Capacitor across flyback primary / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Capacitor across flyback primary

1 2 next

Move Thread

LAN_403

cjanik

Fri Jan 11 2013, 03:54AM

Registered Member #4460 Joined: Sat Apr 07 2012, 08:31AM
Location:
Posts: 9

I saw someone on youtube put a capacitor across the primary of their flyback which was powered by a 555 driver. I tried the same thing on my TL494 full bridge flyback driver which I built from the schematic on scopeboy's site and I got excellent results with a .15 uf capacitor. It stopped working after a while though and I hooked it up to the scope to see that my my TC4429 was not outputting a signal any longer and was also getting hot enough to burn me. I was using 12v to power the logic and a separate 24v 16a supply for the mosfets so I was driving the circuit terribly hard. My question is, will that capacitor across the primary ruin my inverter? Or is something else wrong? I don't really want to burn up another TC4429 until I figure out whats going on. I would appreciate anyone's insight!

Proud Mary

Fri Jan 11 2013, 07:24AM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

The addition of the capacitor has turned the primary into a parallel resonant tuned circuit, across which a high potential difference will appear will appear when it is close to resonance.

StormInABottle

Fri Jan 11 2013, 09:20AM

Registered Member #9252 Joined: Fri Jan 04 2013, 06:27AM
Location: Andromeda
Posts: 253

I have had such a exp but with a cfl.. I just take the the cfl output wires that goes into the tube. put a ceramic cap on the 2 output wires and into the original flyback primary coil. = more than 15 times long arcs than with no cap

Just don't do what i just said. once you take away the hv- wire away from the red wire the cfl goes boom and the house circuit breaker goes off. so be careful
tho maybe if you heat sink the transistors on a 110 watt cfl. it would live longer. also it's better to have the fbt under oil and the hv -ve grounded if you plan on doing that.
i noticed the cfl lives much longer that way. :)

Also that is because of resonance

cjanik

Sat Jan 12 2013, 12:02AM

Registered Member #4460 Joined: Sat Apr 07 2012, 08:31AM
Location:
Posts: 9

Proud Mary wrote ...

The addition of the capacitor has turned the primary into a parallel resonant tuned circuit, across which a high potential difference will appear will appear when it is close to resonance.

Ok that makes sense. What I still cant figure out though is how my TC4429 got toasted, since it only connects to the mosfet gates, its not connected to the primary at all

Proud Mary

Sat Jan 12 2013, 12:44AM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Circuit diagrams often fool us into thinking that electricity flows from left to right, or that electrons will always flow following the lines on the paper. Voltage spikes in particular can do things like travel back up wires and wreck your power supply, or induce currents in neighbouring wires where they are not wanted, and so on, Of course, we can't be sure that this is what did for your chip, but if you get a high AC voltage appearing in one place, and something fizzles out nearby, the possibility should be checked. Is the supply line adequately decoupled, so that rogue spikes get by-passed to Earth, rather than getting coupled into every circuit hanging from the supply rail?

cjanik

Sat Jan 12 2013, 01:07AM

Registered Member #4460 Joined: Sat Apr 07 2012, 08:31AM
Location:
Posts: 9

Proud Mary wrote ...

Is the supply line adequately decoupled, so that rogue spikes get by-passed to Earth, rather than getting coupled into every circuit hanging from the supply rail?

I'm not sure I fully understand the question, but if your refering to the layout, then yes, it could definitely use some work as far as the logic being close to the traces for the primary winding and the drain/source on the mosfets.

I'm also thinking maybe my primary had too few turns (only 6) I switched to a flyback with 10 turns on the primary and so far I haven't burnt another chip

Ash Small

Sat Jan 12 2013, 02:08AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

cjanik wrote ...

I'm also thinking maybe my primary had too few turns (only 6) I switched to a flyback with 10 turns on the primary and so far I haven't burnt another chip

increasing the number of turns on the primary will do two things, as far as I'm aware. Firstly, it will decrease voltage on the secondary, secondly, if the circuit was at resonance with six turns, it will put the circuit 'out of resonance'.

I imagine it's the second that is causing your IC not to burn. What is the arc like compared to before, when the IC burned?

cjanik

Sat Jan 12 2013, 03:01AM

Registered Member #4460 Joined: Sat Apr 07 2012, 08:31AM
Location:
Posts: 9

Ash Small wrote ...

cjanik wrote ...

I'm also thinking maybe my primary had too few turns (only 6) I switched to a flyback with 10 turns on the primary and so far I haven't burnt another chip

Hot, white, and fat. I know changing the number of primary turns will affect the resonant frequency. I readjusted the frequency for optimal arcs again and they're not as intense as before, which is what I expected. Hopefully the smaller arcs don't stress the IC as much.

Proud Mary

Sat Jan 12 2013, 09:38AM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

cjanik wrote ...

Proud Mary wrote ...

Is the supply line adequately decoupled, so that rogue spikes get by-passed to Earth, rather than getting coupled into every circuit hanging from the supply rail?

Read the Wikipedia entry for decoupling capacitor.

Antonio

Sat Jan 12 2013, 12:50PM

Registered Member #834 Joined: Tue Jun 12 2007, 10:57PM
Location: Brazil
Posts: 644

The use of this capacitor is as old as induction coils with mechanical switches, and a totally correct idea. Without it the pulse that appears over the switching element is only limited by breakdown of the device, and will destroy a transistor without protection very easily. But with a capacitor, the voltage over the transistor first rises to a limited level and then falls, as the capacitor resonates with the coil, and tries to go below the ground line. If a bipolar transistor is being used, the base-collector junction will be forward-biased and a large current will flow from the driving circuit, possibly destroying it. The solution is to add a fast high-voltage diode across the transistor. This can be seen in most schematic diagrams of CRT TVs and monitors, and the required elements can be obtained from the same boards from where these flyback transformers can be salvaged (along with all the driving circuit and a complete switching power supply, if you want).

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob