Help wiring a flyback transformer.

[LaZeRz]

I'm still looking around for a large 2000w or so microwave to pull the MOT and capacitor out of. Just need one more cap to get some resonance going on

 

[Fiddy]

Buy and harvest one

 

[LaZeRz]

Buy and harvest one

Meh, new microwaves have different transformers all together.

 

[Things]

^ * Inverters with short protection and open circuit protection

 

[Fiddy]

DAM those fools!

 

[Fiddy]

Got my second cap

Might place a 30uF capacitor across the primary to improve the power factor.

 

[Things]

LOL, I was looking at that picture like, "I sure hope that shifter is plastic or something!!" Good to see it's only being used as a weight and not a chicken stick

 

[Fiddy]

haha yeah, dont think id be here if it was my chicken stick

heres my jacobs ladder:

 

[Atomicrox]

For my first HV project I wired a flyback from a computer monitor to get some arcs using the simplest circuit I could find - I followed the instructions from this video:

I harvested all the components from the monitor PCB, googling the datasheets to find the descriptions that fitted best what the guy from the video said. I attached pictures of the circuit and the components and adapted the video's schematics to fit my components.

Finding the HV- was easy. I also found 2 coils, one of them connected to 4 pins and the other one connected to 3 pins. Couldn't figure out which one was the primary and which one was the feedback coil, much less which pins to use, so I just tried a *lot* of different wirings.
Most of them didn't work. Some that worked "whined" a lot, arced only 1mm and heated the transistor very quickly. Three wirings got me a pretty stable 1cm arc with lower noise and lower transistor heat. Then I tried to switch the primary and feedback coils and got a 2cm arc but it only worked twice, the transistor got very hot and it wouldn't arc anymore, so I went back to the one that got the 1cm stable arc. It worked, so I tried to make a small jacob's ladder and it just wouldn't work anymore - no whine, no arc and a lot of transistor heat. All stable wirings stopped whining and arcing. The 1mm ones still whine and arc, but slightly less.

Did I fry something? Is there anything I can do to fix it?

 

[Fiddy]

For my first HV project I wired a flyback from a computer monitor to get some arcs using the simplest circuit I could find, which was this:

I harvested all the components from the monitor PCB, googling the datasheets to find the descriptions that fitted best what the guy from the video said. I attached pictures of the circuit and the components and adapted the video's schematics to fit my components.

Finding the HV- was easy. I also found 2 coils, one of them connected to 4 pins and the other one connected to 3 pins. Couldn't figure out which one was the primary and which one was the feedback coil, much less which pins to use, so I just tried a *lot* of different wirings.
Most of them didn't work. Some that worked "whined" a lot, arced only 1mm and heated the transistor very quickly. Three wirings got me a pretty stable 1cm arc with lower noise and lower transistor heat. Then I tried to switch the primary and feedback coils and got a 2cm arc but it only worked twice, the transistor got very hot and it wouldn't arc anymore, so I went back to the one that got the 1cm stable arc. It worked, so I tried to make a small jacob's ladder and it just wouldn't work anymore - no whine, no arc and a lot of transistor heat. All stable wirings stopped whining and arcing. The 1mm ones still whine and arc, but slightly less.

Did I fry something? Is there anything I can do to fix it?

Nice work!

Are you an Aussie?

You may have burnt out either your transistor or winding in the flyback?

 

[Atomicrox]

I'm not the guy from the video, I just followed his instructions. Better edit to make that clear

Mine didn't work nearly as well as his.

I will see if the coils still pass the continuity test on the multimeter. Any simple way to test the transistor?

 

[Sigurthr]

Yeah check the resistance between the collector and the emitter, if it is low (<100k) it is probably busted.

Too few primary turns and the impedance of the primary inductor drops and you start pushing a lot of current through the transistor. Too many turns and you limit current by overdoing the impedance, and also increases leakage inductance, which eats away power.

As for reversing the feedback phasing I don't know if it would have any real effect in this circuit, or did you mean to switched which coils you were using for primary and feedback? A common setup for feedback from a flyback is a 5+5 turn primary where the center is tapped and provides feedback.

My best suggestion? Contact Jared (LeQuack) about flyback help... he makes some REAL mean flyback drivers... I have three of his, each with a different topology! Two are half bridges and one is a resonant single transistor build. The resonant single is basically an adjustable frequency 555 timer in astable oscillator mode driving an IRFP450 MOSFET at the tank circuit's resonant frequency. Primary cap in parallel with primary winding and you turn a pot adjust frequency and find the resonant point. Simple circuit, elegantly designed, and very strong. I get long hot arcs of quite high voltage from it, even powering from only 12v. It will even run as low as 5V, but the MOSFET only gets in to the linear region and losses/heat is signifigant.

If you want to put some parts in to it... a center-tap-feedback half bridge of IRFP260s driven by a GDT with mosfet drive chips is the way to go! You'll likely be able to burn out the transformer if you punch enough current through it.

 

[DashApple]

Nice video , i tryed a sigle transistor desing once could never get it to work well . moved over to a ZVS and its great i built mine for 10 pounds and i have enough parts to make 3 more . Using IRF540N Fets i can push 40 volts into ZVS but with heat generation in mosfets but that gives 7" Arcs , but at 30 Volts in i get no heat what so ever and get 4-5" Arcs from flyback .

My previouse ZVS used IRF150N mosfets and i used a 12 volt supply for gates and then 55 Volts for mian supply , i got 8" arcs and after 3 minuits of Arcs the flyback swelled up and cracked and failed



Electric Arc by TwirlyWhirly555, on Flickr

 

[Atomicrox]

Yeah check the resistance between the collector and the emitter, if it is low (<100k) it is probably busted.

Too few primary turns and the impedance of the primary inductor drops and you start pushing a lot of current through the transistor. Too many turns and you limit current by overdoing the impedance, and also increases leakage inductance, which eats away power.

As for reversing the feedback phasing I don't know if it would have any real effect in this circuit, or did you mean to switched which coils you were using for primary and feedback? A common setup for feedback from a flyback is a 5+5 turn primary where the center is tapped and provides feedback.

My best suggestion? Contact Jared (LeQuack) about flyback help... he makes some REAL mean flyback drivers... I have three of his, each with a different topology! Two are half bridges and one is a resonant single transistor build. The resonant single is basically an adjustable frequency 555 timer in astable oscillator mode driving an IRFP450 MOSFET at the tank circuit's resonant frequency. Primary cap in parallel with primary winding and you turn a pot adjust frequency and find the resonant point. Simple circuit, elegantly designed, and very strong. I get long hot arcs of quite high voltage from it, even powering from only 12v. It will even run as low as 5V, but the MOSFET only gets in to the linear region and losses/heat is signifigant.

If you want to put some parts in to it... a center-tap-feedback half bridge of IRFP260s driven by a GDT with mosfet drive chips is the way to go! You'll likely be able to burn out the transformer if you punch enough current through it.

Thanks a lot, did your test and the transistor was dead - I'm glad it wasn't the flyback. Resistance between any 2 pins of the same coil is near 1ohm, which is about the same I had before.

I was using the internal coils and the pins had no markings, I meant trying both coils as primary/feedback, not reversing the polarity. Reversing the polarity in both coils made a lot of difference, though. Maybe they added some diodes inside the flyback?

This flyback is full of plastic, I'm not sure I could wind even 5+5 turns around the exposed ferrite. Maybe I can break some of the plastic away.

I'm willing to try a more complex circuit but only with the schematics and extremely easy to find components. If they're even moderately rare I won't be able to buy around here - that's the main reason I quitted the hobby when younger.

I'd like to at least get a small jacob's ladder working on this simple circuit before trying something else. There are other transistors the board but I couldn't figure out if they'd work, any ideas?
ST MRC DMV32B C714A (ST looks like a trade mark)
F C5802 949 (this one is huge, F looks like a trade mark)
F 944 IRF634A (lots of these, F looks like a trade mark)
TIF F 122 9HT


@ionlaser555: That's HUGE!

 

[Hiemal]

Alright; If you want to make a proper flyback transformer driver, you're going to have to not scavenge for parts from a TV... it's not going to happen, for a lot of reasons. One of them is that TV's are made using the cheapest parts possible. That doesn't bode well with flyback drivers as the transistors need to be...hardy, otherwise they'll explode/die/short whatever.

http://www.pocketmagic.net/wp-content/uploads/2009/05/555-flyback-driver.jpg

I know it looks a bit...difficult. However, that is the same exact flyback driver I used when I first started out with my hobby.

If you get those parts, and some prototyping board you'll be making arcs in no time. Quick note though;

You'll want to add some filtering capacitance after the 7812. Just a 1000 uF or so capacitor would be fine. You can also leave out the neon bulb as well. If you can't find those MPSA42/92 transistors, you can substitute 2N3904 and 2N3906 instead. It'll work exactly the same.

I would also recommend getting an IRFP250 instead of the mosfet shown in the schematic.

You'll also need a decent power supply. A wall wort is NOT going to cut it. An ATX PSU supply would work perfectly.


Once you get this driver done and working, then you can move on to a better, much more powerful driver. You have to kind of...work your way up I guess.

Hope I helped!

 

[Atomicrox]

Cool, I'll see if I can find those components around here.

Can you help me with a some details?
-Can I use a regular breadboard or is the current too high?
-Do I use regular small resistors or high-wattage ones?
-Any specific type/voltage for the capacitors?
-That 5-25t means I have to wind 5 to 25 turns around the ferrite? That's gonna be hard to do, there's a lot of plastic in the way..

I happen to have a working ATX PSU here, that's not going to be a problem