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 next

Move Thread

LAN_403

Ash Small

Sat Apr 16 2011, 03:02AM

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

James wrote ...

As far as I know, the transformer HAS to be gapped for flyback converter to work. Increasing the gap reduces the permeability of the core and increases leakage inductance.

Here's some info you might find useful.

Thanks, James, I'll read through it over the weekend.

(BUX348 might be the best BJT to use, 450 V (max 850), 45 Amp. Not sure what the gain and max frequency are though.)

Patrick

Sat Apr 16 2011, 04:54AM

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

Ash Small wrote ...

I doubt I'll get 90 Amps in.....I want SOME safety margin for the transistor.....Can I avoid saturation by gapping the cores?

well yes you need a gap and youll need to consider your volts per turn, at that means some math.

If im not mistaken there are two flyback types. The flybacks like in TV's were all familar with. and the generic topology, like wall wart smps cell phone chargers that dont use a gap.

Steve Conner

Sat Apr 16 2011, 07:51AM

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

Ash, you don't come across as confrontational, so much as the classic Internet weenie with a Tsar Bomba sized opinion on a subject of which he has no practical experience.

In practice it would work better and cheaper with a half or full bridge of MOSFETs running off 48V at least. Low voltage helps keep the magic smoke in, but too low makes it hard to generate lots of power. 24V is the absolute minimum worth bothering with for these kinds of circuits as far as I'm concerned. I think you can still get the HUF75xxx "UltraFets" from Maplin. 55V, 75A or something. But for 1kW I'd just go straight to a full bridge of IRFP460 type devices running off 160 or 320V.

If you want to know how fast a BJT will go, you need to find the "storage time" on the datasheet. That is the limiting factor. If the datasheet doesn't give it, then the device isn't meant for switching use at all.

Ash Small

Sat Apr 16 2011, 11:59AM

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

Patrick wrote ...

.
If im not mistaken there are two flyback types. The flybacks like in TV's were all familar with. and the generic topology, like wall wart smps cell phone chargers that dont use a gap.

I was under the impression that there is only one basic flyback topology. I understood there were two factors defining whether the cores need to be gapped, the permeability of the core (different core materials) and the power (whether or not the core is saturated). Also, while it is easier to just separate the core halves and have two gaps, this is not ideal, one gap (ground cores) performs better, although the disadvantages of two gaps can be reduced by adding an extra 'single turn winding' usually comprising of copper foil to counteract the effects of stray inductance.

Regarding your point on maths, volts per turn, etc. I'm not averse to doing the maths when required, but before you do the maths you need to understand what you are doing and the limitations. Someone here has a signature something like 'In theory, theory and practice are the same, but in practice they are different'. the maths is only ever an approximation, as anyone who has, for example, calculated the windings for an inductor will verify.

Also, and I'm adressing a point Steve made here, while a different topology may be more efficient (push-pull, H-bridge, etc) this would require a greater number of turns on the secondary, thus increasing the capacitance of the secondary, resulting in more ringing, etc (at least that's what I understand from recent posts here) which can have adverse effects.

Nothing is ever 'black or white', (2 + 2 = ~4)

H-bridges also require isolated gate drivers (I'm sure I'll build an H-bridge at some point)

They must have used BJT's before FET's became available, saying BJT's have no place here is like saying vacuum tubes (valves) have no place here either (am I being confrontational, Steve?

)

I also agree with your point that I'm inexperienced with these things. Surely it's better to start with BJT's and flybacks, rather than jumping straight in with FET's and H-bridges?

Ash Small

Sat Apr 16 2011, 03:43PM

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

Steve McConner wrote ...

.

If you want to know how fast a BJT will go, you need to find the "storage time" on the datasheet. That is the limiting factor. If the datasheet doesn't give it, then the device isn't meant for switching use at all.

Looks like the BUX348 is the best BJT I can find, storage time 2.75 uS Typ, 4.5uS Max, 45 Amp (60 Amp Max) and 450 V. Not sure what the gain is though.

Maybe I'll get a BUX348 and a BUP49 and see which one blows first?

Patrick

Sat Apr 16 2011, 05:32PM

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

Steve McConner wrote ...

In practice it would work better and cheaper with a half or full bridge of MOSFETs running off 48V at least. Low voltage helps keep the magic smoke in, but too low makes it hard to generate lots of power. 24V is the absolute minimum worth bothering with for these kinds of circuits as far as I'm concerned. I think you can still get the HUF75xxx "UltraFets" from Maplin. 55V, 75A or something. But for 1kW I'd just go straight to a full bridge of IRFP460 type devices running off 160 or 320V.

Well I was going to say it if Steve didnt, but I think your going to be having trouble with a 1kW flyback, I would use a normal transformer and a normal h-bridge. At high power (above 300 watts) i dont see the advatage in storing power inside the core as a flyback does. All other transformers transfer but do not store the power in the same way.

Arcstarter

Sat Apr 16 2011, 07:20PM

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

Making a 1kw flyback converter will be much more difficult than making a simple half or fullbridge. You aren't taking anyone's advice, not even Steve Conner whom is considered one of the Tesla coil gurus, so why even bother asking?

It is a very simple topology, but as sizes get much over 100W things begin to get tricky

How tricky do you mean, James? ATX PSUs are often flyback, and they are in the multiple hundred watts, and the circuit is still very basic. Also, CRT TV SMPSs are a couple hundred watts or thereabouts, and they are usually flyback (i have only seen halfbridges in other, newer TVs).

Sure, it is best to start small with something like flyback topology, but starting with BJTs makes no sense... They are no more complex than a mosfet when used in a circuit. BJTs *don't* have a place here unless it is proof of concept or if it's all you have. Go take apart a CRT TV and check out the SMPS unit. They are almost always based on a flyback topology, and they almost always use mosfets.

If you don't like the idea of making a GDT, which is much simpler than you are making it sound, you could use a *very* simple halfbridge driver IC.

It only requires one power supply. I don't mean to be mean when i say this, but if you refuse to take my advice, i don't see the point in trying to help.

Ash Small

Sat Apr 16 2011, 08:04PM

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

Steve and Patrick, I realise an H-bridge would be the best option for most power, also that push-pull has advantages over flyback. I worked through all the theory back in the winter.

I realise rectified mains is the ideal solution here and I'm currently building a batch of heatsinks, some for a mains rectifier. I've also started collecting 450 V 1000 uF electrolytics. I also need to pick up some more stuff I've got in storage a couple of hundred miles away. (toroids for GDT's, assorted MOSFET's, etc.)

I've two sets of those cores, enough for 5 kW in push-pull mode, but I need at least 8 larege heatsinks for a full bridge, maybe more (4 for the bridge rectifier, 4 for the MOSFET's or IGBT's). You'll see why these are taking so long to build once I've finished some and posted details here.

In the meantime I'm building some smaller heatsinks using only handtools as I no longer have the facilities I had when I built the last batch. These, along with the 90mm fans, are specifically for TO3 packages.

It therefore occured to me to get something like a BUP49, BUX348 or BUV20 and see what I can do with only one heatsink, hence the questions. I'm sure you'll agree that when it comes to power circuits one of the most important considerations, if not the most important consideration, is heatsinks and cooling.

I could go on to say that, in some situations where space is a premium, a circuit that has only one heatsink will have advantages over a circuit that requires 8 or more.

I do have another question, though.

Plenty of peolpe here claim to have put 1.5 kW through the Mazilli ZVS circuit. It is referred to as a ZVS flyback circuit, but does it actually operate in flyback mode or push-pull mode?

Patrick

Sat Apr 16 2011, 08:22PM

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

Ash Small wrote ...

Plenty of peolpe here claim to have put 1.5 kW through the Mazilli ZVS circuit. It is referred to as a ZVS flyback circuit, but does it actually operate in flyback mode or push-pull mode?

Ok, now youve put my back up! I too see these claims all the time. I am highly suspicous, me thinks they are putting 1.5kW into the primary and assuming a high degree of transfer to the secondary. I am betting that when they put 1.5kW into the primary, there only transfering 200-500 watts to the secondary at best, maybe much less that 500 watts. That airgap can change things you know. Also, I wonder if their meters on the input side are seeing more current and voltage then is really there.

Ash Small

Sat Apr 16 2011, 08:34PM

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

But the Mazilli circuit, to me, looks more like a push-pull circuit (centre tapped primary, two transistors....) Does it operate in flyback mode or push-pull?

first previous 2 3 4 5 next

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