My first flyback, Help!

Changing the turns ratio to 2:1 works; make L2 88 uH. But then you could use the original circuit!

That IC operates in peak-current mode, so like to run discontinuous.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics
Reply to
John Larkin
Loading thread data ...

Thanks I had almost everything backwards, but I think I have it straightened out now. Say, do you have a fav fast HV diode? As usual I'm overwhelmed by the choices at DK. They have a butt load of these,

formatting link

600V trr = 35 ns

GH

Reply to
George Herold

For low currents, I like MMBD5004S-7. It's a SOT-23 dual series diode,

400 volts per diode, 800 in series. It's nice and fast.
--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com
Reply to
John Larkin

t

tion in

-ND/1642650

Thanks, Totally unrelated but this bit by Kurt Vonnegut was in the local paper yesterday. (It's still snowing here!)

Kurt spent some time in upstate NY. (Cornell)

?One sort of optional thing you might do is to realize that there a re six seasons instead of four. The poetry of four seasons is all wrong for this part of the planet, and this may explain why we are so depressed so m uch of the time. I mean, spring doesn?t feel like spring a lot of t he time, and November is all wrong for autumn, and so on.

Here is the truth about the seasons: Spring is May and June. What could be springier than May and June? Summer is July and August. Really hot, right? Autumn is September and October. See the pumpkins? Smell those burning leav es? Next comes the season called Locking. November and December aren? ?t winter. They?re Locking. Next comes winter, January and Februa ry. Boy! Are they ever cold!

What comes next? Not spring. ?Unlocking? comes next. What e lse could cruel March and only slightly less cruel April be? March and Apri l are not spring. They?re Unlocking.?

I'm sure waiting for the end of Unlocking.

George H.

Reply to
George Herold

Those are a good standby, SMT UF400x's, give or take. They do have more capacitance than you might want/need.

For small stuff, you might like:

formatting link
, though t_rr is kind of embarassing.

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
Website: https://www.seventransistorlabs.com/
Reply to
Tim Williams

Thanks... my DK order is already placed :^)

I guess if the capacitance is a problem I can put some in series... what's another volt or two?

Say do you use film caps for HV? That's what I ordered anyway.

George h.

Reply to
George Herold

Sort of. Then you need matched recovery. Otherwise, whichever one slides open first, gets all the voltage, avalances and gets really hot, which slows it down more, making it even hotter...

Monolithic pairs would be very good, but I don't think you can find those at all, outside of RF (pairs, bridges and DBMs). Matched dies, maybe? There are matched-die BJTs, dunno about diodes.

JL's experience with the MMBD5004S has been good enough, at least.

Yes. In fact, film caps are very much equivalent to high voltage aluminum polymers!

And C0Gs, if you're a big spender, or need any of various attributes they have that films don't.

X7Rs are just about useless at any kind of voltage -- 500V+ C0Gs actually have more capacitance per buck, at rated voltage. I just checked today.

Electrolytics are fine when you need more, given their limitations. You'd probably not need or want them on something this small, of course.

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
Website: https://www.seventransistorlabs.com/
Reply to
Tim Williams

Yes, and that's what the higher-voltage UF400x are; or, maybe there's some other reason that the forward voltage jumps from about a volt to about two volts, at UF4005

Reply to
whit3rd

What's worrying about the data sheet on the VPH5-0155-R inductor is that th ere's nothing about the parallel capacitance of the individual windings.

It's easy enough to measure - you just measure complex impedance as a funct ion of frequency and find the self-resonant frequency (which proper data sh eets tend to include). For wound multi-layer windings 100pF tends to be in the right ball park.

For the 22.3uH of the VPH5-0155-R that would give about 300kHz.

Given six windings, my inclination would have been to use two of them to ma ke a centre-tapped Baxandall and stack up the other four to get six times ( actually 6.283 times) the driving voltage. With a Cockroft-Walton voltage m ultiplier you could get your 300V out of a 24V supply.

--
Bill Sloman, Sydney
Reply to
bill.sloman

Nah, that's normal. Smaller diodes come up to 1.5 or 2kV; larger, slower diodes (including industrial hockey pucks) come up to 6.6kV (and even more in SiC). The downside is higher Vf (up to I think ~5V for the latter kind) and really long recovery time (two microseconds is "fast recovery", when it's holding off /that/ much voltage).

The small, high speed (actually fast, not just relatively so) high voltage diodes, like you'll find in a CRT's FBT, are stacks of small area, 1kV or so, dies. They have voltage drops like 3V, 6V and up. Also anything in the upper kV and low amps, like commercial plate supply rectifier stacks.

The trick is this: look at the knee of the V(I) curve. Fast diodes have higher ESR and N, giving them softer knees and higher voltage drops /at high currents/. The voltage drop at low currents is still quite modest. Like, at random here I've got an FMPG2F damper diode, which measures 0.42V on diode test (a bit under 1mA). It'd be a really weird stack to still measure that little voltage, at that current, over multiple dies in series. :)

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
Website: https://www.seventransistorlabs.com/
Reply to
Tim Williams
[about stacked rectifiers for lower capacitance]

It might be normal, but there's more than one datasheet for UF400x parts, and this one

calls the UF4005 a 20pF diode, and the UF4006 a 10pF diode...it seems like a two-pellet stack to me.

Reply to
whit3rd

Nah, just doping difference. 4001-4005 are specified as identical and they just don't care. And 4006s are 4007s.

Think that's also the borderline between regular-diffused and PIN diodes. Much longer junction = lower capacitance.

Back in the day you might've had more actual grades, more curves on the datasheet. Or, hmm, actually, back in the day, you wouldn't, all you got was a data table!

Anyway, you're welcome to ignore the characteristics and physics; you might not want to advertise to others that you're of such a habit though. :)

Cheers, Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
Website: https://www.seventransistorlabs.com/
Reply to
Tim Williams

there's nothing about the parallel capacitance of the individual windings.

ction of frequency and find the self-resonant frequency (which proper data sheets tend to include). For wound multi-layer windings 100pF tends to be i n the right ball park.

make a centre-tapped Baxandall and stack up the other four to get six times (actually 6.283 times) the driving voltage. With a Cockroft-Walton voltage multiplier you could get your 300V out of a 24V supply.

Right the spec sheet is a bit thin. This fly back looks OK, a few hundred mV of ripple on the output. I'll put another filter on that... I'm thinking a simple RC. I'll post 'scope shots once it's built. (I'll most likely run into more 'issues'.)

George H.

Reply to
George Herold

Right the spec sheet is a bit thin. This fly back looks OK, a few hundred mV of ripple on the output. I'll put another filter on that... I'm thinking a simple RC. I'll post 'scope shots once it's built. (I'll most likely run into more 'issues'.)

You'll find it's much worse than that, spikey due to maybe 5nH (or worse if build poorly) of ESL in the capacitor. An LC filter after is required to get under a volt.

Do take measures to ensure the input and output bypasses share the same ground (without switching currents routed between them). This prevents the spikes appearing in the common mode as well.

Or if nothing else, shove everything into a grounded metal box with LC filters on the two signals.

Simple things, but if you don't understand how it happens, you'll be pulling your hair out (or what's left?) trying to fix it. :)

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
Website: https://www.seventransistorlabs.com/
Reply to
Tim Williams

Isn't he the one who wrote "Chariots of the Gods"?

Reply to
John S

ls

e

ent

e
d

uation in

e

SCT-ND/1642650

l

re are six seasons instead of four. The poetry of four seasons is all wrong for this part of the planet, and this may explain why we are so depressed so much of the time. I mean, spring doesn?t feel like spring a lot of the time, and November is all wrong for autumn, and so on.

be springier than May and June? Summer is July and August. Really hot, rig ht? Autumn is September and October. See the pumpkins? Smell those burning leaves? Next comes the season called Locking. November and December aren? ??t winter. They?re Locking. Next comes winter, January and Feb ruary. Boy! Are they ever cold!

at else could cruel March and only slightly less cruel April be? March and April are not spring. They?re Unlocking.?

I don't think so, Slaughter House Five, Cat's Cradle, Sirens of Titian, Player Piano... OK just look at wiki. His later stuff was OK.

George H.

Reply to
George Herold

Looking this circuit would be much nicer with scroll bars. But my LTspice XVII has none

Reply to
LM

You just drag the interior of the schematic pane and it moves wherever you like.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC / Hobbs ElectroOptics 
Optics, Electro-optics, Photonics, Analog Electronics 
Briarcliff Manor NY 10510 

http://electrooptical.net 
http://hobbs-eo.com
Reply to
Phil Hobbs

yeah, or zoom out of where you don't want to be and zoom into where you want to go. (point the mouse and use the wheel)

--
     ?
Reply to
Jasen Betts

Pretty tricky circuit.

Why not put all that text in an external file?

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com
Reply to
John Larkin

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.