Improved single transistor flyback driver / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Improved single transistor flyback driver

first previous 2 3 4 5 next last

Move Thread

LAN_403

Dr. Dark Current

Wed Apr 23 2008, 07:08AM

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

Antonio wrote ...

I recommend adding another diode in series with a zener diode between emitter and base of the transistor, to avoid excessive inverse voltages there.

Well, the large bipolars (e.g. 2N3055) are rated maybe an amp or more of reverse base current, so I left it out. Adding it does not hurt of course.

Firefox - for me, for some reason the higher voltage transistors (>200V) don't work very well, get hot and sometimes fail to oscillate. If it works for you, it's good. Go as high supply voltage as you dare, the transistor should theoretically stand 50V or more (watch supply current and transistor temp).
Maybe you can try reversing both primary and feedback coils, this sometimes significantly improves performance with rectified transformers.

Steve Conner

Wed Apr 23 2008, 09:48AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Antonio, are you saying that the cycle will end when the transistor goes out of saturation, even if the core doesn't saturate?

I guess that makes sense, since the transistor out of saturation is a constant current sink, constant current implies di/dt = 0, and hence there can be no induced voltage in the transformer to provide the base drive.

Antonio

Wed Apr 23 2008, 03:56PM

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

Dr. Spork wrote ...

Antonio, are you saying that the cycle will end when the transistor goes out of saturation, even if the core doesn't saturate?

I guess that makes sense, since the transistor out of saturation is a constant current sink, constant current implies di/dt = 0, and hence there can be no induced voltage in the transformer to provide the base drive.

Yes. The operation is, essentially:
1 - The collector current rises in a ramp from zero while the base current is constant.
2 - When the transistor leaves saturation, the collector current ceases to increase.
3 - The base current goes to zero because dic/dt is zero.
4 - A large positive pulse appears at the collector, caused by the current in the upper inductance charging the capacitor and the load capacitance, while a negative pulse appears at the base. The amplitude of this pulse is clipped if the be junction ruptures. Maybe good to put a diode in series with the base to prevent this, and a resistor between base and emitter to discharge the base. The zener+diode that I suggested before would just cause clipping at a smaller voltage.
5 - The pulse falls and tends to become negative, being clipped by the diode across the transistor. When the pulse returns to zero the cycle repeats.

Similar oscillators can be found in the standby power supply in the power supply of most PCs.

Dr. Dark Current

Wed Apr 23 2008, 04:27PM

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

Maybe I'm wrong but this is how I think the circuit works;

1) Transistor gets turned on by the resistor divider.
2) Voltage gets induced on the feedback coil. This voltage forces even more current through the transistor base.
3)As the transformer core saturates, the base currents starts to fall until the transistor goes out of saturation.
4) As the transistor turns off, the magnetic field collapses, and induces a high voltage pulse of OPPOSITE polarity on all coils - so this forces the base voltage to go negative, shutting the transistor off, untill the voltage "flies back" . The cycle repeats.

If the primary-feedback phasing was opposite, the circuit would fail to oscillate, because the HV spike of opposite polarity would force the transistor ON instead of OFF - so it would never happen.

EDIT - Here are quick MS Paint waveforms, in no way accurate, but they illustrate my idea good enough:

Antonio

Wed Apr 23 2008, 09:50PM

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

The waveforms assuming that the transformer does not saturate are similar, but the current grows linearly. I show below the result of a simulation with SwitcherCAD III, showing the voltages at the collector and at the base, and the currents trough the collector and the diode across the transistor. The simulator assumed that the be junction does not break down.
Oscilador

cbfull

Fri Apr 25 2008, 11:57AM

Registered Member #187 Joined: Thu Feb 16 2006, 02:54PM
Location: Central Ohio
Posts: 140

Just a quick note about that circuit, you can put a diode in series with the 22R segment to eliminate the wasted 30-40mA that drains through those two resistors. It's nothing major but if you start going up in power (which most of us do), the wasted power can become significant.

I might have to double check my archives but I have a circuit diagram with this diode included.

quicksilver

Fri Apr 25 2008, 02:35PM

Registered Member #1408 Joined: Fri Mar 21 2008, 03:49PM
Location: Oracle, AZ
Posts: 679

I could be way off base here but doesn't all this have an interdependency on the flyback in question? I've used the "simple driver" with a 2N3773 and it worked a bit better than the the same driver with the MJE13009 with specific transformers.

Then the "enhanced" driver appeared to function much better with a third (NTE284) transistor. The differences between flybacks appears to be a greater variable than just the driver design alone.

This had just been my experience from my sampling....but is there validity to this???

Proud Mary

Fri Apr 25 2008, 03:49PM

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

quicksilver wrote ...

I could be way off base here but doesn't all this have an interdependency on the flyback in question? I've used the "simple driver" with a 2N3773 and it worked a bit better than the the same driver with the MJE13009 with specific transformers.

Then the "enhanced" driver appeared to function much better with a third (NTE284) transistor. The differences between flybacks appears to be a greater variable than just the driver design alone.

This had just been my experience from my sampling....but is there validity to this???

Very good points I think. TV LOPTs have been optimized not only for operation at ~16kHz, but usually to drive multiple loads from primary taps and often multiple tapped secondaries in addition to the EHT winding, and all at the same time. PC Monitor flybacks will have been optimized for operation at twice this frequency and more. I would expect this optimization to be reflected in the composition of the magnetic core, and whatever gaps it may have.

dingo27

Fri May 02 2008, 10:37AM

Registered Member #890 Joined: Tue Jul 10 2007, 10:06PM
Location: Slovakia
Posts: 180

i tried that schematics, but little bit different... i dont have any diode or capacitor, resistors i have 220 and 22 ohms... primary winding 5 turns and secondary 3 turns on HV transformer from CRT...
and... i am running it from 12V PC PSU

arcs are 1cm long, but not flammable (e.g. safe) and it is doing squeeking sound ( i cant WRITE exactly :) ... i am powering with it my mini SGTC ( available at teslacoils, theme TC at 12V? )

transistor should be cooled good enough, because it is slightly overheating (2n3055)

photos as soon as i get charged my batteries (8-9 hours, maybe i buy two)

Dr. Dark Current

Fri May 02 2008, 12:59PM

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

Dingo, the capacitor makes the biggest difference. Your circuit looks to me like the original schematic. With the capacitor my transistor does not run hot and I can draw out more power.

first previous 2 3 4 5 next last

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