Push pull generator.

Hi Guys,

I would like to buy a ready-made control for a push-pull circuit, so that I can easily control 2 mosfets or IGBTs directly.

I've already seen a few controls on ebay where you can adjust the pulse width and frequency via buttons and see on a digital display which settings have been made.
As it looks but are the pure 1 channel PWM controls, I have unfortunately received no response from the sellers.

Why and what I need it for:

I would like to bring the 3-phase transformer, which serves as a ferronic resonance transformer, into resonance via the drive winding via a push-pull signal in order to find the sweetspot for the lowest operating energy.
Since the transformer has a bifilar winding, it can be operated in simple push-pull operation.

Under rectangle generator, I found only that, but I do not think that this is the right one.

https://www.ebay.de/itm/Rechteckgenerator-1-Hz-150-kHz-einstellbares-Tastverhaltnis-0-100/223003507660?hash=item33ec0b7bcc:g:MPQAAOSwQW1a0kQt

As far as the grounding of the FR-transformer is concerned, as much talk about grounding, I have to say if I operate the transformer at 1250V URMS in resonance and the grounding wire to the core hold small sparks, I thought only that would be short circuits between the isolated laminated core but with a simple screwdriver you can not replicate that. The oscilloscope also records the energy peaks, but the earthing cable only ever stays in contact for a short time. Nothing happens for just a brief touch. A small blue spark jumps over, unfortunately not visible to the video camera.

regards

Sven

HI Sven, 
 This is difficult to find, I have also looked.   Anyhow if you had a push pull fron a timer chip it would be an uneven duty cycle.  The best way it to use a complimentary pair of transistors-NPN-PNP gated from the timer.
 

 I wish the translator worked better because I don't understand the grounding if this is showing an advantage or not?  The advantage I have found with the grounding is best with sharp flyback pulses.  Also I have found an advantage to a small HV film capacitor of around 1nf.

Sven, have you looked at the UCC21222? 

 Ron
 

 http://www.ti.com/lit/ds/symlink/ucc21222-q1.pdf http://www.ti.com/lit/ds/symlink/ucc21222-q1.pdf

Hi Guys,

since there is no response to this topic will write again.

Unfortunately, I have not found any suitable finished control boards and will now try again with a sg3525 controller to build a pulse width modulation in Pushpull principle.

I have already set up the circuit via a plug-in board with 2 potentiometers. from 5 to 40% pulse width and a frequency of 35 to 450 hertz. I think you can cover everything with this area, I think.

I want to build the controller modular so driver with optocoupler TLP250 for each mosfet, I do not want to use any IGBTs anymore as they are more sensitive.

So driver board and control board separately. At the low frequency, the interference should be limited by longer lines.

Until now, I have always broken the mosfets with the drivers, I want to design the driver board so that this does not happen so easily.

On Thamids blog, the TLP250 is described as used.

Best regards

Sven

Hi Sven,

 "So driver board and control board separately. At the low frequency, the interference should be limited by longer lines."
Long lines give more interference than short lines. The longer lines can have more area to be swept by ambient fields.

"Until now, I have always broken the mosfets with the drivers, I want to design the driver board so that this does not happen so easily."
Then some protection circuit is needed like in commercial products. The SG3525 has a shutdown input that can be used for shut down when an error is encountered. A current monitoring circuit and an over voltage monitoring circuit and perhaps a temperature sensing circuit could be used to shut down the MOSFETs before being destroyed.


As transformers have symmetrical Vs [Volt x second] product the input signal has to be symmetric if not used as just an isolated inductor. This is normally done by using a HF carrier wave that is PWM modulated with a sine wave. The carrier wave has to be perhaps of 10 to 20 times greater frequency than that of the modulation signal. If it is to be used as an inductor the PWM signal will work as step up or step down of the voltage or current.

Regards
Ole 

---In [email protected], <s.friedrich@...> wrote :

 Hi Guys,

since there is no response to this topic will write again.

Unfortunately, I have not found any suitable finished control boards and will now try again with a sg3525 controller to build a pulse width modulation in Pushpull principle.

I have already set up the circuit via a plug-in board with 2 potentiometers. from 5 to 40% pulse width and a frequency of 35 to 450 hertz. I think you can cover everything with this area, I think.

I want to build the controller modular so driver with optocoupler TLP250 for each mosfet, I do not want to use any IGBTs anymore as they are more sensitive.

So driver board and control board separately. At the low frequency, the interference should be limited by longer lines.

Until now, I have always broken the mosfets with the drivers, I want to design the driver board so that this does not happen so easily.

On Thamids blog, the TLP250 is described as used.

Best regards

Sven

Hello everyone,

I have continued to work on the pushpull converter and now have a frequency range of 37Hz to over 400Hz which is certainly sufficient for my heavy iron core transformer. Pulse width is adjustable from 780μs to 6ms at 37Hz. I think a lot can be experimented in this area.

Since I did not want to repeat the same mistakes as at that time, I concentrated strongly on the driver control and had to say if I increase the supply voltage for the mosfets at some point a point comes where there are dropouts after a long search, I took out the gate resistance at some point and the error with incandescent lamps was gone, now I am with 2 transformers as inductance testing and from now I increase the pulse width and a channel slowly starts to flicker, with lower voltage it is better again and the same current at higher frequency, it is also worse.

I have also hung ferrite beads on the gate lines and had a super clean signal but with very small pulse width, the signal collapsed, the ferrites seem to act as gate resistors, unfortunately there are no small pulse widths to produce

I soon have the feeling that the low voltage supply does not create or am I wrong.

Mosfets are IXFN44N100P

The mosfets are very hot despite the large heat sink at just 4 amps although they are designed for 37 amps per Mosfet I find that very strange.

Best regards

Sven

Sven,

Determine if the mosfets are hot or the heat sinks are hotter. 
Sometimes with these strange energies the heat sink is too hot to touch
and the fet can be touched no problem.  Maybe you have infrared
thermometer and can test each temperature easily.

Hi Sven
 Try a 10K 1/4W resistor from gate to source, not the 1K you have....
 I don't know about that small cap across there too, I would think that might screw things up and do the opposite of what the pull-down resistor does!.
 

 Also what s the voltage-feed to your mosfet gate? (reading at the gate with the resistor and the ferrite and everything you have connected to protect the gate)
 

 Maybe your voltage feeding the mosfet is too low, and it will switch but not turn off qucily because of too low of voltage at gate....the mosfets I use have maximum 18V or they blow, and 15V is supposed to be optimum for very quick chop to the turn off....while 9V or 12V "will work" but not most optimum...
 Kone

Hi Sven
 Use thin magnet wire wrapped inside tiny ferrite tubes  just a few turns ,maybe 4 turns....use very thin magnet wire around 28 or 30 GA  thickness (thin)...then connect this wire to mosfet gate.
 Also they say to include in series a low ohm resistor too, such as 5 ohm 1/4W size to help dampen any ringing...the ferrite squelches the HF stuff, the resistor the LF stuff is what I understand.
 Maybe your mosfets do not turn off fast enough, this would make for excess heat that is for sure!....are you using a driver chip to the mosfets??
 Try 4421 or 4422 types...
 Also if no drivers feeding the gates, you better for sure have a 10K "pull sown" resistor stretching across mosfet  from gate to source on the mosfet....I like to include the 10K resistor even if using a 4422 or 4421 driver chip just to be double safe....
 Kone
 ...

Empty body

Empty body

SvenI'm not sure what we are encountering, but for diodes, I try to keep the ratings above 1000 volts, even though I may be only actually measuring 100-150 volts.This is true of Joule thief, and Rene chargers too.Cheers Warren

Sent from Yahoo Mail on Android 
 
  On Thu, 5 Jul 2018 at 11:43 AM, [email protected] [EVGRAY]<[email protected]> wrote:       

 Hi Mick, very interesting your attempt, I really just want to build an experimental control for my transverter which actually only has to drive two mosfets to work in push-pull mode, of course fully adjustable,

I just do not get it up, I raise the drain voltage of the Mosfets to over 150 volts a short time out now, of course, a worst case a mosfet is scrap despite 6 amp ampere fuse.

Always at a higher voltage strange but all Lowside switched, I have now also another power supply installed with more Ampereleistung but even voltage 15 volts driving voltage but no improvement.

15Volt is that too little?

I can not find anything on the Internet on this topic or disturbs my scope but even without this effect I have already tested.

Maybe someone has an idea.

I'll do a driver circuit diagram of my circuit.

regards

SvenGesendet: Donnerstag, 05. Juli 2018 um 17:15 Uhr
Von: "Mick [email protected] [EVGRAY]" <[email protected]>
An: [email protected]
Betreff: Re: [EVGRAY] Re: Push pull generator. 
 

Sven,

Determine if the mosfets are hot or the heat sinks are hotter.  Sometimes with these strange energies the heat sink is too hot to touch and the fet can be touched no problem.  Maybe you have infrared thermometer and can test each temperature easily.

Hi Sven,

Don't put ferrite beads on the gate circuit as they add inductance. Try to avoid inductance by making the distance between the gate driver and the gate as small as possible an minimize any area between the wires that act like a coil winding adding inductance. The gate-source is a capacitor. Adding inductance makes it an LC-tank which makes oscillations when storing energy. The oscillations are amplified at the drain. Also the inductive kickback from any inductance can go above the gate-source breakdown voltage which degrades and destroys the MOSFET sooner or later. By adding extra capacitance in parallel to the gate-source capacitance slows down the MOSFET as the drain signal follows the gate signal. The series resistor at the gate is for absorbing the energy of the formed LC-tank to prevent oscillations of the gate and thus also oscillations on the drain.

Opto couplers are slow devices. The TLP250 has a delay of 0.5 us and a max frequency of 25 kHz.

A fuse normally isn't fast enough to protect semiconductors from destruction. Use an electronic protection that removes the gate signal to shut off the device. For over voltage protection use some kind of snubber active or passive. A crowbar protection made by using a thyristor (SCR) that shorts out the voltage supply rails and blows a fuse is an effective way to prevent over voltages.

'Fundamentals of MOSFET and IGBT Gate Driver Circuits:' (http://www.ti.com/lit/ml/slua618/slua618.pdf http://www.ti.com/lit/ml/slua618/slua618.pdf).
Just follow those rules to make a good gate driver. If not doing that all kinds of trouble may arise.

If the impedance of the capacitors are less than that of B2 the current mainly takes the path shown. If R2 is omitted this may cause ringing when charging the gate capacitance. R2 has to absorb the energy to prevent oscillations in the tank formed by the inductance of the two loops and the capacitance of the gate-source capacitor. For fast response R2 is chosen for critical damping which means no ringing.

Regards
Ole
 

---In [email protected], <s.friedrich@...> wrote :

 Hi Douglas,
 
 I use an opto driver TLP250 very simple.
 
 At the mosfet I use 1k resistors from gate to source.
 
 On TLP250 directly 1μF capacitor MKP and parallel 68μF Elko.
 
 I removed the 10 Ohm gate resistor because I could not realize short pulse widths below 1 ms with it, ferrite beads work just as clean scope image but no short pulse widths possible.
 
 In addition, a 10nF capacitor between gate and source connected to relieve the gate, thereby minimal improvement.
 
 I have to drive at least 4 amps at 300 volts.
 
 The parts are all oversized.
 
 I can shorten the gate lines a bit but if that's the solution I doubt it.
 
 regards
 
 Sven   Gesendet: Donnerstag, 05. Juli 2018 um 17:03 Uhr
 Von: "konehead@... [EVGRAY]" <[email protected]>
 An: [email protected]
 Betreff: [EVGRAY] Re: Push pull generator.
    
 Hi Sven
 Use thin magnet wire wrapped inside tiny ferrite tubes  just a few turns ,maybe 4 turns....use very thin magnet wire around 28 or 30 GA  thickness (thin)...then connect this wire to mosfet gate.
 Also they say to include in series a low ohm resistor too, such as 5 ohm 1/4W size to help dampen any ringing...the ferrite squelches the HF stuff, the resistor the LF stuff is what I understand.
 Maybe your mosfets do not turn off fast enough, this would make for excess heat that is for sure!....are you using a driver chip to the mosfets??
 Try 4421 or 4422 types...
 Also if no drivers feeding the gates, you better for sure have a 10K "pull sown" resistor stretching across mosfet  from gate to source on the mosfet....I like to include the 10K resistor even if using a 4422 or 4421 driver chip just to be double safe....
 Kone
 ...

The message got truncated and the attachment discarded!
Here's another try.


Hi Sven,

Don't put ferrite beads on the gate circuit as they add inductance. Try to avoid inductance by making the distance between the gate driver and the gate as small as possible an minimize any area between the wires that act like a coil winding adding inductance. The gate-source is a capacitor. Adding inductance makes it an LC-tank which makes oscillations when storing energy. The oscillations are amplified at the drain. Also the inductive kickback from any inductance can go above the gate-source breakdown voltage which degrades and destroys the MOSFET sooner or later. By adding extra capacitance in parallel to the gate-source capacitance slows down the MOSFET as the drain signal follows the gate signal. The series resistor at the gate is for absorbing the energy of the formed LC-tank to prevent oscillations of the gate and thus also oscillations on the drain.

Opto couplers are slow devices. The TLP250 has a delay of 0.5 us and a max frequency of 25 kHz.

A fuse normally isn't fast enough to protect semiconductors from destruction. Use an electronic protection that removes the gate signal to shut off the device. For over voltage protection use some kind of snubber active or passive. A crowbar protection made by using a thyristor (SCR) that shorts out the voltage supply rails and blows a fuse is an effective way to prevent over voltages.

'Fundamentals of MOSFET and IGBT Gate Driver Circuits:' (http://www.ti.com/lit/ml/slua618/slua618.pdf http://www.ti.com/lit/ml/slua618/slua618.pdf).
Just follow those rules to make a good gate driver. If not doing that all kinds of trouble may arise.

If the impedance of the capacitors are less than that of B2 the current mainly takes the path shown. If R2 is omitted this may cause ringing when charging the gate capacitance. R2 has to absorb the energy to prevent oscillations in the tank formed by the inductance of the two loops and the capacitance of the gate-source capacitor. For fast response R2 is chosen for critical damping which means no ringing.

Regards
Ole 

---In [email protected], <onielsen@...> wrote :

 Hi Sven,

Don't put ferrite beads on the gate circuit as they add inductance. Try to avoid inductance by making the distance between the gate driver and the gate as small as possible an minimize any area between the wires that act like a coil winding adding inductance. The gate-source is a capacitor. Adding inductance makes it an LC-tank which makes oscillations when storing energy. The oscillations are amplified at the drain. Also the inductive kickback from any inductance can go above the gate-source breakdown voltage which degrades and destroys the MOSFET sooner or later. By adding extra capacitance in parallel to the gate-source capacitance slows down the MOSFET as the drain signal follows the gate signal. The series resistor at the gate is for absorbing the energy of the formed LC-tank to prevent oscillations of the gate and thus also oscillations on the drain.

Opto couplers are slow devices. The TLP250 has a delay of 0.5 us and a max frequency of 25 kHz.

A fuse normally isn't fast enough to protect semiconductors from destruction. Use an electronic protection that removes the gate signal to shut off the device. For over voltage protection use some kind of snubber active or passive. A crowbar protection made by using a thyristor (SCR) that shorts out the voltage supply rails and blows a fuse is an effective way to prevent over voltages.

'Fundamentals of MOSFET and IGBT Gate Driver Circuits:' (http://www.ti.com/lit/ml/slua618/slua618.pdf http://www.ti.com/lit/ml/slua618/slua618.pdf).
Just follow those rules to make a good gate driver. If not doing that all kinds of trouble may arise.

Yahoo having a bad day again.

Hi Sven,

Don't put ferrite beads on the gate circuit as they add inductance. Try to avoid inductance by making the distance between the gate driver and the gate as small as possible an minimize any area between the wires that act like a coil winding adding inductance. The gate-source is a capacitor. Adding inductance makes it an LC-tank which makes oscillations when storing energy. The oscillations are amplified at the drain. Also the inductive kickback from any inductance can go above the gate-source breakdown voltage which degrades and destroys the MOSFET sooner or later. By adding extra capacitance in parallel to the gate-source capacitance slows down the MOSFET as the drain signal follows the gate signal. The series resistor at the gate is for absorbing the energy of the formed LC-tank to prevent oscillations of the gate and thus also oscillations on the drain.

Opto couplers are slow devices. The TLP250 has a delay of 0.5 us and a max frequency of 25 kHz.

A fuse normally isn't fast enough to protect semiconductors from destruction. Use an electronic protection that removes the gate signal to shut off the device. For over voltage protection use some kind of snubber active or passive. A crowbar protection made by using a thyristor (SCR) that shorts out the voltage supply rails and blows a fuse is an effective way to prevent over voltages.

'Fundamentals of MOSFET and IGBT Gate Driver Circuits:' (http://www.ti.com/lit/ml/slua618/slua618.pdf http://www.ti.com/lit/ml/slua618/slua618.pdf).
Just follow those rules to make a good gate driver. If not doing that all kinds of trouble may arise.

If the impedance of the capacitors are less than that of B2 the current mainly takes the path shown. If R2 is omitted this may cause ringing when charging the gate capacitance. R2 has to absorb the energy to prevent oscillations in the tank formed by the inductance of the two loops and the capacitance of the gate-source capacitor. For fast response R2 is chosen for critical damping which means no ringing.

Regards
Ole

Good post onielson! The first one was fine.
protecting the Gate is very important, maybe a ~20 volt Zener, diode, (or back to back Zeners) and resistor. Seem to work for me.I still like the PVI5050 as (Gate input) it is stand alone and can be run with a micro-controller. It is much slower..

    On Thursday, July 5, 2018, 4:50:36 p.m. MDT, [email protected] [EVGRAY] <[email protected]> wrote:  
 
     
The message got truncated and the attachment discarded!
Here's another try.


Hi Sven,

Don't put ferrite beads on the gate circuit as they add inductance. Try to avoid inductance by making the distance between the gate driver and the gate as small as possible an minimize any area between the wires that act like a coil winding adding inductance. The gate-source is a capacitor. Adding inductance makes it an LC-tank which makes oscillations when storing energy. The oscillations are amplified at the drain. Also the inductive kickback from any inductance can go above the gate-source breakdown voltage which degrades and destroys the MOSFET sooner or later. By adding extra capacitance in parallel to the gate-source capacitance slows down the MOSFET as the drain signal follows the gate signal. The series resistor at the gate is for absorbing the energy of the formed LC-tank to prevent oscillations of the gate and thus also oscillations on the drain.

Opto couplers are slow devices. The TLP250 has a delay of 0.5 us and a max frequency of 25 kHz.

A fuse normally isn't fast enough to protect semiconductors from destruction. Use an electronic protection that removes the gate signal to shut off the device. For over voltage protection use some kind of snubber active or passive. A crowbar protection made by using a thyristor (SCR) that shorts out the voltage supply rails and blows a fuse is an effective way to prevent over voltages.

'Fundamentals of MOSFET and IGBT Gate Driver Circuits:' (http://www.ti.com/lit/ml/slua618/slua618.pdf).
Just follow those rules to make a good gate driver. If not doing that all kinds of trouble may arise.

If the impedance of the capacitors are less than that of B2 the current mainly takes the path shown. If R2 is omitted this may cause ringing when charging the gate capacitance. R2 has to absorb the energy to prevent oscillations in the tank formed by the inductance of the two loops and the capacitance of the gate-source capacitor. For fast response R2 is chosen for critical damping which means no ringing.

Regards
Ole


---In [email protected], <onielsen@...> wrote :

Hi Sven,

Don't put ferrite beads on the gate circuit as they addinductance. Try to avoid inductance by making the distance between thegate driver and the gate as small as possible an minimize any areabetween the wires that act like a coil winding adding inductance. Thegate-source is a capacitor. Adding inductance makes it an LC-tank whichmakes oscillations when storing energy. The oscillations are amplifiedat the drain. Also the inductive kickback from any inductance can goabove the gate-source breakdown voltage which degrades and destroys theMOSFET sooner or later. By adding extra capacitance in parallel to thegate-source capacitance slows down the MOSFET as the drain signalfollows the gate signal. The series resistor at the gate is forabsorbing the energy of the formed LC-tank to prevent oscillations ofthe gate and thus also oscillations on the drain.

Opto couplers are slow devices. The TLP250 has a delay of 0.5 us and a max frequency of 25 kHz.

Afuse normally isn't fast enough to protect semiconductors fromdestruction. Use an electronic protection that removes the gate signalto shut off the device. For over voltage protection use some kind ofsnubber active or passive. A crowbar protection made by using athyristor (SCR) that shorts out the voltage supply rails and blows afuse is an effective way to prevent over voltages.

'Fundamentals of MOSFET and IGBT Gate Driver Circuits:' (http://www.ti.com/lit/ml/slua618/slua618.pdf).
Just follow those rules to make a good gate driver. If not doing that all kinds of trouble may arise.
  #yiv4113843395 #yiv4113843395 -- #yiv4113843395ygrp-mkp {border:1px solid #d8d8d8;font-family:Arial;margin:10px 0;padding:0 10px;}#yiv4113843395 #yiv4113843395ygrp-mkp hr {border:1px solid #d8d8d8;}#yiv4113843395 #yiv4113843395ygrp-mkp #yiv4113843395hd {color:#628c2a;font-size:85%;font-weight:700;line-height:122%;margin:10px 0;}#yiv4113843395 #yiv4113843395ygrp-mkp #yiv4113843395ads {margin-bottom:10px;}#yiv4113843395 #yiv4113843395ygrp-mkp .yiv4113843395ad {padding:0 0;}#yiv4113843395 #yiv4113843395ygrp-mkp .yiv4113843395ad p {margin:0;}#yiv4113843395 #yiv4113843395ygrp-mkp .yiv4113843395ad a {color:#0000ff;text-decoration:none;}#yiv4113843395 #yiv4113843395ygrp-sponsor #yiv4113843395ygrp-lc {font-family:Arial;}#yiv4113843395 #yiv4113843395ygrp-sponsor #yiv4113843395ygrp-lc #yiv4113843395hd {margin:10px 0px;font-weight:700;font-size:78%;line-height:122%;}#yiv4113843395 #yiv4113843395ygrp-sponsor #yiv4113843395ygrp-lc .yiv4113843395ad {margin-bottom:10px;padding:0 0;}#yiv4113843395 #yiv4113843395actions {font-family:Verdana;font-size:11px;padding:10px 0;}#yiv4113843395 #yiv4113843395activity {background-color:#e0ecee;float:left;font-family:Verdana;font-size:10px;padding:10px;}#yiv4113843395 #yiv4113843395activity span {font-weight:700;}#yiv4113843395 #yiv4113843395activity span:first-child {text-transform:uppercase;}#yiv4113843395 #yiv4113843395activity span a {color:#5085b6;text-decoration:none;}#yiv4113843395 #yiv4113843395activity span span {color:#ff7900;}#yiv4113843395 #yiv4113843395activity span .yiv4113843395underline {text-decoration:underline;}#yiv4113843395 .yiv4113843395attach {clear:both;display:table;font-family:Arial;font-size:12px;padding:10px 0;width:400px;}#yiv4113843395 .yiv4113843395attach div a {text-decoration:none;}#yiv4113843395 .yiv4113843395attach img {border:none;padding-right:5px;}#yiv4113843395 .yiv4113843395attach label {display:block;margin-bottom:5px;}#yiv4113843395 .yiv4113843395attach label a {text-decoration:none;}#yiv4113843395 blockquote {margin:0 0 0 4px;}#yiv4113843395 .yiv4113843395bold {font-family:Arial;font-size:13px;font-weight:700;}#yiv4113843395 .yiv4113843395bold a {text-decoration:none;}#yiv4113843395 dd.yiv4113843395last p a {font-family:Verdana;font-weight:700;}#yiv4113843395 dd.yiv4113843395last p span {margin-right:10px;font-family:Verdana;font-weight:700;}#yiv4113843395 dd.yiv4113843395last p span.yiv4113843395yshortcuts {margin-right:0;}#yiv4113843395 div.yiv4113843395attach-table div div a {text-decoration:none;}#yiv4113843395 div.yiv4113843395attach-table {width:400px;}#yiv4113843395 div.yiv4113843395file-title a, #yiv4113843395 div.yiv4113843395file-title a:active, #yiv4113843395 div.yiv4113843395file-title a:hover, #yiv4113843395 div.yiv4113843395file-title a:visited {text-decoration:none;}#yiv4113843395 div.yiv4113843395photo-title a, #yiv4113843395 div.yiv4113843395photo-title a:active, #yiv4113843395 div.yiv4113843395photo-title a:hover, #yiv4113843395 div.yiv4113843395photo-title a:visited {text-decoration:none;}#yiv4113843395 div#yiv4113843395ygrp-mlmsg #yiv4113843395ygrp-msg p a span.yiv4113843395yshortcuts {font-family:Verdana;font-size:10px;font-weight:normal;}#yiv4113843395 .yiv4113843395green {color:#628c2a;}#yiv4113843395 .yiv4113843395MsoNormal {margin:0 0 0 0;}#yiv4113843395 o {font-size:0;}#yiv4113843395 #yiv4113843395photos div {float:left;width:72px;}#yiv4113843395 #yiv4113843395photos div div {border:1px solid #666666;min-height:62px;overflow:hidden;width:62px;}#yiv4113843395 #yiv4113843395photos div label {color:#666666;font-size:10px;overflow:hidden;text-align:center;white-space:nowrap;width:64px;}#yiv4113843395 #yiv4113843395reco-category {font-size:77%;}#yiv4113843395 #yiv4113843395reco-desc {font-size:77%;}#yiv4113843395 .yiv4113843395replbq {margin:4px;}#yiv4113843395 #yiv4113843395ygrp-actbar div a:first-child {margin-right:2px;padding-right:5px;}#yiv4113843395 #yiv4113843395ygrp-mlmsg {font-size:13px;font-family:Arial, helvetica, clean, sans-serif;}#yiv4113843395 #yiv4113843395ygrp-mlmsg table {font-size:inherit;font:100%;}#yiv4113843395 #yiv4113843395ygrp-mlmsg select, #yiv4113843395 input, #yiv4113843395 textarea {font:99% Arial, Helvetica, clean, sans-serif;}#yiv4113843395 #yiv4113843395ygrp-mlmsg pre, #yiv4113843395 code {font:115% monospace;}#yiv4113843395 #yiv4113843395ygrp-mlmsg * {line-height:1.22em;}#yiv4113843395 #yiv4113843395ygrp-mlmsg #yiv4113843395logo {padding-bottom:10px;}#yiv4113843395 #yiv4113843395ygrp-msg p a {font-family:Verdana;}#yiv4113843395 #yiv4113843395ygrp-msg p#yiv4113843395attach-count span {color:#1E66AE;font-weight:700;}#yiv4113843395 #yiv4113843395ygrp-reco #yiv4113843395reco-head {color:#ff7900;font-weight:700;}#yiv4113843395 #yiv4113843395ygrp-reco {margin-bottom:20px;padding:0px;}#yiv4113843395 #yiv4113843395ygrp-sponsor #yiv4113843395ov li a {font-size:130%;text-decoration:none;}#yiv4113843395 #yiv4113843395ygrp-sponsor #yiv4113843395ov li {font-size:77%;list-style-type:square;padding:6px 0;}#yiv4113843395 #yiv4113843395ygrp-sponsor #yiv4113843395ov ul {margin:0;padding:0 0 0 8px;}#yiv4113843395 #yiv4113843395ygrp-text {font-family:Georgia;}#yiv4113843395 #yiv4113843395ygrp-text p {margin:0 0 1em 0;}#yiv4113843395 #yiv4113843395ygrp-text tt {font-size:120%;}#yiv4113843395 #yiv4113843395ygrp-vital ul li:last-child {border-right:none !important;}#yiv4113843395

Mick,
Tesla also could charge capacitors near a spark gap.  He could also hold a capacitor in proximity to the spark gap and charge it to a very high capacity. In fact, he could charge them until their capacity was exceeded and an explosion would occur. Considering all of these effects, he realized this technology would allow for the wireless transmission of power.
George

    On Friday, 6 July 2018, 1:15:31 am AEST, Mick [email protected] [EVGRAY] <[email protected]> wrote:  
 
     
 

Sven,
 
Determine if the mosfets are hot or the heat sinks are hotter.  Sometimes with these strange energies the heat sink is too hot to touch and the fet can be touched no problem.  Maybe you have infrared thermometer and can test each temperature easily.

George!  Absolutely correct!  Joel & I were able to charge capacitors in 
the field of the MRA with only one lead attached to ground potential. If 
you pulse one side of a capacitor there is a definite displacement that 
occurs in the dielectric.  BTW! where the charge is stored (not on the 
plates).


On 7/6/2018 12:44 AM, george gray [email protected] [EVGRAY] wrote:
> Mick,
>
> Tesla also could charge capacitors near a spark gap.  He could also 
> hold a capacitor in proximity to the spark gap and charge it to a very 
> high capacity. In fact, he could charge them until their capacity was 
> exceeded and an explosion would occur. Considering all of these 
> effects, he realized this technology would allow for the wireless 
> transmission of power.
>
> George
>
> On Friday, 6 July 2018, 1:15:31 am AEST, Mick [email protected] 
> [EVGRAY] <[email protected]> wrote:
>
>
> Sven,
>
> Determine if the mosfets are hot or the heat sinks are hotter.  
> Sometimes with these strange energies the heat sink is too hot to 
> touch and the fet can be touched no problem.  Maybe you have infrared 
> thermometer and can test each temperature easily.
>
> From my experiments with the low voltage wireless energy I have found 
> the effect is superior if the whole system is in a resonant loop 
> rather than being forced to resonate outside of it's natural phase by 
> a 555 or equivalent.
>
> I am filling motor run caps capacitors at 140 volts from two feet away 
> with a power supply of 10 v 400ma input using a rectenna.
>
>
>
> On 7/5/2018 6:16 AM, [email protected] <mailto:[email protected]> 
> [EVGRAY] wrote:
>>
>> Hello everyone,
>>
>> I have continued to work on the pushpull converter and now have a 
>> frequency range of 37Hz to over 400Hz which is certainly sufficient 
>> for my heavy iron core transformer. Pulse width is adjustable from 
>> 780μs to 6ms at 37Hz. I think a lot can be experimented in this area.
>>
>> Since I did not want to repeat the same mistakes as at that time, I 
>> concentrated strongly on the driver control and had to say if I 
>> increase the supply voltage for the mosfets at some point a point 
>> comes where there are dropouts after a long search, I took out the 
>> gate resistance at some point and the error with incandescent lamps 
>> was gone, now I am with 2 transformers as inductance testing and from 
>> now I increase the pulse width and a channel slowly starts to 
>> flicker, with lower voltage it is better again and the same current 
>> at higher frequency, it is also worse.
>>
>> I have also hung ferrite beads on the gate lines and had a super 
>> clean signal but with very small pulse width, the signal collapsed, 
>> the ferrites seem to act as gate resistors, unfortunately there are 
>> no small pulse widths to produce
>>
>> I soon have the feeling that the low voltage supply does not create 
>> or am I wrong.
>>
>> Mosfets are IXFN44N100P
>>
>> The mosfets are very hot despite the large heat sink at just 4 amps 
>> although they are designed for 37 amps per Mosfet I find that very 
>> strange.
>>
>> Best regards
>>
>> Sven
>>
>
>

Hello Ole,

I looked at the PDF and added one from Toshiba where on the last page you should put a capacitor between gate and source, I used a very small one with 10nF.


For me, C1 is 1μF MKP and C2 is 68μF Elko.
R2 2.2 ohms and R3 now 10K.

I've twisted all the wires and connected, unfortunately, I can not keep the lines so short by my design, but I only work with frequencies from 37 to just over 600 Hz.

I send some Scope pictures I have the feeling that the Mosfet switch clean and the small disturbances do not affect them, at least if you look at the scope pictures from the power section.

What surprises me is that my Scope probably disturbances in the system brings at least at small pulse widths a signal from whatever. The scope is grounded if there are leakage currents flowing over it?!?

regards

Sven

Re-post with the picture attached.

Regards
Ole
 
---In [email protected], <onielsen@...> wrote :

 Hi Sven,

The capacitor in parallel to the parasitic gate-source capacitor can be omitted if using a low impedance gate driver placed close to the MOSFET for minimizing inductive loops. Twisting the gate source wires forms a transmission line of less impedance than the big loops on the picture. Twisting the wires of different gate circuits together couples them together. This can be seen in the scope shot where spikes are transferred between the two gate circuits. This also happens between the big loops that work as loop antennas which both send and receive the signals. See the picture below if it makes it through Yahoo.

"For me, C1 is 1μF MKP and C2 is 68μF Elko.
R2 2.2 ohms and R3 now 10K."
This doesn't tell anything without knowing how or what they're connected to.

Regards
Ole