Storing Cookies (See : http://ec.europa.eu/ipg/basics/legal/cookies/index_en.htm ) help us to bring you our services at overunity.com . If you use this website and our services you declare yourself okay with using cookies .More Infos here:
https://overunity.com/5553/privacy-policy/
If you do not agree with storing cookies, please LEAVE this website now. From the 25th of May 2018, every existing user has to accept the GDPR agreement at first login. If a user is unwilling to accept the GDPR, he should email us and request to erase his account. Many thanks for your understanding

User Menu

Custom Search

Author Topic: 3v OU Flashlight  (Read 563409 times)

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: 3v OU Flashlight
« Reply #345 on: April 30, 2014, 03:36:41 PM »
Hi TinselKoala,

Do you have the time to test your circuit according to the attached circuit?

GL.

Yes, I have time and that test is on my schedule, already anticipated.    ;)

Thanks for your work on the other circuit schematic by the way, I'll be starting a build on that one later this afternoon.

Meanwhile, I've cleaned up a few discrepancies and am now able to get a much more stable performance over a wider "tuning" range. Here's a shot of the current performance. This is at 3.0 V, about 70 mA input. (Since this is a pwm circuit, the current draw is set by the tuning and can go anywhere from zero (with LEDs dark) to around 400 mA with brilliant LEDs (no burst oscillations)).

Unfortunately the LEDs still fade out over a couple of seconds when the input power is removed.

ETA: I removed the 2.2 pF cap across the R2.

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: 3v OU Flashlight
« Reply #346 on: April 30, 2014, 03:39:14 PM »
hi TinselKoala,

It looks like you are following the Akula variation.Have you considered implementing resistor in series with Led(Limit current to led) and discharge resistor(around 560...3.3k ) connected in parallel with C5 100uf to ensure the sine wave looks like nice sine wave. :)

-----------------------------------

Finally received the 4.7cm diameter Ferrite pot core as attached after waiting for nearly one month.Now i shall patiently wait for copper foil to magically appear at my door step.  :D




 


No, I hadn't considered it, but now I will. Thanks for the suggestions.

ETA: Bridging the C5 with reasonable resistance values had no real effect that I could tell. But adding a 100R at the cathode end of the LED string to ground did help to stabilise the envelope! Thanks for that suggestion. It added a little current but made the burst oscillation/sinus waveforms a lot more stable, actually. Thanks again!

Unfortunately... the LEDs still dim and go out when the power is removed.

wattsup

  • Hero Member
  • *****
  • Posts: 2606
    • Spin Conveyance Theory - For a New Perspective...
Re: 3v OU Flashlight
« Reply #347 on: April 30, 2014, 04:51:33 PM »
@all

Very sorry for interjecting here. Just wanted to know if you guys are having any trouble getting to page 69 on the Akula 30 thread. Hmmmmmmmmm. All other pages are fine except the last page 69. Maybe the number says it all. hahaha

While I'm at it, as some are working this circuit, please take a known object on the circuit, get its dimensions, then extrapolated the dimensions of the ccore Akula is using. Very important to do now.

wattsup

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: 3v OU Flashlight
« Reply #348 on: April 30, 2014, 04:51:41 PM »
What is the result? LEDs are lit without 3v power?

Not for me. Only for Akula, in Akula's laboratory.




"yet"

 :P

MenofFather

  • Hero Member
  • *****
  • Posts: 943
Re: 3v OU Flashlight
« Reply #349 on: April 30, 2014, 04:53:46 PM »
Acording what akula about 1W lantern 2 (with supply) and 3 (without supply) one frenquency is 4-7 herc and duty cycle 50 precents, other frenquency is 400-460 herc. But ferroresonance frenquency can be higher, for example 1.6 megaherc. So duty cycle of 400 kiloherc, let say, frenquency must be 12 precents.


But inportant is fast closing one and other mosfet, to not make mistakes, better use wery fast closing one and other mosfet. Then we hear sound of 5-15 kiloherc in video, here is sound of mixed two frenquencies, how, something like that, explain Akula.

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: 3v OU Flashlight
« Reply #350 on: April 30, 2014, 05:02:47 PM »
@all

Very sorry for interjecting here. Just wanted to know if you guys are having any trouble getting to page 69 on the Akula 30 thread. Hmmmmmmmmm. All other pages are fine except the last page 69. Maybe the number says it all. hahaha
It doesn't seem to be working for me either, I can't get to any pages there. Probably somebody posted a huge image or something.

Or the MiBs have gotten to it finally....
Quote

While I'm at it, as some are working this circuit, please take a known object on the circuit, get its dimensions, then extrapolated the dimensions of the ccore Akula is using. Very important to do now.

wattsup
Huh? Can you translate that into a language I might understand?

I am using a flyback transformer ferrite, made of two identical halves. The round part that goes inside the coil is 14.4 mm diameter and the square part outside the coil is 14.4 mm x 11.5 mm. The overall length when assembled is 62.8 mm and the width is 40.0 mm. It has the stock spacer glued to one face of the squareish part, mica I think, 0.33 mm thick, about. It's held together by a spring clip that fits into grooves in the outside surfaces of the squarish part. Right now I am using it without any extra spacers. I like it very much.

As far as I can tell this is not an irradiated Soviet ferrite. It measures 50 ohms/cm on the surface, contacted with point probes. I think it came from a Japanese TV set, Mitsubishi big screen, IIRC.

magpwr

  • Hero Member
  • *****
  • Posts: 1168
    • Youtube Channel
Re: 3v OU Flashlight
« Reply #351 on: April 30, 2014, 05:04:56 PM »

No, I hadn't considered it, but now I will. Thanks for the suggestions.

ETA: Bridging the C5 with reasonable resistance values had no real effect that I could tell. But adding a 100R at the cathode end of the LED string to ground did help to stabilise the envelope! Thanks for that suggestion. It added a little current but made the burst oscillation/sinus waveforms a lot more stable, actually. Thanks again!

Unfortunately... the LEDs still dim and go out when the power is removed.

hi Tinselkoala,

It seems there is indeed improvement in your waveform. :)

Do try to ignore the 270hz ,as the core value used is different from Akula.If you force it to be even close the 270hz ,you might be compromising efficiency unknowingly.

I was hoping you used 220...330ohms in series with Led.But i think it's ok with 100 ohms. :-X
"blue Led do consume little more current than typical green or white led."

The high efficient green led i used in the original circuit connected to 4069 is 27kohms instead of 1kohms and yet i can see it's bright enough.The old green led purchased over 10years ago don't even come close."Pri objective was to lower circuit current draw before even receiving core or MC34063."

The shape of blue waveform  near the top peak looks a little strange.I think i have seen this waveform before likely related to capacitor value 100uf or 1000uf which might be little high do try mixing capacitor eg:47uf+22uf=69uf or 47uf+47uf(low ESR this way) to observe the effect.

lost_bro

  • Sr. Member
  • ****
  • Posts: 282
    • youtube channel
Re: 3v OU Flashlight
« Reply #352 on: April 30, 2014, 05:05:41 PM »
Hello All

Sorry Off topic:

Page 69 of akula0083-30-watt-self-running-generator doesn not open for me this morning...

Anyone else with same problem?

thanks
take care, peace
lost_bro

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: 3v OU Flashlight
« Reply #353 on: April 30, 2014, 05:08:16 PM »
Acording what akula about 1W lantern 2 (with supply) and 3 (without supply) one frenquency is 4-7 herc and duty cycle 50 precents, other frenquency is 400-460 herc. But ferroresonance frenquency can be higher, for example 1.6 megaherc. So duty cycle of 400 kiloherc, let say, frenquency must be 12 precents.
This is theory, not useful to me at present.
Quote

But inportant is fast closing one and other mosfet, to not make mistakes, better use wery fast closing one and other mosfet. Then we hear sound of 5-15 kiloherc in video, here is sound of mixed two frenquencies, how, something like that, explain Akula.

This is empirical fact, very useful. The configuration of the dual 494s in the circuit is arranged to produce this kind of "beating", that is the sound of mixed frequencies creating a third frequency from reinforcement and cancellation of the two input signals. Beat notes.

Now yesterday as I recall you were saying that only one of the mosfets needs the Miller clamp for fast shutoff. But now are you saying they both should have the clamp? It's a good thing I bought two PNP transistors, then....

But this thread is about a different circuit altogether. You are talking about the other circuit, with 2x TL494 and 2 mosfets. This thread is about the circuit using the MC34063 and the hex inverter chip, single mosfet (transistor).

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: 3v OU Flashlight
« Reply #354 on: April 30, 2014, 05:18:33 PM »
@magpwr: At first I thought that the 270 Hz was due to the circuit picking up some stimulus from the environment. I still might think that, I'm not sure. But if it is, I clearly have the same stimulus in my environment here.

So it seems that you are saying that there is some property of the core, not related to inductance, that determines "efficiency" and that my core, being different.... even though it makes nearly identically the same frequency, same waveforms by using the same external component values.... only the pot setting could be different.... might need to be tuned to some other frequency to reach the necessary "efficiency" to keep the LEDs lit when power is removed?

Sorry, I don't buy that, it's hand-waving special pleading. Let's see some working device, from someone other than Akula, that uses this schematic, works at a different frequency and keeps the LEDs lit. That would be evidence. Or even.... send Akula a Japanese ferrite and let him show how it makes the LEDs stay on at a different frequency.

LOL.....

 :P

wattsup

  • Hero Member
  • *****
  • Posts: 2606
    • Spin Conveyance Theory - For a New Perspective...
Re: 3v OU Flashlight
« Reply #355 on: April 30, 2014, 05:29:58 PM »
@TK

About the core size, if you take any component on a photo like a mosfet that we know has x dimension, you can figure out the core dimension Akula is using. Your measurements are close to what I have as well but is it close to the one Akula is using.

I am saying this because yesterday when I received my ETD29 ecores for the AKula 30 device, I realized that this particular ecore, that others are using as well, is way to small. Even your ccore is at least twice the size. So just make sure the core sizes are comparable to the Akula device you are working on.

All the best.

wattsup


MenofFather

  • Hero Member
  • *****
  • Posts: 943
Re: 3v OU Flashlight
« Reply #356 on: April 30, 2014, 06:54:55 PM »
This is theory, not useful to me at present.
This is empirical fact, very useful. The configuration of the dual 494s in the circuit is arranged to produce this kind of "beating", that is the sound of mixed frequencies creating a third frequency from reinforcement and cancellation of the two input signals. Beat notes.

Now yesterday as I recall you were saying that only one of the mosfets needs the Miller clamp for fast shutoff. But now are you saying they both should have the clamp? It's a good thing I bought two PNP transistors, then....

But this thread is about a different circuit altogether. You are talking about the other circuit, with 2x TL494 and 2 mosfets. This thread is about the circuit using the MC34063 and the hex inverter chip, single mosfet (transistor).
I seems make mistake, need fast close one and other mosfet. But maybe more important fast close mosfet who works on feroresonance (200-400 kiloherc can be mosfet frenquency).
I make mistake and write into not needed topic. But no is topic about 1 W akula divice (lantern 2 and 3), so were me write this words or is that topic?

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: 3v OU Flashlight
« Reply #357 on: April 30, 2014, 07:22:45 PM »
I seems make mistake, need fast close one and other mosfet. But maybe more important fast close mosfet who works on feroresonance (200-400 kiloherc can be mosfet frenquency).
I make mistake and write into not needed topic. But no is topic about 1 W akula divice (lantern 2 and 3), so were me write this words or is that topic?
No matter, I can use the Miller clamp on both or just the lower one. I have transistors!

It's being discussed in the "Akula 30 W " thread, which is having problems today, I think.

Meanwhile here are two zooms into the oscillation/noise portion of the signal. The first one is at 500 microseconds per division, the second one is at 20 microseconds/div.  The square-topped pulses are the normal output of the MC34063 chip. They are happening at a variable frequency controlled by the circuit's feedback loops and are happening here at a frequency of about 25 kHz max in the "bunched up" bursts.  The spike and ringdown on the very top... seen in the second picture...  there are about four peaks in 20 microseconds, for a frequency of about 200 kHz for this resonant ringdown. If I did the math right.

Is this the "ferroresonance" you are talking about?

(Both shots are from the same data capture, I just zoomed in to make these images. The dotted red lines are the trigger location, so you can orient the two shots.)

ETA: The bursts themselves are happening at about 300 Hz in this shot, and this is the frequency one can feel and hear in the core vibrations.

MenofFather

  • Hero Member
  • *****
  • Posts: 943
Re: 3v OU Flashlight
« Reply #358 on: April 30, 2014, 07:33:40 PM »
No matter, I can use the Miller clamp on both or just the lower one. I have transistors!

It's being discussed in the "Akula 30 W " thread, which is having problems today, I think.

Meanwhile here are two zooms into the oscillation/noise portion of the signal. The first one is at 500 microseconds per division, the second one is at 20 microseconds.  The square-topped pulses are the normal output of the MC34063 chip. They are happening at a variable frequency controlled by the circuit's feedback loops and are happening here at a frequency of about 25 kHz max in the "bunched up" bursts.  The spike and ringdown on the very top... seen in the second picture...  there are about four peaks in 20 microseconds, for a frequency of about 200 kHz for this resonant ringdown. If I did the math right.

Is this the "ferroresonance" you are talking about?

(Both shots are from the same data capture, I just zoomed in to make these images. The dotted red lines are the trigger location, so you can orient the two shots.)
I not shore is here feroresonance. Here is other circuit. To find ferorerosonance use one turn on primary and one turn on secondary, on secondary put 10 omh resistor and oscilioscope. And put into primary from range 800-2000 kiloherc frenquency of 5 volts and were you get maximum amplitude on that resistor, here and is ferroresonance. Seems i write coresct. Akula say about one turn... but other part of his words I now not remember. Need relisten conference....
Or you can use other numbers of turns, if you get same frenquency of fading out vibration, then here can be ferroresonance...

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: 3v OU Flashlight
« Reply #359 on: April 30, 2014, 07:58:19 PM »
But... but.... TK.... you may say.... what about the envelope? Yours is all wavy and Akula's maintains a constant amplitude.

Ok... then I'll just have to turn off Bandwidth Limiting on my scope.



No changes in Any Settings, except turning off BWL in the oscilloscope: