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: Kapanadze Cousin - DALLY FREE ENERGY  (Read 11715836 times)

Marsing

  • Sr. Member
  • ****
  • Posts: 300
Re: Kapanadze Cousin - DALLY FREE ENERGY
« Reply #4830 on: July 23, 2014, 10:38:03 AM »

just 200kHz (oscilator frequncy) is 500nS still out of nano range!



edit   :)

 T = 1/F

example

1 / 300 KHz  = 3,3333333333333333333333333333333e-6   -->    3,3 micro second
1 / 2   MHz  = 0,000 000 5                                                  -->    500  nano second 

etc ...

If  F = 300 KHZ  AND Duty cycle = 10 %    then     3,3 uS / 10 = 330 nano second
IF  F = 2 MHZ    AND Duty cycle = 10 %    then     500 nS / 10 =  50 nano second.

etc ...   

So . we already got nanosecond pulse using tl494, with condition duty cycle = 10 % AND F = 300 KHz


...

itsu

  • Hero Member
  • *****
  • Posts: 1845
Re: Kapanadze Cousin - DALLY FREE ENERGY
« Reply #4831 on: July 23, 2014, 11:36:52 AM »
Itsu,

Its strange you still get this 'special' mode as I never saw it at any voltage. One thing that might be worth trying is to increase the inductance of your choke to see if you can lose the effect.


Hi Hoppy,

the choke is extremely sensitive concerning this special mode.
The used 250uH choke is just on the border, when i tested with 800uH, 5.5mH and 7mH chokes, they all caused the Mazzilli to go into this special mode (@24v INPUT).

I will reposition this choke (close to the input terminals) on my circuit to see if the present location perhaps is causing the feedback/oscillations.


Thanks,  regards Itsu

zcsaba77

  • Full Member
  • ***
  • Posts: 245
Re: Kapanadze Cousin - DALLY FREE ENERGY
« Reply #4832 on: July 23, 2014, 12:26:30 PM »

edit   :)

 T = 1/F

example

1 / 300 KHz  = 3,3333333333333333333333333333333e-6   -->    3,3 micro second
1 / 2   MHz  = 0,000 000 5                                                  -->    500  nano second 

etc ...

If  F = 300 KHZ  AND Duty cycle = 10 %    then     3,3 uS / 10 = 330 nano second
IF  F = 2 MHZ    AND Duty cycle = 10 %    then     500 nS / 10 =  50 nano second.

etc ...   

So . we already got nanosecond pulse using tl494, with condition duty cycle = 10 % AND F = 300 KHz


...
but 300kHz with 10% duty cycle is on border, and just 330nS!!!
and TL494's datasheet show just 200kHz!!!
maybe with 2MHz if exist

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: Kapanadze Cousin - DALLY FREE ENERGY
« Reply #4833 on: July 23, 2014, 12:32:19 PM »

edit   :)

 T = 1/F

example

1 / 300 KHz  = 3,3333333333333333333333333333333e-6   -->    3,3 micro second
1 / 2   MHz  = 0,000 000 5                                                  -->    500  nano second 

etc ...

If  F = 300 KHZ  AND Duty cycle = 10 %    then     3,3 uS / 10 = 330 nano second
IF  F = 2 MHZ    AND Duty cycle = 10 %    then     500 nS / 10 =  50 nano second.

etc ...   

So . we already got nanosecond pulse using tl494, with condition duty cycle = 10 % AND F = 300 KHz


...


330 nanoseconds is not "nanosecond pulsing" it is "hundred nanosecond pulsing". You are still off by two orders of magnitude. Besides, it isn't the _length_ of the pulse that is critical, it is the rise and fall times. Certainly, short pulses must have even shorter rise and fall times, but if you can oscillate at 300 kHz _and_ make your rise and fall times be under 10 ns, you will be in the right ballpark.  The genuine TI-manufactured TL494, with careful circuitry, can be made to oscillate reliably at more than 800 kHz and with careful output stage pulse shaping can deliver a final power pulse train with the needed nanosecond-class rise and fall times. Don't believe me? See my "TinselKoil 2" videos.

magpwr

  • Hero Member
  • *****
  • Posts: 1168
    • Youtube Channel
Re: Kapanadze Cousin - DALLY FREE ENERGY
« Reply #4834 on: July 23, 2014, 02:12:13 PM »
but 300kHz with 10% duty cycle is on border, and just 330nS!!!
and TL494's datasheet show just 200kHz!!!
maybe with 2MHz if exist

hi zcsaba77,

I have done comparison between TL494 and 3525 PWM I/C and find that both have the nearly the same spec.
The finding on 3525 is it will produce nice waveform to around 200khz any higher i would recommend using PWM I/c:3825 which is able to do 1Mhz max.

For the recent basic Don Smith demonstration to itsu i was using 3825.

I have attach my PWM circuit based on 3825.

-----------------------------------
Side track project-
I have just assembled the 5mhz induction heater today but i'm using slightly inferior available mosfet IRFP450(higher Rds on resistance) instead of IRFP460.
http://4hv.org/e107_plugins/forum/forum_viewtopic.php?142264

Officially this the first time i replicated something that made in Russia and it works base on schematic. :)

I was able to get metal nail really hot but wasn't able to see glow on nail using variable 30....40volts 10Amps power supply.

I'm using some excess silver-mica capacitor 1600volts 2000pf for this induction heater project which was previously purchased from Ukraine around 2 months back.
"This is to be used in the snubber circuit for the Don smith as found in page 260 of pjkbook.pdf."

For 1.5nf i am using available mica capacitors 560nf x 3 in parallel.

Placing oscilloscope probe nearby the frequency seems started with around 5mhz then it slowly started drifting slightly higher to around 5.2mhz(I think this is due to the silver-mica capacitor is getting hotter)

Current consumption is around 40voltsx 1.6Amp and slowly increasing.




 

zcsaba77

  • Full Member
  • ***
  • Posts: 245
Re: Kapanadze Cousin - DALLY FREE ENERGY
« Reply #4835 on: July 23, 2014, 02:14:28 PM »
The genuine TI-manufactured TL494, with careful circuitry, can be made to oscillate reliably at more than 800 kHz and with careful output stage pulse shaping can deliver a final power pulse train with the needed nanosecond-class rise and fall times. Don't believe me? See my "TinselKoil 2" videos.

I dont found it. Please link! Thanks

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: Kapanadze Cousin - DALLY FREE ENERGY
« Reply #4836 on: July 23, 2014, 02:57:53 PM »
I dont found it. Please link! Thanks

Here's the 830 kHz version. Later I made a new resonator and brought the frequency down to a less stressful 350 kHz or so. I found that when "overclocking" the TL494 the two complimentary outputs aren't so symmetrical any more, so I just used the + output and split-inverted that using a couple of 2n7000 minimosfets after the TL494, to drive an intermediate current amplifier stage before the main mosfet H-bridge. If you use a mosfet driver chip you can eliminate the intermediate stage but for overclocking I would still recommend just using the single 494 output and splitting/inverting externally to get your complementary square wave drive signals.
Also, if overclocking, make sure your chips are genuine Texas Instruments TL494.

http://www.youtube.com/watch?v=7BMRJoqdy6E

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: Kapanadze Cousin - DALLY FREE ENERGY
« Reply #4837 on: July 23, 2014, 03:06:24 PM »
(snip)
-----------------------------------
Side track project-
I have just assembled the 5mhz induction heater today but i'm using slightly inferior available mosfet IRFP450(higher Rds on resistance) instead of IRFP460.
http://4hv.org/e107_plugins/forum/forum_viewtopic.php?142264

Officially this the first time i replicated something that made in Russia and it works base on schematic. :)
This is really "cool" project, I'm starting one myself. I think that some improvements can be made, especially with layout. It looks like an effective, simple and cheap small induction heater. My wireless systems also work well as induction heaters but are too low-frequency to be very convenient for small items that I want to heat, and this one seems ideal.
Quote
I was able to get metal nail really hot but wasn't able to see glow on nail using variable 30....40volts 10Amps power supply.
I think he was using 50 V or so for his input. Did you also put a light bulb in series to see the current?
Quote

I'm using some excess silver-mica capacitor 1600volts 2000pf for this induction heater project which was previously purchased from Ukraine around 2 months back.
"This is to be used in the snubber circuit for the Don smith as found in page 260 of pjkbook.pdf."

For 1.5nf i am using available mica capacitors 560nf x 3 in parallel.
Uh... you must mean 560 pF  .... ?
Quote

Placing oscilloscope probe nearby the frequency seems started with around 5mhz then it slowly started drifting slightly higher to around 5.2mhz(I think this is due to the silver-mica capacitor is getting hotter)

Current consumption is around 40voltsx 1.6Amp and slowly increasing.
Or the mosfet itself. As it heats its Rdss rises and this will change the frequency too. I think also the coil and where it is tapped will be critical for good performance too. As the frequencies of our devices rises, our construction, layout and small differences in components become increasingly significant. I usually don't work much with frequencies in the MHz range for that reason. 5 MHz is pretty high for clipleads and breadboards.

magpwr

  • Hero Member
  • *****
  • Posts: 1168
    • Youtube Channel
Re: Kapanadze Cousin - DALLY FREE ENERGY
« Reply #4838 on: July 23, 2014, 03:33:33 PM »
This is really "cool" project, I'm starting one myself. I think that some improvements can be made, especially with layout. It looks like an effective, simple and cheap small induction heater. My wireless systems also work well as induction heaters but are too low-frequency to be very convenient for small items that I want to heat, and this one seems ideal. I think he was using 50 V or so for his input. Did you also put a light bulb in series to see the current?Uh... you must mean 560 pF  .... ?Or the mosfet itself. As it heats its Rdss rises and this will change the frequency too. I think also the coil and where it is tapped will be critical for good performance too. As the frequencies of our devices rises, our construction, layout and small differences in components become increasingly significant. I usually don't work much with frequencies in the MHz range for that reason. 5 MHz is pretty high for clipleads and breadboards.

hi TinselKoala,

This  5mhz mosfet base oscillator is indeed a cool circuit.Later i will connecting the drain to the bottom of tesla coil to see if i can get any special effect using 5mhz :)

You will need to use light bulb for sure but later part i just remove bulbs and use multimeter which is set in 10Amp range.I'm using 22volts TVS diode(Planned to use IGBT later on) in parallel with 1uf ceramic cap to maintain a stable voltage."I have tested it in multisim but can't get 5mhz in virtual lab since i'm not using right coil".
I'm not using 10uf but i'm using 10,000uf 50volts capacitor.There is also 330nf 63v cap connected at the circuit power supply.

The bulb needs to be dim but upon inserting it needs to be brighter and not another way around."You will find out soon if the osc just stops if variable voltage is set too high"

Oops it was typo error suppose to be 560pf x 3 to get around 1.5nf.

I am using 14awg for the coil.total 8 turns ,3(center tap to +)5 turns.

All the pins needs to be soldered closely at this frequency ,breadboard at this frequency is out of the picture.

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

Latest update-1N4148 base AVR plug to white led does glow if placed around 1inch away from work coil.


Marsing

  • Sr. Member
  • ****
  • Posts: 300
Re: Kapanadze Cousin - DALLY FREE ENERGY
« Reply #4839 on: July 23, 2014, 03:53:45 PM »

330 nanoseconds is not "nanosecond pulsing" it is "hundred nanosecond pulsing". You are still off by two orders of magnitude. Besides, it isn't the _length_ of the pulse that is critical, it is the rise and fall times. Certainly, short pulses must have even shorter rise and fall times, but if you can oscillate at 300 kHz _and_ make your rise and fall times be under 10 ns, you will be in the right ballpark.  The genuine TI-manufactured TL494, with careful circuitry, can be made to oscillate reliably at more than 800 kHz and with careful output stage pulse shaping can deliver a final power pulse train with the needed nanosecond-class rise and fall times. Don't believe me? See my "TinselKoil 2" videos.

Sure, 330 nS are in hundreds and i didn't yet consider fall times and rise times. that's just an example of calculation. and i don't think we need to make a signal with 1 ns precession , by now duty cycle is the key i guess.

what do you mean " You are Still off By two orders of magnitude" ?,   HA...    HA...

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

maybe we can make 1 % duty cycle.

if  F =  300 Khz    AND   Duty cycle=  1 % then    3,3us / 100 = 33ns
if  F =  1 Mhz      AND   Duty Cycle=  1 % then    1  us /  100 = 10ns   

now we have  PWM 3525 to generate signal up to 1Mhz by magpwr and TK 830 Khz.
...

edit....    '3825'    not  '3525',    sorry.
« Last Edit: July 23, 2014, 06:41:13 PM by Marsing »

magpwr

  • Hero Member
  • *****
  • Posts: 1168
    • Youtube Channel
Re: Kapanadze Cousin - DALLY FREE ENERGY
« Reply #4840 on: July 23, 2014, 04:04:45 PM »
Sure, 330 nS are in hundreds and i didn't yet consider fall times and rise times. that's just an example of calculation. and i don't think we need to make a signal with 1 ns precession , by now duty cycle is the key i guess.

what do you mean " You are Still off By two orders of magnitude" ?,   HA...    HA...

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

maybe we can make 1 % duty cycle.

if  F =  300 Khz    AND   Duty cycle=  1 % then    3,3us / 100 = 33ns
if  F =  1 Mhz      AND   Duty Cycle=  1 % then    1  us /  100 = 10ns   

now we have  PWM 3525 to generate signal up to 1Mhz by magpwr and TK 830 Khz.
...

hi,

It's suppose to be "3825" for max 1mhz.The lowest duty cycle i have got is around 4% at 230khz.

Thaelin

  • TPU-Elite
  • Hero Member
  • *******
  • Posts: 1093
Re: Kapanadze Cousin - DALLY FREE ENERGY
« Reply #4841 on: July 23, 2014, 10:10:57 PM »
Enjoy your RF burns at that freq. That is why most of the circuits
stay fairly low as not to emit too much. Not going to be good for
the body.

thay


zcsaba77

  • Full Member
  • ***
  • Posts: 245
Re: Kapanadze Cousin - DALLY FREE ENERGY
« Reply #4842 on: July 24, 2014, 09:15:29 AM »
hi,

It's suppose to be "3825" for max 1mhz.The lowest duty cycle i have got is around 4% at 230khz.

... and at 1MHz how low was duty cycle %?

magpwr

  • Hero Member
  • *****
  • Posts: 1168
    • Youtube Channel
Re: Kapanadze Cousin - DALLY FREE ENERGY
« Reply #4843 on: July 24, 2014, 10:32:52 AM »
... and at 1MHz how low was duty cycle %?

hi zcsaba77,

The highest i have done was nearly 500khz while selecting right capacitors for the PWM I/C.There is no requirement for me to go higher because the average common mosfet and igbt would stop working after certain frequency is reached.Base on datasheet the lowest duty cycle is 3%.

My most expensive CREE mosfet in my possession cost $16USD per piece as sold in digikey is able to produce nice waveform up to around 310khz using 20volts signal at gates.

If you are just playing around with yoke and other ferrite core the 3525 or TL494 will do.

Air core would need frequency higher than 200khz base on my findings.

Hoppy

  • Hero Member
  • *****
  • Posts: 4135
Re: Kapanadze Cousin - DALLY FREE ENERGY
« Reply #4844 on: July 24, 2014, 11:14:05 AM »

Hi Hoppy,

the choke is extremely sensitive concerning this special mode.
The used 250uH choke is just on the border, when i tested with 800uH, 5.5mH and 7mH chokes, they all caused the Mazzilli to go into this special mode (@24v INPUT).

I will reposition this choke (close to the input terminals) on my circuit to see if the present location perhaps is causing the feedback/oscillations.


Thanks,  regards Itsu



Hi Itsu,

You say that the 250uH is just on the border. Have you tried reducing the inductance? I have less than 250uH.

Regards
Hoppy