long side of the board. They're colour coded red and blue.
These are not high efficiency LEDs. They are the dime a dozen kind, bought literally as a bag of 100, for a few bucks.
I think I'll also put up a video showing what their diet is in terms of conventional DC. Because they're hungry little buggers compared to the order of magnitude of the rest of what's going on in this circuit.
It's certainly a lot more than 37mW. That's just silly.
And there is no step up converter, here.
That is actually why I put in the two quotes from the Matrix. Back when I was still putting the video together, right after the test was done, I was using that reference when talking to another buddy of mine elsewhere in the world (who I have been mentoring in various technologies for a few years), in describing how this simply isn't a Joule Thief, because there is no coil (“there is no spoon”).
-”hey are using a joule thief which is powered by itself once it starts up (that is, the oscillator power is supplied by the transformer.) This will mean it can oscillate down to ~0.3V. These circuits have been around for ages and without actual power measurements no claim to extracting more than the batteries' energy can be upheld.”
Nope. No JT, here, dood. Just check the two schematics side by side and you can clearly see that, assuming you are fluent in schematic. This video is not so typical of my regular output on that channel. Check out a few of my other ones and you'll quickly see a higher production value. I typically do high detail custom schematics and diagrams for my output, always each symbol and part from scratch, using software like Inkscape instead of the expected prefab circuit software options currently out there.
It won't oscillate below the PNP's gate threshold voltage, which you can confirm is between 0.7 and 0.9 V DC per the datasheet - part # 2SA949, Toshiba. And this is why you see the test last for only a short while after the 0.905 V point had been reached.
And I gave you power measurements at the start and at the end. Don't know what you're talkin' about there, dood.
- “I am very dubious about the identity of that large round component at the top cleverly covered in black tape.
It looks like 5 battery cells in series to me. The video is long on hype and mystery and pretty devoid of hard facts.
I say humbug. Move on, nothing to see here.”
The part # for that component, on DigiKey, is 495-3867-ND - Ferrite Toroid 38.1mm OD.
They're pretty big. And there's simply no room in there for anything else. Besides, CapIndRes has already done the 'taking the thing apart at the end of the demo' thing, to take this whole issue into the Land of the Moot.
This isn't a deception. It's a REPLICATION. I'm suprised that more of you haven't taken the time to check out the output of both Larskro and CapIndRes regarding this technology. Your comments would drammatically change, should you take the time and show some respect.
I s'pose I could have glued the 5 toroids together in order to get rid of some tape and expose the outside paintjobs of the pretty cores, but the naysayers would still try to point out that I could have hollowed them out and snuck a small battery inside each one.
But if you took the time to really attempt this circuit, you'd find that it really does work, just like I did.
- “the black taped thing is apparently A brass tube surrounded by ferrite cores.”
If you pay attention while watching the video, I describe in detail what they really are.
I also detail that you could substitute what I used for a brass tube and get the same result. It was a point about how the resistance of whatever connects the part of the circuit going through the core set is essentially zero when the battery is pushing current though during the ON phase.
I use 4 8 foot lengths of 28 gauge EM wire, soldered at each end.
That results in a 4 X 28 gauge parallel Litz wire. And the final length, after twisting the wires together, is closer to 5 feet. So it's the conventional resistance of a single 5 foot length of roughly 18 or 16 gauge. But you get a better magnetoelectric skin effect transfer toward the dielectric using what I use. That is coiled up into around 8 or 10 turns on its way through the cores. Do I still not sound like I know what I'm talking about?
But again, you can just use a straight ol' piece of brass - or a few other things; just check out their videos to see that as well.
- “It looks like just another stupid amateur ad-hoc YouTube video. There are millions of them out there.”
OK. That is just ignorant. And quite trollish.
Seriously, guys. I am known by many in the community.
Endorsement (most recent one) by Matt McMahon, in his review of some of my recent output:
"Stone, this is yet another beautiful creation of yours. You're one of the smartest guys I know and there can really be no value put on the quality of your work. Great job. Beautiful contribution to the HHO community."
By all means let me know if you have any other doubts, questions, respectful comments.
Thanks for taking a look,
Stone
Home
Help
Search
Login
Register
