BUP49 transistor - Max frequency? / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

BUP49 transistor - Max frequency?

first previous 2 3 4 5

Move Thread

LAN_403

Patrick

Sun Apr 17 2011, 08:13PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Steve McConner wrote ...

My point is, why make life harder by using BJTs. They really suck for switched mode.

I have to agree, why not use todays transistors? I did use some bipolars for a smaller power (500 watt) supply, back in 1998-ish, I would not do it again.

James

Mon Apr 18 2011, 01:30AM

Registered Member #3610 Joined: Thu Jan 13 2011, 03:29AM
Location: Seattle, WA
Posts: 506

The "Mazzilli" driver is a push-pull driver. Technically it would be called a current-fed push-pull Royer oscillator but this gets a bit long winded. The reason it's often called a "flyback" driver is because it is often used to drive converted flyback transformers.

It is possible to put some serious power through variants of the circuit. I just recently finished building one of these

which if you look at the schematic, looks very much like the Mazzilli driver.

Every flyback topology converter I've seen has used a gapped core, including the small wall warts. You could probably get by with a powdered iron core which is inherently gapped, but ungapped ferrite will saturate, DC in the core is inherent to that topology.

I have yet to see a flyback topology ATX power supply. It's frequently used in the standby +5V portion, but the main power supply has been a push-pull design based on a TL494 or similar PWM IC in every one that I've opened up. The only flyback PC power supplies I've come across were the PC/XT era stuff back in the early 80s, and those were mostly under 100W. CRT monitors and TVs are rarely much over 100W. I used to work on a lot of those and even by 27" or so most have gone to push-pull driven power supplies and those are only 120W or so. I still have an old 56" HD rear projection TV and it draws about 200W according to my kill a watt. My 22" CRT computer monitor ranges from 80-120W depending on what's on the screen and that's about as big and power hungry as PC monitors get. IIRC it does use a flyback type PSU but it's been a long time since I've been in it.

Arcstarter

Mon Apr 18 2011, 02:57AM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

James wrote ...

I have yet to see a flyback topology ATX power supply. It's frequently used in the standby +5V portion, but the main power supply has been a push-pull design based on a TL494 or similar PWM IC in every one that I've opened up. The only flyback PC power supplies I've come across were the PC/XT era stuff back in the early 80s, and those were mostly under 100W. CRT monitors and TVs are rarely much over 100W. I used to work on a lot of those and even by 27" or so most have gone to push-pull driven power supplies and those are only 120W or so. I still have an old 56" HD rear projection TV and it draws about 200W according to my kill a watt. My 22" CRT computer monitor ranges from 80-120W depending on what's on the screen and that's about as big and power hungry as PC monitors get. IIRC it does use a flyback type PSU but it's been a long time since I've been in it.

I have about 10 ATX PSUs, all of which are well over 100 watts, and at least half are flyback. Those which are flyback use a mosfet around 800v and 9 amps usually, it would seem. I do have some that use push-pull topology, and at least a few seem to use a couple of BJTs.

Most ATX PSUs i have either use some variation of the TL494 or a uc384x. The uc384x series is made for flyback topology.

James

Mon Apr 18 2011, 03:11AM

Registered Member #3610 Joined: Thu Jan 13 2011, 03:29AM
Location: Seattle, WA
Posts: 506

Hmm, well ok so it must be more common than I thought, I'm familiar with the UC384x series and they are indeed made for flyback supplies. Still, ~100W is about the point where it starts to make sense to use something else. Not saying one couldn't make a 10kW flyback power supply if they wanted, you could modify a car to run on Bacardi 151 too, but why would you when other options are cheaper, easier and more efficient?

Steve Conner

Mon Apr 18 2011, 08:59AM

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

Yes, we had a debate on this a while ago. The textbooks say that flyback converters aren't worth the hassle above a few hundred watts.

The reason is that the energy stored in the transformer's leakage inductance is lost at every switching event, and this goes up as the size of the transformer and the square of the primary current. So the overall misery level in the designer's life goes up as the cube of the rated power.

The wasted energy can be recycled by active snubbers, but the complexity and added cost of these is similar to just going for a bridge-driven forward converter. Now, a forward converter needs two magnetics: a transformer and a filter choke. But the combined size and cost of them is probably about the same as a flyback transformer of the same rating.

Now, on to TV flybacks.

As the name would suggest they are designed to be driven as a flyback converter: in fact that is how the flyback converter got its name. To make the electron beam "fly back" quickly to the left-hand side of the screen, the current in the deflection coils had to be ramped down quickly.

Suddenly changing the current in an inductor generates a very high voltage according to Faraday's law of induction, and sometime in the 1950s, TV designers had the idea of harnessing this "flyback pulse" to generate the picture tube anode voltage, instead of wastefully snubbing it.

TVs drive the flyback from a ~120V rail with a 1200V rated transistor. (In late-era CRT TVs and monitors, this 120V DC came from a separate SMPS, which can be a flyback too, or a forward converter. I've seen both.)

Originally, the stored energy was mostly in the deflection coils, and the flyback transformer was connected in parallel with them and wouldn't do much on its own. But I've worked on some high-end monitors where the flyback transformer stores all its own energy, and the deflection coils are driven by a separate circuit, all running off an off-line SMPS, making three converters in total.

But somehow TV flybacks work anyway, when driven from a Mazzilli driver, which is a forward converter. I think they are behaving like Tesla coils, with the voltage step-up by resonant rise in the secondary. Or maybe the lower voltage, higher current output is better for drawing the hot arcs that HV hobbyists love.

James

Mon Apr 18 2011, 05:06PM

Registered Member #3610 Joined: Thu Jan 13 2011, 03:29AM
Location: Seattle, WA
Posts: 506

Steve McConner wrote ...

Yes, we had a debate on this a while ago. The textbooks say that flyback converters aren't worth the hassle above a few hundred watts.

The reason is that the energy stored in the transformer's leakage inductance is lost at every switching event, and this goes up as the size of the transformer and the square of the primary current. So the overall misery level in the designer's life goes up as the cube of the rated power.

The wasted energy can be recycled by active snubbers, but the complexity and added cost of these is similar to just going for a bridge-driven forward converter. Now, a forward converter needs two magnetics: a transformer and a filter choke. But the combined size and cost of them is probably about the same as a flyback transformer of the same rating.

Now, on to TV flybacks.

Great post, I love the equation for power output vs misery level :)

The combined HV and horizontal deflection circuit in TV sets was genius, a masterpiece of analog engineering, bouncing the energy around in a resonant tank, and bleeding some off to provide the anode voltage instead of throwing it away after scanning each line. Virtually every "multiscan" CRT computer monitor uses completely independent HV and deflection circuits. That includes 95% of SVGA CRT monitors made starting from the late 80s.

I kinda miss CRT displays as they rapidly disappear, I always did like working with them.

The voltage output from a flyback transformer / LOPT driven by the Mazzilli driver is much lower than driving it in flyback mode. The power level can be much higher though resulting in a big hot arc. The output in flyback mode tends to look more like a small Tesla coil, pulses of very high amplitude and thin crackly blue sparks. The high current output from the Mazzilli push-pull drive produces a very hot firey arc resembling that of a large NST and can approach that of a MOT for short runs. The voltage is lower, but the current is amazing, limited mostly by the DC resistance of the secondary winding. The attached picture is a ~5 second exposure of a Jacobs Ladder using a SVGA monitor flyback / LOPT driven push-pull by a "Mazzilli" ZVS circuit. Input was about 42V with a draw of 4-8 Amps depending on arc length.

first previous 2 3 4 5

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