A Pulse Of Annoyance About Oscillators, Followed By A Flyback Of A Rant

Everyone likes to play with high voltages, right?. Even though the danger of death goes up with every volt, it’s likely that a few readers will have at some time or other made fancy long sparks. You’re reading this so you lived to tell the tale, and we’d only ever counsel only doing so safely, but the point of this piece lies not in the volts themselves but in a touch of frustration at the voltage generators. There’s a circuit I see so often which annoys me every single time, so here if you don’t mind I’m going to deliver both a little rant and a look into flyback converters.

It’s Got Coils, so It’s A Transformer

A power supply with the lid removed, visible is a large transformer
Linear power supplies with a mains transformer are a surprisingly rare sight now. Dilshan Jayakody, CC BY-SA 2.0.

How does a transformer work? An alternating current in a primary winding induces an opposite current in its secondary winding. The voltage out is equal to the turns ratio times the voltage in. Thus if you want to make a high voltage, it’s simply a case of finding a transformer with the right turns ratio, and applying the right AC to the input.

A handy choice for a high voltage transformer has been for years a TV line output transformer, also sometimes known as a flyback transformer. You could find these in CRT displays and TVs, and they consist of a square ferrite core with a big chunky high voltage overwinding for the CRT anode circuit and a load of lower voltage windings. TV designers were always out to save on parts costs, so they often had windings for all the voltage rails inside the set as well as the anode voltage, using the timebase as a crude switching power supply.

Schematic of a two transistor feedback oscillator driving a transformer.
It would not be fair to pick on [Skyler]’s circuit alone, as this is one that can be found in many places online. But it’s not a flyback circuit. (CC BY-SA 3.0)
Given a line output transformer, making some high voltage on that EHT winding should just be a case of applying the right AC to the primary and watching the pretty but dangerous sparks. And here comes the circuit which annoys me, because the simplest way to do that is to drive a primary winding from a transistor or maybe two transistors whose base is in turn driven by a feedback circuit. In other words, turn the transformer into the output device in an oscillator, and that generates the required AC.

This works. It makes the volts, and we can go home happy, right? Wrong, at least if you’re me. The oscillator is an analogue circuit driving the transistor in its linear mode, and with a near 50 % duty cycle. In other words, the transistor spends roughly half its time on and half off, with a bit in the middle where it’s neither on nor off. This means it gets very hot, and you immediately need a much bigger transistor. The transformer gets hot because the core is saturating, and the whole thing takes several amps of current. it’s a great heater, but a very inefficient high voltage generator. You can even buy a module on AliExpress that’s pretty much this circuit, and I have to say right now that I hate it.

Not All Transformers Work In A Conventional Manner

A schematic diagram of a TV horizontal driver.
This Philips TV from back in the day supplies the basic circuit of a flyback driver. The diodes on the right hand side are an integrated voltage multiplier, interacting with the capacitance between windings for their operation.

If it’s so bad then, what’s the alternative? The answer should come in looking at the circuit of the TV the transformer came from. There, a transistor switches the primary alongside a deflection coil, but it’s way more clever than a simple AC transformer circuit. It’s both a TV timebase moving the spot horizontally across the screen, and a completely different kind of power supply, a flyback supply.

When the transistor turns on, the current rises through the coils as a ramp. This provides the sawtooth required to move the dot across the screen, but it also slowly builds up the magnetic field in the transformer core. It’ll produce a voltage in the secondary, but not the big voltage needed by the CRT. At the end of the dot’s travel it needs to return to the other side of the screen, so the transistor turns off and the current in the deflection coil falls precipitously. This causes the stored magnetic field in the transformer core to collapse in a very short time, and it induces the huge voltage in that high voltage overwinding. Now a huge high voltage spike is created between the CRT anode output and ground. The CRT is designed to be a high voltage capacitor which acts as a reservoir to build up a constant DC voltage. I’ve always understood the use of the term “Flyback” to describe any supply working in this manner to refer to the period in which the dot on the CRT flies back across to the start of its next line.

My annoyance then every time I see a self oscillating converter using a flyback transformer is this: why can’t they use a flyback circuit, all they have to do is find a TV service manual or two and maybe add a 555 as a pulse generator! But aside from that, the elegant hack devised by a TV designer decades ago has of course morphed into a circuit that’s a vital weapon in the power supply designer’s arsenal. Most flyback supplies you’ll see these days aren’t used to make the huge voltages for a CRT, instead they find a place wherever a higher voltage is required than can easily be made with a boost converter. There’s a little more complexity in that the simple inductor is replaced by a flyback transformer, and if you’re paying attention it’s not quite wired as you’d expect a simple transformer step-up supply to work, but when the right magnetic is paired with the right control chip it’s an extremely efficient way to make more volts than you can otherwise get your hands on.

For their curious position somewhere between the analogue and digital worlds, switching regulators have always fascinated me. There’s no better document I can think of for anyone new to the field, and particular new to flyback converters, than Jim Williams’ 1987 app note 25 for Linear Technology: “Switching Regulators for Poets“. Enjoy making volts, and stay safe!

Header image: ZngZng, CC BY-SA 3.0.

37 thoughts on “A Pulse Of Annoyance About Oscillators, Followed By A Flyback Of A Rant

  1. “A handy choice for a high voltage transformer has been for years a TV line output transformer, also sometimes known as a flyback transformer. ”

    In 40 yrs as an electrical engineer, I have never heard the term “TV line output transformer.” In my orbit, the device in question has always been called a “flyback transformer” (though in TV repair parlance the adjective “flyback,” by itself, served equally well as a noun.)

    Is “TV line output transformer” regional terminology?

    1. Output high voltage in flyback trasformers was, at that times, a supplementary (positive) conseguence about horizontal 15Khz oscillator associated with trasformer.
      Energy that should be lost in heat during return of ray tarcing raster was used to generate high voltage for the tube. (Excuse my poor english)

    2. I once watched a friend touch a rubber-handle-coated metal tool to the output of a flyback. He watched as a spark simply jumped right over the end of the rubber coating and onto his hand. He was promptly thrown forcibly backwards against a wall and then landed with a thud.

      1. When I was very young, I watched a TV repairman working in the back of our set. He did the same thing, although with an adequately insulated screwdriver handle. Could have been a quick HV test although it also served to tell me “Don’t mess around back here, kid.”

    3. Now ANOTHER point is that the whole shebang ALSO has a specific resonance point. That sharp cutoff allows the high voltage winding to RING, which produces MORE than a simple, single voltage spike. The horizontal sync for the old NTSC American TV’s was a tad under 16KHz, and flybacks were designed for that. A REALLY cool fact is that it’s a resonant transformer, a descendant of the classic Tesla Coil! I had made a pulse driver with the 555 driving a HexFET (MOSFET), and yes, the high voltage winding DID display the classic exponential oscillation decay curve. (I used a simple capacitive pickup on the ‘scope lead at distance.) If I connected the high voltage output to a small metal plate (with the rectifier diode(s) removed), enough RF was radiated to light up a small flourecent tube about12″ away…

  2. an easy way to get high voltage at home for this diy ‘get a piece of alumium foil hovering’ project is to place a square foil (with a copper wire) infront of the crt tv and another cable to a big mass and than switch the tv on/off. And your foil starts hovering.

      1. ‘Easy, only if you happen to have a CRT TV.’
        What do you mean by that? I’m writing these lines from a ussr bunker and haven’t seen daylight in 30 years. And this 64k modem isn’t the fastest either…. ;-)

  3. The old analogue TV set was a masterpiece of circuit design genius, a BOM pared down through long experience to a set of key components doing multiple jobs in clever ways. I miss those days.

  4. Isn’t the reason for not using a 555 timer and using that 2 transistor circuit is that the 2 transistor circuit automatically operates at it’s resonant frequency? In TVs the goal is a fixed, defined voltage. In the 2 transistor circuit it’s usually highest voltage, or power transfer for induction heater circuits.

    1. Indeed. The flyback transformer is also modified with a custom primary winding, so all that it’s really used for is the nice potted secondary and the beefy ferrite core.

      Another modification that’s necessary is to remove the air gap in the core, by disassembling the ferrite pieces, and removing any kind of spacer that’s in between. If the ferrite core cannot be removed, it’s not going to be very efficient when used in push-pull configuration.

      1. It’s interesting to note that a flyback transformer will not work in its intended manner without the gap. A few years ago I got curious and looked up the reason on Google. The best article I found involved some simple formulas about inductors and the use of calculus. I was happy that I remembered my college calculus well enough to understand every step of the calculation and agree that it was correct. Sadly, this left me with no intuitive feeling as to why this was the case. I’d share the URL, but that was some years ago, and I didn’t save it anywhere.

        1. FWIW, the best “intuitive” description of the situation seems to be that without the gap, the core becomes saturated, and that limits the energy storage. But when the gap prevents this and why you can store more energy in the magnetic field because of this just doesn’t seem obvious intuitively. It was quite clear in the math IU found though.

        2. The main reason for an air gap is the DC component introduced by a single-ended driver or half-wave rectification that shifts the working point on the magnetic characteristics curve to an unfavorable position and thus reduces the useable magnetic AC amplitude. It’s basically the same problem as with the output transformer of a single-ended tube amplifier. Choosing the correct air gap is a trade-off between core size, core losses and distortion.
          Many LOPTs do not have a visible air gap or removable spacers, the air gap is distributed in the iron powder or ferrite core.

  5. Being zapped by a high voltage tv output as a kid (don’t calibrate your tv without the casing on), and still being in this universe, I find this fascinating to read. thanks for the article :)

    1. It’s a rite of passage. The tube and associated components isn’t really what was likely to kill you; more often it was when somebody touched the mains side of the power supply. The tube anode discharge would certainly scare the hell out of you with a 3-inch lightning bolt that sneaks out and grabs you by the hand, but it wasn’t very likely to stop your heart.

      1. It’s when the back of your hand hits the antenna terminals mount hanging down from the top or the razor edge of the HV cage. It’s not the shock but the reflex you do and what’s behind your hand or elbow.

    2. Worked on a video display terminal as my first job out of college. Rule #1 was to discharge the tube after opening the case. Screwdriver with a clip lead to ground and an insulated handle, blade shoved under the anode cap until SNAP! and then it was safe. The sound alone inspired the necessary respect.

        1. Not exactly building up a charge again, but close enough: the 2nd anode of a television CRT consists of a coating of “Aquadag”, a conductive paint, on the inside of the bell of the tube. This is conductive all right, but not so highly conductive that you discharge the capacitance in an instant. Let’s say it’s a 10kV CRT. You reach in with a grounded wire to discharge it, and !!!***ZAP***!!! you get a lightning bolt about an inch long. So far so good, but you haven’t discharged it completely; you’ve only discharged that local part of the capacitance near the terminal, where the voltage at that point on the Aquadag has decreased to below what it takes to sustain that spark. But that doesn’t take long, and the REST of the inner surface of the CRT is still up around 8kV. Enough to do this a few more times.

          I don’t know if this is why they chose a high-resistance Aquadag, but the reason we didn’t have inquisitive kids falling over dead every day, was that due to this resistance, the energy available at the anode terminal was limited to what was stored close to that terminal, with the rest current-limited by the resistance.

          1. It’s not only CRTs that rebuild charge.. all capacitors do it.. look up dielectric absorbtion.. electrolytics are especially dangerous.. for HV stuff I’d never assume one short will do it.. nearly always takes two.. I find that a fast short discharge (e.g. the screwdriver) on an electrolytic is never as good as a slow bleed with a resistor..

  6. Glad to see one of my pet peeves being shared! I worked as a TV technician in the ’70s and ’80s and am quite familiar with flyback circuits. Another interesting point is that this is quite similar to the old Kettering ignition system used by gasoline engines.

  7. The self oscillating converter doesnt have to be running transistors in linear mode. The schematic above is very similar to a Mozilli resonant converter – this is operating the mosfets in saturation, and is a resonant converter, so generates a sinewave output – this also means that the mosfets are by definition switching at the point on the waveform where it is passing through zero, which means that the brief moment when the mosfet is transitioning through the linear region has very little power, resulting in very little dissipation. Subsequently, these converters are very efficient. I have built one using a different transformer, which generates 100W output at 4kV, the mosfets dont even get warm. If you have the mosfets stuck in linear mode then they just arent being driven hard enough

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.