PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2017-09-23, 15:35:08
News: A feature is available which provides a place all members can chat, either publicly or privately.
There is also a "Shout" feature on each page. Only available 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 ... 48
Author Topic: Akula0083 30 watt self running generator.  (Read 479730 times)
Full Member
***

Posts: 164
My question is:  Has anyone proto-typed (breadboard) the proposed drive circuit and traced the signals with a o-scope?

I have NOT had time to do so myself, so my comments are only based on experience and not signal tracing the proposed circuit.

take care, peace.
lost_bro

On the Russian forums, yes.. here, I don't think so.  I myself like to breadboard first, but everyone has their own preferences.

Russian forums:
http://realstrannik.ru
http://www.freenergy.com.ua
   
Full Member
***

Posts: 116
On the Russian forums, yes.. here, I don't think so.  I myself like to breadboard first, but everyone has there own preferences.

Good Day 4 Tesla

Yes, I have learned the hard way about prototyping first:  highly recommended....... and to each their own.

I know that it sucks to waste time and components...

take care, peace.
lost_bro
   
Full Member
***

Posts: 164
I've learned a lot by surfing the Russian forums.  If anyone here using Firefox, I highly recommend this translator add-on:

S3.Google Translator
https://addons.mozilla.org/en-US/firefox/addon/s3google-translator/

It automatically translates the pages using Google Translate.  Not perfect, but helpful.  Also, I think Chrome might have this feature built in.
   
Group: Guest
4Tesla,

With the reading you have been doing, are any of those guys revealing what component fails when the unit quits running?  We heard about the one where the core crumbled, any other symptoms you have spotted like burnt wiring, blown semiconductors...?  When the unit quits are some able to restart it with no changes?
   
Full Member
***

Posts: 164
MOSFETS/Transistors.  I think it is because for this to work the core may have to be saturated which creates a short.. from my understanding.

I don't know about restarting.. if components have to be replaced or just started again with battery.
   
Group: Guest
I have designed a PCB based on the original version of the 30W schematic...

This is excellent work GL.  It has my vote as "The Reference Design" for testing this device.  Also a very kind offer putting four of them out there for grabs, which by now all should have been claimed.

Exciting times ahead.
   
Full Member
***

Posts: 164
   

Group: Administrator
Hero Member
*****

Posts: 3519


Buy me some coffee
I'm also looking at placing an order for 20-30 pcb's and am taking orders, the pcb's are free, I'm in the UK and will post a number to someone in the US who can then distribute them, maybe Chet.

So please post here if you want one so i can get an idea how many to get made up, at the moment i don't think anyone has claimed any of Groundloops and so far no one has asked for one from me.
   
Full Member
***

Posts: 164
I'm also looking at placing an order for 20-30 pcb's and am taking orders, the pcb's are free, I'm in the UK and will post a number to someone in the US who can then distribute them, maybe Chet.

So please post here if you want one so i can get an idea how many to get made up, at the moment i don't think anyone has claimed any of Groundloops and so far no one has asked for one from me.

+1 I'm in US.  GL in UK?  I'll let locals get GL's

Thanks you!
   
Sr. Member
****

Posts: 337
Good Day 4 Tesla;

Those are very interesting schematics that you posted.  I will study them when I find time.

Hello All.

Short of sounding like a broken record, I post my thoughts on the drive portion of the 'Final' Akula TPU circuit.  

First I will say that it is very encouraging to see such enthusiasm with any project, let alone one as complicated as this one.

I am NOT trying to be critical, and only offering suggestions based on my observations and experience with the TL494 chip.

I am NOT diss-ing GroundLoop for his work and time donated to the cause................................

My question is:  Has anyone proto-typed (breadboard) the proposed drive circuit and traced the signals with an o-scope?

I have NOT had time to do so myself, so my comments are only based on experience and not signal tracing the proposed circuit.

take care, peace.
lost_bro





 

lost_bro,

You comment on my drawing is not correct. The mosfet is on (conducting current) when the C2/E2 transistor is off.
When the C2/E2 transistor is on again then the mosfet gate is discharged to ground, switching off the mosfet. The
output of the C2/E2 transistor can only sink current.

You don't need a pull down resistor at the mosfet gate because the C2/E2 transistor in the TL494 is pulling
down. The transistor never pulls up. This because R3 in the circuit drawing is the pull up. The BD139
transistor will carry most of the current to charge the mosfet gate when the TL494 transistor output
is not active, so I think it is needed.

The 1/2 frequency output is on purpose, I guess that was what Akyla0083 wanted. Also I wanted
to keep my design as close to the Akyla0083 posted design as possible.

GL.
   

Group: Moderator
Hero Member
*****

Posts: 1637


Buy me a cigar
Dear lost_bro and all.

Quote.
First I will say that it is very encouraging to see such enthusiasm with any project, let alone one as complicated as this one.
Unquote.

Sometimes it is enthusiasm that provides the catalyst for great things to happen !! The rallying cry, so to speak. I may be nearing 58 years but I have always been young at heart !! Taken the bull by the horns, jumped in with both feet, you know where I am coming from !!

Groundloops efforts, I feel, will soon be rewarded particularly with the team we have now assembled here at OUR. Away from those people that try to dissuade and dampen the enthusiasm that is really needed IMO to see something through.

So  "Cry 'Havoc!', and let slip the dogs of war".

Lets build a self runner. 3 days will be fine for me !!  ;D

Cheers Grum.


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

Posts: 337
I'm also looking at placing an order for 20-30 pcb's and am taking orders, the pcb's are free, I'm in the UK and will post a number to someone in the US who can then distribute them, maybe Chet.

So please post here if you want one so i can get an idea how many to get made up, at the moment i don't think anyone has claimed any of Groundloops and so far no one has asked for one from me.

None has claimed any PCBs from me yet, so I have still 4 to give away.

GL.
   
Full Member
***

Posts: 164
@GL

Is it possible to add jumpers so we can configure either scheme?  Transistor or MOSFET.
   
Sr. Member
****

Posts: 337
@GL

Is it possible to add jumpers so we can configure either scheme?  Transistor or MOSFET.

4Tesla,

No change on my boards because I have already ordered them.

But anything is possible. If Peter has not ordered yet and want changes done
to the board then OK by me.

GL.
   
Full Member
***

Posts: 116
lost_bro,

You comment on my drawing is not correct. The mosfet is on (conducting current) when the C2/E2 transistor is off.
When the C2/E2 transistor is on again then the mosfet gate is discharged to ground, switching off the mosfet. The
output of the C2/E2 transistor can only sink current.

You don't need a pull down resistor at the mosfet gate because the C2/E2 transistor in the TL494 is pulling
down. The transistor never pulls up. This because R3 in the circuit drawing is the pull up. The BD139
transistor will carry most of the current to charge the mosfet gate when the TL494 transistor output
is not active, so I think it is needed.

The 1/2 frequency output is on purpose, I guess that was what Akyla0083 wanted. Also I wanted
to keep my design as close to the Akyla0083 posted design as possible.

GL.

Hello Ground Loop

Thanks for the comments:
Actually my comments are CORRECT:   Please reference attached copy of original posted pic.

 Ground Loop: "The mosfet is on (conducting current) when the C2/E2 transistor is off. When the C2/E2 transistor is on again then the mosfet gate is discharged to ground, switching off the mosfet.
 The output of the C2/E2 transistor can only sink current."

Yes, I agree as stated in the pic. that I posted:

"MOSFET is ON when TL494  Output  LOW"    and  "TL494 is SINKING CURRENT thru RED trace"     ( Red trace is drawn thru C2 & E2)     so  We are in agreement

----You can place a SPST between Pin #13 & Pin #14 to choose at random the output mode : push pull or single pulse = 1/2 or complete Frequency of PWM.

Ground Loop: "You don't need a pull down resistor at the mosfet gate because the C2/E2 transistor in the TL494 is pulling
down. The transistor never pulls up."

I am NOT discussing the action of the BD139, I am referrencing a possibility of a High Noise Floor generated by any possible cause including but not limited to:  parasitic inductance of traces on PCB or feedback from the L1 or L2. Depending on the Gate threshold values of the MOSFET in question it is ALL the more important.

Please see excerpt below from Tamids Blog: http://tahmidmc.blogspot.mx/2012/10/magic-of-knowledge.html

Importance of the gate-to-source resistor:  

"It prevents accidental turn on of the MOSFET by external noise usually at startup when the gate is floating. The MOSFET may sometimes turn on with a floating gate because of the internal drain to gate "Miller" capacitance. A gate to source resistor acts as a pull-down to ensure a low level for the MOSFET. I have had MOSFETs blowing up in high voltage circuits, without the resistor in place. In most of the commercial power supplies / inverters I have seen, there is a 1k resistor used."


Sorry if I offended you, not my intention....

take care, peace.
lost_bro
   
Sr. Member
****

Posts: 337
Hello Ground Loop

Thanks for the comments:
Actually my comments are CORRECT:

 Ground Loop: "The mosfet is on (conducting current) when the C2/E2 transistor is off. When the C2/E2 transistor is on again then the mosfet gate is discharged to ground, switching off the mosfet. The
output of the C2/E2 transistor can only sink current."

Yes, I agree as stated in the pic. that I posted:

"MOSFET is ON when TL494  Output  LOW"    and  "TL494 is SINKING CURRENT thru RED trace"     ( Red trace is drawn thru C2 & E2)     so  We are in agreement

----You can place a SPST between Pin #13 & Pin #14 to choose at random the output mode : push pull or single pulse = 1/2 or complete Frequency of PWM.

Ground Loop: "You don't need a pull down resistor at the mosfet gate because the C2/E2 transistor in the TL494 is pulling
down. The transistor never pulls up."

I am NOT discussing the action of the BD139, I am referrencing a possibility of a High Noise Floor generated by any possible cause including but not limited to:  parasitic inductance of traces on PCB or feedback from the L1 or L2. Depending on the Gate threshold values of the MOSFET in question it is ALL the more important.

Please see excerpt below from Tamids Blog: http://tahmidmc.blogspot.mx/2012/10/magic-of-knowledge.html

Importance of the gate-to-source resistor:  

"It prevents accidental turn on of the MOSFET by external noise usually at startup when the gate is floating. The MOSFET may sometimes turn on with a floating gate because of the internal drain to gate "Miller" capacitance. A gate to source resistor acts as a pull-down to ensure a low level for the MOSFET. I have had MOSFETs blowing up in high voltage circuits, without the resistor in place. In most of the commercial power supplies / inverters I have seen, there is a 1k resistor used."


Sorry if I offended you, not my intention....

take care, peace.
lost_bro

lost_bro,

I'm not offended at all. :-) Use to be fisherman in my young days LOL.

I think we are talking about the same issues. I also think the 1K resistor and 100nF (C1 and R4) snubber
circuit will take care of any transients. Also the anti series zener diodes will prevent some of the "spikes".

Lets build it first and test it. It is possible to change things around by adding components and cutting in
the PCB afterwards.  One thing I think needs changing is the 1N4148 diode. We will probably need a Schottky
diode here.

GL.
   
Full Member
***

Posts: 116
Hello Ground Loop:

Yes, you are correct.... until it is built it is ALL TALK and NO WALK ......................

When I have time I will prototype it also.....

I am inclined to believe along the lines suggested by Matt the other day:   That the use of a Discrete MOSFET Driver chip would actually be the best solution if a super fast Rise and Fall time is required for the pulse signal.

I have heard talk on the Russian Forums of 'Nano Pulsing'......  If that is the case then there are other 'PWM' chips and discrete MOSFET driver options that will make the 'Nano Pulsing' a reality.


Take care, peace
lost_bro
   
Sr. Member
****

Posts: 337
Hello Ground Loop:

Yes, you are correct.... until it is built it is ALL TALK and NO WALK ......................

When I have time I will prototype it also.....

I am inclined to believe along the lines suggested by Matt the other day:   That the use of a Discrete MOSFET Driver chip would actually be the best solution if a super fast Rise and Fall time is required for the pulse signal.

I have heard talk on the Russian Forums of 'Nano Pulsing'......  If that is the case then there are other 'PWM' chips and discrete MOSFET driver options that will make the 'Nano Pulsing' a reality.


Take care, peace
lost_bro

lost_bro,

I have read about that too. Maybe the forward diode is for that purpose?
So we probably need a more powerful diode than the 1N4148.

BTW: I did check the data sheet for the TL494 and the RC oscillator values
that Akyla0083 did use puts the frequency off the factory scale for low Ohm
values. For highest Ohm value the frequency is just inside the window.

GL.

   
Full Member
***

Posts: 164
Hi all,

Not good with graphics, but this is what I was thinking for PCB.  If others like the idea... please let us know so GL could draw up.. if he is cool with the idea.

Thanks!
   
Sr. Member
****

Posts: 337
Hi all,

Not good with graphics, but this is what I was thinking for PCB.  If others like the idea... please let us know so GL could draw up.. if he is cool with the idea.

Thanks!

4Tesla,

It is doable. It depend on what Peter says? If he want me to redesign
the PCB for such a breadboard layout then I will do it. But no time until
next weekend for this.

Is there any other people here that are familiar with the EAGLE CAD design program?

GL.
   
Group: Guest
Is there any other people here that are familiar with the EAGLE CAD design program?

I can muddle through it GL, though I'm more comfortable with DipTrace.
   
Full Member
***

Posts: 116
lost_bro,

I have read about that too. Maybe the forward diode is for that purpose?
So we probably need a more powerful diode than the 1N4148.

BTW: I did check the data sheet for the TL494 and the RC oscillator values
that Akyla0083 did use puts the frequency off the factory scale for low Ohm
values. For highest Ohm value the frequency is just inside the window.

GL.



Hello Groundloop

Yes, the spec sheet list 300Khz as the MAX freq. for the TL494......  
So. Akula wanted this beast to run in the upper spectrum of the chips capacity.


About the Nano Pulsing:   First thing I would do is the Bypass the integral Body diode in the MOSFET using an External Super Fast Diode.....   Place series Schottky at Drain ( cathode to Drain on MOSFET) and then run UF or SiC Schottky diode( in Paralled with MOSFET) from upper side of schottky (which is series with the Drain) to the Source of the MOSFET.  Anode of parallel SiC Schottky connects to Source of MOSFET.
  
This effectively bypasses the internal body diode (Now parallel external Schottky or UF will respond in place of body diode)  and will clean up the  recovery time for the MOSFET by maybe an order of magnitude .

This will also have the effect to REDUCE the heat generated by the MOSFET  because now the EXTERNAL UF or SiC Schottky is bypassing and not the internal body diode......... You can even heatsink the SiC Schottky separately for better dissipation.

This is a trick used in SS Tesla coil driver circuits and works really well.

Take care, peace.
lost_bro
   
Full Member
***

Posts: 164
Hello Groundloop

Yes, the spec sheet list 300Khz as the MAX freq. for the TL494......  
So. Akula wanted this beast to run in the upper spectrum of the chips capacity.


About the Nano Pulsing:   First thing I would do is the Bypass the integral Body diode in the MOSFET using an External Super Fast Diode.....   Place series Schottky at Drain ( cathode to Drain on MOSFET) and then run UF or SiC Schottky diode( in Paralled with MOSFET) from upper side of schottky (which is series with the Drain) to the Source of the MOSFET.  Anode of parallel SiC Schottky connects to Source of MOSFET.
  
This effectively bypasses the internal body diode (Now parallel external Schottky or UF will respond in place of body diode)  and will clean up the  recovery time for the MOSFET by maybe an order of magnitude .

This will also have the effect to REDUCE the heat generated by the MOSFET  because now the EXTERNAL UF or SiC Schottky is bypassing and not the internal body diode......... You can even heatsink the SiC Schottky separately for better dissipation.

This is a trick used in SS Tesla coil driver circuits and works really well.

Take care, peace.
lost_bro


Sounds good!  Can you draw out schematic?  This is why I think a little bread on the PCB would be cool, because so many mods in that area of the circuit.
   
Full Member
***

Posts: 116
Sounds good!  Can you draw out schematic?  This is why I think a little bread on the PCB would be cool, because so many mods in that area of the circuit.

Hello 4Tesla

As requested;

see below.

take care, peace
lost_bro
   
Group: Guest

Hi People,

Thanks to everybody. Thanks to Lost_Bro, for his comments.
For the moment, I'm interested in very basic "theoretical"  stuffs.
I just made a new quick experiment:



--------------------------
It works = you can get voltage out of an air core coil wrapped with aluminium foil (as to build a kinda "capacitor")
even if this coil is just 'excited' by this fictitious displacement current. :)

You have to find the right ("resonant") freq.
Grounding/Eathing one particular end of the coil strongly increase the output voltage and change the "resonant" freq.

First results. No load. No "OU" claimed.
The "capacitor" : From 260 pF to 410 pF according to 2 different f' cheap meters.
The coil: Air core coil. From 19 to 22 Milli Henrie according to 2 different cheap bl' meters.

The Freq. Gen is given for: Max freq 3 MHZ. (Mine can goes a little  bit further (3.15)).
Max output 10 volts ???

Actually, according to my scope:
At 3 MZH the Freq. Gen. output voltage (with not load) is only about 5 volts PK-to-PK.
At freqs less than about 500 KHZ  this voltage is about 20 Volts PK-to-PK.
---------
Sine wave as CCT input. CCT Output volts-wise only. With no load.
Not grounded/'earthed'
Feq. Gen. output (= CCT input)about 20 volts PK-to-PK.
CCT output about 2 volts PK-to-PK at about 150 Khz.

Grounded/'earthed':
Feq. Gen. output (= CCT input) about 5 volts PK-to-PK.
CCT output about 20 volts PK-to-PK at about 3,07 Mhz.
--------------------
Something seems to be going on, here. No.?

These are just first results. The numbers have to be confirmed. The circuit (CCT) have to be loaded.
----------
Very very easy to reproduce. This experiment is for the slackers like me. :P
 
Take any off-the-shelf enameled copper wire spool.
It has to have his 2 ends available, of course. :)
Wrap it with some kitchen aluminium foil around it.
You have made a "capacitor".
Use a crocodile clip for connecting one "plate" of you cap.
Follow the picture and utilize your favorite Freq. Gen. and Oscilloscope.

Cheers,
Jean
   
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 ... 48
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2017-09-23, 15:35:08