Single Mosfet Flyback Driver / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Single Mosfet Flyback Driver

previous 1 2

Move Thread

LAN_403

Steve Ward

Mon May 15 2006, 11:38PM

Registered Member #146 Joined: Sun Feb 12 2006, 04:21AM
Location: Austin Tx
Posts: 1055

mine gives about an inch. The resistor (if you understand design) is to limit the instantaneous current from going to infinity. I don't understand why so many of you guys don't see that.

What i dont understand is why you dont know the equation V=L*di/dt, which would show you right away that the instantaneous current (when the fet first turns on) is 0, and builds up at a rate given by that equation (unless you saturate, but thats a completely different problem now). Your 1 ohm resistor really is just a waste of power.

A better thing to do would to be to give the 555 its own large lytic (say 1000uF), and run its power though 5 or 10 ohms. This RC filter would decouple the 555 supply from the noise generated from the "power" side of the flyback driver. Right now your are kind of decoupling the wrong side.

Clamping circuit is completly unnecesarry for the low voltages given, just diode across the transistor can be used but it's not needed too.

Diode across flyback primary eats up the inductive spike and just burns it in a resistor.

This flyback driver works off of the "inductive kick" produced when you quickly shut off the fet. This makes di/dt very high, hence producing a large voltage spike. Since the coupling between primary and secondary is not unity (or even close), even if you were to short the output, you will still get a spike across the fet, given by the leakage inductance of the primary. As to your suggestion about the diode across the coil, yes, this would clamp the spike, but would also really hinder performance and may cause saturation (because the current would probably freewheel until the next pulse, and add up with each incoming pulse).

The optimal choice would be a capacitor straight across the primary (or mosfet) chosen so that the cap stores as much energy (at a "safe" voltage) as the flyback does. This way, magnetic energy is converted to potential energy if its not used up by the output of the flyback, or other losses. This starts to get into class E stuff, which would probably work fairly well for a flyback driver, though its overkill.

teravolt

Tue May 16 2006, 05:21AM

Registered Member #195 Joined: Fri Feb 17 2006, 08:27PM
Location: Berkeley, ca.
Posts: 1111

ferite cores have a natural resonance. try feeding that mosfet with a 555 and A 7474 divide by 2 sqare wave circuit with a siff power supply that should work. You may need a reverse diode across the fet . Anybody can correct me on this but I Think 2n3055 fly back circuit gets it's frequency from the core

Marko

Tue May 16 2006, 08:33AM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

This flyback driver works off of the "inductive kick" produced when you quickly shut off the fet. This makes di/dt very high, hence producing a large voltage spike. Since the coupling between primary and secondary is not unity (or even close), even if you were to short the output, you will still get a spike across the fet, given by the leakage inductance of the primary. As to your suggestion about the diode across the coil, yes, this would clamp the spike, but would also really hinder performance and may cause saturation (because the current would probably freewheel until the next pulse, and add up with each incoming pulse).

I used 30V input with 140V bjt and never needed any clamping or ever fried the transistor (but I did saturate and fry it's base drive circuit

)

Putting almost ''anything'' across the primary usually just reduces the output.
As Steve ward explained you want inductive kick to happen here, it's the point of flyback operation.

In TV's they usually feed the flyback with 80V supply and sometimes use even 1200 or 1700V transistors, just to make sure that peak voltage of primary never gets even close to actual breakdown of transistor.

There is also one thing: kick on the secondary occurs in one polarity with much more peak voltage than another. You want the diode to rectify that part, so if it is wrong you yust need to swap the primary pins.

Dr. Dark Current

Tue May 16 2006, 01:56PM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

Firkragg wrote ...

Today I made an optimized STFD, the only problem is that the fly back gets hot (not the transistor!)

Are you sure it's not arcing? Some pics of the setup, arcs..?

It is not arcing (if you meant, internally.) It is a small potted DC flyback, but easily makes a 4cm spark. I will build this driver incl. mains xfmr and flyback into PC PSU case, but I will post it in the Flyback arc thread because I think this is getting a bit OT...

wrote ...

Putting almost ''anything'' across the primary usually just reduces the output.
As Steve ward explained you want inductive kick to happen here, it's the point of flyback operation.

Not quite. In my flyback driver with a tl494 driven Mosfet a small (few nF) cap across the D-S junction: 1) reduces (by a magnitude) power draw when the HV output is unloaded (resonance with primary); 2) reduces FET heating; 3) output voltage is slightly higher. This is just my experience that may vary from yours.

Marko

Tue May 16 2006, 04:59PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Not quite. In my flyback driver with a tl494 driven Mosfet a small (few nF) cap across the D-S junction: 1) reduces (by a magnitude) power draw when the HV output is unloaded (resonance with primary); 2) reduces FET heating; 3) output voltage is slightly higher. This is just my experience that may vary from yours.

I said primary, not the transistor

That what you are describing would now be class E if it is 'resonating with primary', I don't know can you be sure about that.

Have you scoped it, some pics?

Dr. Dark Current

Tue May 16 2006, 05:23PM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

Firkragg wrote ...

I said primary, not the transistor

This is actually quite the same. (if you have good (foil) cap parralel to the supply rails)

Firkragg wrote ...

That what you are describing would now be class E if it is 'resonating with primary', I don't know can you be sure about that.

It is not class E, since the LC (with C being the D-S cap) resonates many times higher frequency than the mosfet is switching.

J.M.

Steve Ward

Tue May 16 2006, 06:25PM

Registered Member #146 Joined: Sun Feb 12 2006, 04:21AM
Location: Austin Tx
Posts: 1055

It is not class E, since the LC (with C being the D-S cap) resonates many times higher frequency than the mosfet is switching.

Yeah, the cap is probably just large enough to keep the fet from avalanching from over-voltage, so this could greatly reduce heating.

Sulaiman

Tue May 16 2006, 06:33PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

As far as I can imagine,
a single transistor power oscillator/invertor will need to
saturate the core each cycle for timing/feedback.
This will inevitably cause a lot of core heating/inefficiency.

previous 1 2

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