PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2024-11-26, 22:20:29
News: Check out the Benches; a place for people to moderate their own thread and document their builds and data.
If you would like your own Bench, please PM an Admin.
Most Benches are visible only to members.

Pages: 1 2 3 4 5 6 7 8 9 10 11 12 [13] 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 ... 48
Author Topic: Akula0083 30 watt self running generator.  (Read 975095 times)
Sr. Member
****

Posts: 336
Well,  that's not that easily done, as it needs substantial (re)moving parts/connections on my tightly soldered print  :(
Perhaps its faster done in a simulator.

Would this feedback link to pin 15 not be more appropriate to be connected to pin 3 (feedback)?

Anyway, it will have to wait till tonight as i am at work right now.

Regards Itsu
 

Itsu,

OK, I understand.

>>Would this feedback link to pin 15 not be more appropriate to be connected to pin 3 (feedback)?
I do not know. I hope those simulator guys can give it a try.

GL.
   
Group: Guest
I like what user MenofFather says on overunity.com here:

http://www.overunity.com/14378/akula0083-30-watt-self-running-generator/msg394343/#msg394343

He points out that this R5 (and C5) in the original diagram could be an automatic duty cycle adjuster depending on the load, he says
as a reaction on this post from T1000:

T1000
MenofFather:
So a changing voltage on pin 15 via R7 could/should vary the d.c.

Any thoughts?

Regards Itsu
Hi Itsu,

If we look into the TL494 specs, we see that each amplifier has a 90dB+ gain at DC. So, for DC, the amplifier at pins 1 and 2 is operating at full gain, because of capacitor C7. The same applies to amplifier at pins 15 and 16. This explains the jump in the output, because the gain is so high.

There are some possibilities here:
1) a resistor is missing between pins 2 and 3, so that the DC gain will be lower at DC. In this case, R12+C7 will further limit the gain at AC, starting at around 270 Hz (=1/(2*PI*R12*C7) ).
2) R12 should be connected directly from pin3 to pin 2 and capacitor C7 to from pin 2 to ground.
I'm running simulations on these two situations and let you know what I found.

Cheers,

Black Bird
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 4159

Sounds good Black Bird, let me know the outcome, so i can put some changes in tonight.

Regards Itsu
   
Group: Guest
You guys are sure to nail this down soon, but I have to ask, are you sure it wouldn't make sense to just utilize a pulse generator like this one to drive the looped portion of the circuit:

https://www.bkprecision.com/downloads/datasheets/4030_datasheet.pdf

You would have full control of frequency, duty cycle and could probably get the coil to respond once it's wound correctly.

I'm probably off my rocker.  Just seems like this is becoming a hair more difficult than it needs to be.  I've always tried to take what I know works and use it to figure out the things I don't know.  If we can get something dialed-in, we can always go back and add the components that mimic the pulse generator.

Please take no offense.  I'm just curious if I'm the only one that has a similar idea.
   
Group: Elite
Hero Member
******

Posts: 3537
It's turtles all the way down
I must agree with Matt, using bench equipment to nail down the operating parameters of the coil seems a very flexible approach for those who have the equipment. Understanding the "magic of the coil" if any exists seems the best route, lest we become a "cargo cult".

Both routes are fine, but understanding is key.


---------------------------
"Secrecy, secret societies and secret groups have always been repugnant to a free and open society"......John F Kennedy
   

Group: Moderator
Hero Member
*****

Posts: 2765


Buy me a cigar
Dear ION.

Welcome aboard !!  :)

Your extensive knowledge and expertise will be a great help to this thread.

Best wishes, Grum.


---------------------------
Nanny state ? Left at the gate !! :)
   
Hero Member
*****

Posts: 520
@all

I asked Wesley if he can talk or email Akula for the winds.

While you guys ponder the circuit (I told you guys not to make it yet....but boys will be boys - hehehe), I am preparing a Standard Procedure for OU Coil Winding.

Point is forget what Akula says about 15/45, it means nothing but a range of turns, so you don't make a coil with 100/400. The 15/45 is a 1:3 ratio. The only real fixed component is the E core. The primary and secondary must be wound as per the characteristics of both the wires used and the number of turns required to produce not more and not less of an impulse. My idea is if everyone followed a set procedure to make the primary and secondary winding, your coil will be well balanced regardless of the turns or core type since it will match your core limitations.

15 turns primary. So what if with only 11 turns you get the same or better output at 36% less impress cost plus 36% less mosfet killing flyback plus better core to wind sensitivity. What if 15 turns is saturating your core at only a half impulse and with that other half excess being over the core limit seizes the reactivity of the core. Some are saying their cores have cracked already. Wondering why????

I think it's about time that we started to work smarter on both sides of the pond. With such an OU based standard, our Russian friends or any builder can simply say "This transformer was wound using XXXXX standard except step 3 where I started at 2 turns". Then provide some core, wire specs and possible resonance frequencies as a guide to others. This way anyone in the world will be able to come very close to the same transformer build regardless of number of turns since the main builder used the same standard to wind his coils with which he obtained the results that others are looking to now replicate.

Many here have been around the block a few hundred times already and can attest to the fact that although all come on board with the best intentions, that is not enough. Good intentions are great but making sure your efforts are not unknowingly thwarted by actions based on a lack of foresight is worth gold to any such enterprise. What we want is not winding up again at page 345 by playing musical chairs.
 
One observation like;

1) @Grumage, your coil wire type seems to have its plastic outer coating to be very soft and in a transformer dishing out 30 watts, it will risk shorting in the center winds possible then making your drive circuit overheat and blow. So make sure the wire you are using has a good coating on it that will withstand some heat.

About the standards, if we can have a good 4-5 standards that cover a good range of building methods, then the actual information resulting from these should help everyone stay on the same page.

I am presently working on the LTPU/FTPU and will have some great discoveries to announce soon enough. But I will place the procedure in this thread and at my Understanding Overunity thread since this applies to everyone.

All the best.

wattsup



---------------------------
   

Group: Moderator
Hero Member
*****

Posts: 2765


Buy me a cigar
Dear Wattsup.

I have the greatest respect for your input and agree with your sentiments. YES we could really do with the winding schedule it would make our quest so much easier !!

I also agree with your sentiments as regards my transformer windings. However, for the "Looking for anomalies" phase they are probably adequate to perform that function.

I do really feel  that the preliminary work being done by Itsu and the guys will hold us in great stead when the time comes to hook up  the two parts together.

In the mean time I would just simply like to say that you're all doing a great job, many thanks.  O0

Cheers Grum.


---------------------------
Nanny state ? Left at the gate !! :)
   
Group: Guest
@ all

If and when we receive some firm information on the coil winding, I shall initially breadboard the TL494, Vreg and mosfet circuitry. I'm tempted to put everything on strip-board, as I'm experienced in using this medium. However, its not an advisable method for novices in electronics and as Wattsup suggests, its sensible that we all use the same build method. If we were to all initially use a breadboard approach, then a Janet & John row and column listing could, if deemed necessary, be drawn up for the necessary component placing and inter-connections. Those that will be receiving a PCB (kindly free issued by Groundloop and Peterae) may be best advised to populate this only when some initial groundwork has been carried out on the full build, to verify that there is either no necessity to deviate from Akula's design, or that any necessary variation(s) can be accomodated easily on the PCB. Just my thoughts.

Hoppy
   
Group: Guest
Grum, Peterae, u got my PM?
regards
   
Group: Guest

Hello Gentlemen,

In the Akula's circuit, this aluminium foil wraped (but not short circuited) around the trafo is strange.
Is it not? So, why not testing it? Just 5 pictures. No comments:











Le bonjour vous sied,
Jean
   

Group: Professor
Hero Member
*****

Posts: 3499
But the 5K potmeter does not seem to change the pulse width, although the voltage on pin 1 changes from 1V to 4V.
Oh, C'mon!  You've connected the TL494 to adjust its duty cycle with a pot, before.  See your video and the schematic here.

I seem to have a problem with the 12V voltage regulator as sometimes the voltage output on it drops from 12 to 10V
Of course you do, the minimum difference in voltage between the input and output of the linear voltage regulator (VDO) should be respected as stated in its datasheet.


This circuit has a high yawn factor.  The mystery is in the E-core inductor/transformer.
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 4159
Oh, C'mon!  You've connected the TL494 before to adjust its duty cycle with a pot.  See your video and the schematic here.

Correct, but that was with a different diagram.  We are now trying to get this specific AKULA 30W diagram to work without changing to much, else it would not be called a replication.

Quote
Of course you do, the minimum difference in voltage between the input and output of the linear voltage regulator (VDO) should be respected as stated in its datasheet.


This circuit has a high yawn factor.  The mystery is in the E-core inductor/transformer.

Yep, you are right, when using my bench PS, the 12V regulator output sags from 12V to 10V when the input goes below 14V.

Regards Itsu
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 4159
Hi Itsu,

If we look into the TL494 specs, we see that each amplifier has a 90dB+ gain at DC. So, for DC, the amplifier at pins 1 and 2 is operating at full gain, because of capacitor C7. The same applies to amplifier at pins 15 and 16. This explains the jump in the output, because the gain is so high.

There are some possibilities here:
1) a resistor is missing between pins 2 and 3, so that the DC gain will be lower at DC. In this case, R12+C7 will further limit the gain at AC, starting at around 270 Hz (=1/(2*PI*R12*C7) ).
2) R12 should be connected directly from pin3 to pin 2 and capacitor C7 to from pin 2 to ground.
I'm running simulations on these two situations and let you know what I found.

Cheers,

Black Bird

Black Bird,  i tried the first option and put a 20K pot-meter between pins 2 and 3.
Now i can manipulate the duty cycle between 56% and 100% by varying both this 20K pot and the Original 5K pot.
It never will go below 56%.

Regards Itsu
   
Group: Guest
Correct, but that was with a different diagram.  We are now trying to get this specific AKULA 30W diagram to work without changing to much, else it would not be called a replication.

Yep, you are right, when using my bench PS, the 12V regulator output sags from 12V to 10V when the input goes below 14V.

Regards Itsu

So far, a couple of quite serious 'yawns' have been identified with Akula's circuit. However, you are right that Akula's circuit should be replicated as closely as possible and this means that full coil details are essential as its here as Verpies suggests that any magic is bottled-up.

Regards
Hoppy
   

Group: Professor
Hero Member
*****

Posts: 3499
Correct, but that was with a different diagram.  
Which diagram is the most up to date??
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 4159

Oh there are many,  too many.
I am using the diagram (don't know if its the latest) as can be found in this thread on page 9 in post #204 from Groundloop:

http://www.overunityresearch.com/index.php?topic=2358.msg36967#msg36967




We do work also on this magical transformer now and then, at least i do, and i know Grumage did some testing and he made his squeel, but i did not find any abnormalities (no static electricity either).

See short video here:  https://www.youtube.com/watch?v=1mPy2rjDkhE&feature=youtu.be
See also a screenshot of the signal across the bulb while squeeling on 2.9KHz

Regards Itsu
   
Sr. Member
****

Posts: 336
Oh there are many,  too many.
I am using the diagram (don't know if its the latest) as can be found in this thread on page 9 in post #204 from Groundloop:

http://www.overunityresearch.com/index.php?topic=2358.msg36967#msg36967




We do work also on this magical transformer now and then, at least i do, and i know Grumage did some testing and he made his squeel, but i did not find any abnormalities (no static electricity either).

See short video here:  https://www.youtube.com/watch?v=1mPy2rjDkhE&feature=youtu.be
See also a screenshot of the signal across the bulb while squeeling on 2.9KHz

Regards Itsu

Itsu,

I couldn't hear the 10KHz either. That says something about our age. LOL

@All,

I have updated the "Prototype Board" design here: http://www.overunityresearch.com/index.php?topic=2358.250 / Reply #270
This will be my final design. This board can probably be used to other projects also. I have also split some tracks so that it
it is possible to solder in 555 or MOSFET drivers (8 pin DIL) types. It is also possible to plug a 555 into the top part of the
socket because all 8 pins here is isolated from the rest of the tracks. The board is single sided so all copper tracks is on
the underside of the board. The prototype tracks are all marked on the top side of the board. It is my hope that this board will
inspire people to do hobby research on switching power circuits. Even if we do not find any OU, there will be hours of fun hobby
electronic work. But who knows, maybe we can find a circuit that define some of the man made laws regarding magnetic fields etc.

I will give away 8 boards of this design for free. I can snail mail everywhere in the world. When I said free, I mean free, so I
will also pay for the postal charges. The boards will arrive in the week before Easter. The first 8 people that PM me a snail mail
address and country location will have a free board. If you want to stay anonymous then rent a P.O. box or something, but I
will need a real postal address. Have fun.............. :-)

Added: 8 PCB claimed, 0 boards left!

GL.
« Last Edit: 2014-03-29, 16:50:46 by Groundloop »
   
Sr. Member
****

Posts: 375
We do work also on this magical transformer now and then, at least i do, and i know Grumage did some testing and he made his squeel, but i did not find any abnormalities (no static electricity either).

You won't see any anomalies in this test case because in way it is made and connected. Three things should be there from my point of view:
1) The load must be LEDs, no any other resistive/inductive load;
2) The transformer core must have gap in all 3 sections, not only in 2;
3) The transformer is supposed to be connected like choke as per schematics and both windings should end up in series with capacitor between them when BEMF pulse goes on. Your LEDs should be attached like there too.

Hopefully that helps.

Cheers!
   
Group: Guest
2) The transformer core must have gap in all 3 sections, not only in 2;

So that means, the core must be an E-Core with the center gapped as well as the two outer ends gapped.  Is that correct?  Would that mean to use a non-gapped core with some sort of spacer between all three surfaces?

Thanks much for this info T.
   
Sr. Member
****

Posts: 336
Well just the small section of the circuit.. but if no one likes the idea of this that is fine.  I usually do test whole circuit on breadboard, but thought this was a compromise until we for sure know what is the best schematic in that area of the circuit.  I just thinking if you're going to spend money on PCB, then at least be able to customize the section in question.  The idea wasn't for me, but for those who like the PCB method.

4Tesla,

OK, now that you got your whish come true, do you want a board?

GL.
   
Group: Guest
Black Bird,  i tried the first option and put a 20K pot-meter between pins 2 and 3.
Now i can manipulate the duty cycle between 56% and 100% by varying both this 20K pot and the Original 5K pot.
It never will go below 56%.

Regards Itsu
Hi Itsu,
Have you also added the 4.7k resistor from pin 2 to ground?
As I said before, I was having problems with the simulation until I also grounded pin 15. When I did this, I could see the circuit working and I found a very interesting fact, which is that the circuits shows two frequencies: the first one is the switching frequency and another one, at much lower frequency, which is directly related to the value of the output filter capacitor C3 and the inductor at the input of the 7812 regulator. For C3=2200uF and the inductor = 100uH, this frequency is about 340Hz, and it appears in the transformer current as kind of a sawtooth. If the inductor is increased to 500uH, this frequency lowers to 115Hz, with the switching frequency unchanged.
Many of the alleged working OU circuits mention using two frequencies mixed at the output transformer, and it is interesting that this circuit apparently does the same. This triggers a thought that it might be that the "magical" effects on the transformer will only appear when it is connected to the real circuit.
Concerning the load, I ran simulations with a resistor and with LEDs and did not see any apparent difference, so far, but let's not forget this is only a simulation, not the real thing.

Cheers

Black Bird
PS: I modified the position of C3 according to Menofather information to the junction of the diode that goes to the battery and the inductor. keeping it in the original position give rise to currents of more than 60 amps, and changing the position as described keeps currents below 25 amps.
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 4159
You won't see any anomalies in this test case because in way it is made and connected. Three things should be there from my point of view:
1) The load must be LEDs, no any other resistive/inductive load;
2) The transformer core must have gap in all 3 sections, not only in 2;
3) The transformer is supposed to be connected like choke as per schematics and both windings should end up in series with capacitor between them when BEMF pulse goes on. Your LEDs should be attached like there too.

Hopefully that helps.

Cheers!


thanks for the info T1000,
it was just a quick setup to show what this transformer can do when hard driven by a 12V switching MOSFET.
I know others are also experimenting with the x-former and wanted to encourage them to show what they are doing.

Good to know that only LEDs must be the load, i have ordered some.
As the two outer cores of my e-core had a taped gap, this automatically means the inner core had a similar air gap.

Thanks,  regards Itsu
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 4159
Hi Itsu,
Have you also added the 4.7k resistor from pin 2 to ground?
As I said before, I was having problems with the simulation until I also grounded pin 15. When I did this, I could see the circuit working and I found a very interesting fact, which is that the circuits shows two frequencies: the first one is the switching frequency and another one, at much lower frequency, which is directly related to the value of the output filter capacitor C3 and the inductor at the input of the 7812 regulator. For C3=2200uF and the inductor = 100uH, this frequency is about 340Hz, and it appears in the transformer current as kind of a sawtooth. If the inductor is increased to 500uH, this frequency lowers to 115Hz, with the switching frequency unchanged.
Many of the alleged working OU circuits mention using two frequencies mixed at the output transformer, and it is interesting that this circuit apparently does the same. This triggers a thought that it might be that the "magical" effects on the transformer will only appear when it is connected to the real circuit.
Concerning the load, I ran simulations with a resistor and with LEDs and did not see any apparent difference, so far, but let's not forget this is only a simulation, not the real thing.

Cheers

Black Bird
PS: I modified the position of C3 according to Menofather information to the junction of the diode that goes to the battery and the inductor. keeping it in the original position give rise to currents of more than 60 amps, and changing the position as described keeps currents below 25 amps.

Hi Black Bird,  no i did not add the 4.7k resistor from pin 2 to ground.  Where did that come from?


For the rest of your comments i can not respond to as i do not have those components installed, so i do not see any "lower frequency".
But i agree that probably we need to setup the WHOLE circuit to be able to test the proper functioning of the control section (TL494).

Thanks,  regards Itsu
   
Group: Guest
Hi Black Bird,  no i did not add the 4.7k resistor from pin 2 to ground.  Where did that come from?


For the rest of your comments i can not respond to as i do not have those components installed, so i do not see any "lower frequency".
But i agree that probably we need to setup the WHOLE circuit to be able to test the proper functioning of the control section (TL494).

Thanks,  regards Itsu
The 4.7k resistor from pin 2 to ground is to keep the voltage of pin 2  at about 2.5V, below the threshold voltage of the sawtooth that appears at pin 5.

Cheers,

Black Bird
   
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 [13] 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 ... 48
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2024-11-26, 22:20:29