Flyback vs half-bridge

J

John Larkin 13 years ago

Spice voltage sources are infinitely powerful; wall warts aren't. Lots of switchers have soft-start, too.

John Larkin Highland Technology Inc www.highlandtechnology.com jlarkin at highlandtechnology dot com Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom timing and laser controllers Photonics and fiberoptic TTL data links VME analog, thermocouple, LVDT, synchro, tachometer Multichannel arbitrary waveform generators

Vote

B

Bill Sloman 13 years ago

n

ut

t

I've got a copy of the second edition ISBN 0-07-052236-7

The chapter on magnetics is comprehensive, and even covers the skin effect, and the "proximity effect" which I've never heard of before. Pedagogically , it's rubbish - almost as bad as Snelling. The author assumes that the rea der understands the basics, and concentrates on the complicated detail.

Bill Sloman, Sydney

Vote

P

Phil Hobbs 13 years ago

I think the torture-test is powering up the wall wart and then plugging it into the box. There'll be the usual polyfuse, whose cold resistance helps a bit, and I suppose I could do something fancier like a resistor parallelled by a MOSFET controlled by the POR chip, but that's way too fancy for just a few watts.

It's really the 1N4148s or BAV99s or whatever little diode I use that will be the issue--the turn-on surge will divide between the main -15V reservoir cap and the charge pump caps more or less proportionally with the size of the caps (neglecting ESR and diode drops). So I may use a much bigger reservoir cap and smaller charge pump caps, followed by RC filters ahead of the cap multipliers. Alternatively, just putting 50 or

100 ohms in series with the charge pumps will protect them OK, and will only cost a volt or two.

It's better from a certification standpoint if the voltages inside the box never exceed 60V even in fault conditions, so I'll need to put zeners on the +-45V outputs, so that even with an out-of-spec input voltage and the transzorb at its upper spec limit, I don't get over 60V.

Cheers

Phil Hobbs

Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net

Vote

J

Joerg 13 years ago

That's what most books about advanced technologies do. If they didn't then most of the forests on this planet would be depleted by now :-)

But seriously, a person with lacking EE basics should not be designing switchers without first becoming proficient enough, that's something the pros should do.

Regards, Joerg http://www.analogconsultants.com/

Vote

J

Joerg 13 years ago

That's a buck. If you already have a -15V switcher in there, why not hook into that as you had outlined earlier?

Regards, Joerg http://www.analogconsultants.com/

Vote

P

Phil Hobbs 13 years ago

I'm using it as an inverter, i.e. grounding the far end of the inductor and letting it pump its ground pin down to -15V. Works fine.

Cheers

Phil Hobbs

Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net

Vote

J

Jim Thompson 13 years ago

That's what I was noticing... it's a switch current limit, not an output current limit. But probably safe. ...Jim Thompson

Vote

J

Jim Thompson 13 years ago

Yup. It's the inrush that tests your specifications ;-) ...Jim Thompson

Vote

J

Joerg 13 years ago

Yes, that's the old MC34063 method. But watch out for noise and loop stability.

Regards, Joerg http://www.analogconsultants.com/

Vote

J

Jim Thompson 13 years ago

You might want to reconsider and add soft-start to your "cartoon"....

Subject: Hobb's Cartoon (SED) - HobbsCartoon-with_CL.png Date: Thu, 18 Jul 2013 11:19:26 -0700 Message-ID:

Using the LM2594, you're sort of caught between a rock and a hard-place... large load (on LM2594) capacitor, huge peak currents on your add-on inverter, smaller capacitor... sag your 16V. ...Jim Thompson

Vote

P

Phil Hobbs 13 years ago

Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 USA +1 845 480 2058 hobbs at electrooptical dot net http://electrooptical.net --------------040407090803080109080400 Content-Type: text/plain; charset=windows-1252; name="BuckTripler4.asc" Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="BuckTripler4.asc" Version 4 SHEET 1 1656 900 WIRE -784 48 -832 48 WIRE -656 48 -704 48 WIRE -624 48 -656 48 WIRE -576 48 -624 48 WIRE -368 48 -576 48 WIRE -288 48 -368 48 WIRE -64 48 -192 48 WIRE 80 48 -64 48 WIRE 176 48 80 48 WIRE 256 48 176 48 WIRE 272 48 256 48 WIRE 368 48 352 48 WIRE 448 48 432 48 WIRE 496 48 448 48 WIRE 656 48 560 48 WIRE 784 48 656 48 WIRE 800 48 784 48 WIRE 448 80 448 48 WIRE -832 96 -832 48 WIRE 656 96 656 48 WIRE 800 112 800 48 WIRE -368 128 -368 48 WIRE -272 128 -272 96 WIRE -656 144 -656 112 WIRE -64 144 -64 112 WIRE 160 144 -64 144 WIRE 336 144 160 144 WIRE 448 144 416 144 WIRE 80 176 80 48 WIRE 656 208 656 160 WIRE -368 224 -368 208 WIRE -272 224 -272 208 WIRE -272 224 -368 224 WIRE 800 224 800 192 WIRE -832 240 -832 176 WIRE 256 240 256 48 WIRE 272 240 256 240 WIRE 400 240 336 240 WIRE 400 256 400 240 WIRE -64 272 -64 144 WIRE 0 272 -64 272 WIRE 80 272 80 256 WIRE 80 272 64 272 WIRE 80 304 80 272 WIRE -64 352 -64 272 WIRE -64 352 -128 352 WIRE 400 352 400 336 WIRE 464 352 400 352 WIRE 544 352 464 352 WIRE 656 352 608 352 WIRE 800 352 656 352 WIRE 464 368 464 352 WIRE -128 384 -128 352 WIRE -64 384 -64 352 WIRE 656 400 656 352 WIRE 800 400 800 352 WIRE -800 416 -864 416 WIRE -672 416 -720 416 WIRE -864 432 -864 416 WIRE -672 448 -672 416 WIRE 464 448 464 432 WIRE -128 496 -128 448 WIRE -64 496 -64 464 WIRE -64 496 -128 496 WIRE -64 512 -64 496 WIRE 656 512 656 464 WIRE 800 528 800 480 WIRE -864 544 -864 512 WIRE -672 544 -672 512 WIRE -576 560 -576 48 WIRE 176 560 -576 560 WIRE 464 560 464 528 WIRE 464 560 176 560 FLAG 656 208 0 FLAG -832 240 0 FLAG 784 48 Out- FLAG 656 512 0 FLAG 800 352 Out+ FLAG 80 304 0 FLAG -64 512 0 FLAG 176 48 Switch FLAG 160 144 -15 FLAG 800 528 0 FLAG 800 224 0 FLAG -656 144 0 FLAG 176 560 Vin FLAG -832 48 Vwart FLAG -624 48 Vin FLAG -864 544 0 FLAG -672 544 0 FLAG -672 416 Eff SYMBOL ind2 64 160 R0 WINDOW 0 50 51 Left 2 WINDOW 3 60 86 Left 2 SYMATTR InstName L1 SYMATTR Value 150u SYMATTR Type ind SYMATTR SpiceLine Rser=0.2 Cpar=20p SYMBOL cap 640 96 R0 SYMATTR InstName C1 SYMATTR Value 1u SYMBOL diode 560 64 M270 WINDOW 0 42 32 VTop 2 WINDOW 3 -3 -7 VBottom 2 SYMATTR InstName D1 SYMATTR Value BAV99 SYMBOL voltage -832 80 R0 WINDOW 0 40 54 Left 2 WINDOW 3 -109 209 Left 2 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V1 SYMATTR Value PULSE(0 {Vraw} 100n 10n 10n 1) SYMBOL cap 640 400 R0 SYMATTR InstName C2 SYMATTR Value 1u SYMBOL diode 544 368 R270 WINDOW 0 32 62 VTop 2 WINDOW 3 -7 64 VBottom 2 SYMATTR InstName D2 SYMATTR Value BAV99 SYMBOL diode 448 432 M180 WINDOW 0 55 41 Left 2 WINDOW 3 39 0 Left 2 SYMATTR InstName D3 SYMATTR Value BAV99 SYMBOL diode 432 80 R0 SYMATTR InstName D4 SYMATTR Value BAV99 SYMBOL cap 432 32 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C5 SYMATTR Value 100n SYMBOL cap 336 224 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C6 SYMATTR Value 100n SYMBOL cap 0 256 M90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C4 SYMATTR Value 100u SYMBOL schottky -80 112 M180 WINDOW 0 47 33 Left 2 WINDOW 3 24 0 Left 2 SYMATTR InstName D5 SYMATTR Value 1N5819 SYMATTR Description Diode SYMATTR Type diode SYMBOL res -80 368 R0 WINDOW 3 53 81 Left 2 SYMATTR Value {Load15} SYMATTR InstName R2 SYMBOL pmos -192 96 M270 WINDOW 0 72 72 VLeft 2 WINDOW 3 -17 64 VLeft 2 SYMATTR InstName M1 SYMATTR Value Si9407AEY SYMBOL voltage -368 224 R180 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 WINDOW 3 -269 -26 Left 2 SYMATTR Value PULSE(0 -10 .5u 10n 10n {6.6u*10.3/(8+Vraw)} 6.6u) SYMATTR InstName V2 SYMBOL res -288 112 R0 SYMATTR InstName R4 SYMATTR Value 100 SYMBOL schottky -144 384 R0 WINDOW 0 -58 28 Left 2 WINDOW 3 -104 94 Left 2 SYMATTR InstName D6 SYMATTR Value 1N5819 SYMATTR Description Diode SYMATTR Type diode SYMBOL res 784 384 R0 SYMATTR InstName R1 SYMATTR Value 4k SYMBOL res 784 96 R0 SYMATTR InstName R5 SYMATTR Value {LoadM45} SYMBOL res 256 64 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 0 56 VBottom 2 SYMATTR InstName R3 SYMATTR Value {Rsoft} SYMBOL res 384 240 R0 SYMATTR InstName R6 SYMATTR Value {Rsoft} SYMBOL res -800 64 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 0 56 VBottom 2 SYMATTR InstName R7 SYMATTR Value 2 SYMBOL cap -672 48 R0 WINDOW 0 -26 61 Left 2 WINDOW 3 -19 87 Left 2 SYMATTR InstName C3 SYMATTR Value 47u SYMBOL res 448 432 R0 WINDOW 0 48 38 Left 2 SYMATTR InstName R8 SYMATTR Value {Rsoft} SYMBOL res 432 128 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R9 SYMATTR Value {Rsoft} SYMBOL bv -864 416 R0 WINDOW 3 -37 169 Left 2 SYMATTR Value V=(V(out+)*I(R1)+V(out-)*I(R5)+V(-15)*I(R2))/(-V(Vin)*I(V1)) SYMATTR InstName B1 SYMBOL res -816 432 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 0 56 VBottom 2 SYMATTR InstName R10 SYMATTR Value 1 SYMBOL cap -688 448 R0 WINDOW 0 -45 34 Left 2 WINDOW 3 -57 70 Left 2 SYMATTR InstName C7 SYMATTR Value 100u TEXT 600 640 Left 2 !.tran 8mm TEXT -120 632 Left 2 !.param Load15=100 TEXT -120 600 Left 2 !.param LoadM45=2.2k TEXT 176 600 Left 2 !.MODEL BAV99 D IS=4.858E-9 N=2.038 RS=1.238 XTI=3 TEXT 280 -32 Left 2 ;Add this part to the -15V switcher TEXT -72 -48 Left 2 ;Switcher TEXT -72 -16 Left 2 ;External TEXT -80 16 Left 2 ;Components TEXT -272 360 Left 2 ;-15V Load TEXT -488 -24 Left 2 ;LM2594 -15V switcher chip TEXT -488 -64 Left 2 ;Toy model of the TEXT 736 0 Left 2 ;-43-47V TEXT 744 304 Left 2 ;+44-52V TEXT -304 -144 Left 2 ;ADDING +-40V UNREGULATED SUPPLIES TO THE -15V SWITCHER TEXT 608 -144 Left 2 ;7/18/13 PH TEXT 152 632 Left 2 !.param Rsoft=10 TEXT -504 632 Left 2 !.step param Vraw 16 22 2 RECTANGLE Normal 96 544 -288 336 2 RECTANGLE Normal 112 336 112 336 2 RECTANGLE Normal -96 -64 112 320 2 RECTANGLE Normal -544 -96 -96 272 2 RECTANGLE Normal 720 544 240 -64 2 --------------040407090803080109080400--

Vote

J

Joerg 13 years ago

[...]

Not sure what you did for this post but it came through as a file attachment. No copy - paste, just click it and open into LTSpice. Cool.

Alternatively you might be able to provide an over-voltage lockout. TL431 -> input goes above 18V -> pulls current -> robs current from enabling circuit that serves pin 5 -> switcher goes silent. Pages 29/30 in the datasheet:

formatting link

[...]

Regards, Joerg http://www.analogconsultants.com/

Vote

J

Joerg 13 years ago

P.S.: Didn't you mention you wanted to place cap multipliers at the 45V outputs? Then you could hang zeners off of the bases, which would limit things.

Regards, Joerg http://www.analogconsultants.com/

Vote

J

John Larkin 13 years ago

In my boost-doubler, I just replaced a BAV23S with a couple BAT46W's, and picked up a bunch of efficiency.

In a 9-to-48 volt boost converter, I was seeing over 100 volt spikes at the fet drain, blamed on forward conduction delay in the PN diodes. The BAT schottkies clamp hard at 48.

John Larkin Highland Technology, Inc jlarkin at highlandtechnology dot com http://www.highlandtechnology.com Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom laser drivers and controllers Photonics and fiberoptic TTL data links VME thermocouple, LVDT, synchro acquisition and simulation

Vote

P

Phil Hobbs 13 years ago

I used to do that on Supernews, but they now reject posts with attachments, even text. I got an eternal-september.org account as well, and they apparently allow attachments. Good news.

That's pretty sensible. I'm planning to have the processor control the

+-15V and HV supplies, but a bit of (inexpensive) belt and suspenders never hurts.

Cheers

Phil Hobbs

Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 USA +1 845 480 2058 hobbs at electrooptical dot net http://electrooptical.net

Vote

P

Phil Hobbs 13 years ago

Nice part--like a 100V BAT54. That's a keeper.

Cheers

Phil Hobbs

Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 USA +1 845 480 2058 hobbs at electrooptical dot net http://electrooptical.net

Vote

H

Harry D 13 years ago

oK, Here is a duh question, When in LTSPICE, how do you save that top level *.asc file you show above?

Thanks, Harry

Vote

J

Jim Thompson 13 years ago

I don't think a wall-wart is going to look like 2 Ohms... it's going to be current limited... even if it is made in China ;-) ...Jim Thompson

Vote

P

Phil Hobbs 13 years ago

There are a couple of methods.

(BTW you should get a real newsreader and a (free) Eternal September account--lots of people round here killfile anyone from Google Groups because of the spam and the zillions of blank lines.

formatting link

).

If you have an eternal-september.org account, you can just use your mailer and attach the .asc file as you would any other file.

Otherwise, open the .asc file in a text editor, and cut-n-paste it into your post.

Mike Engelhardt understands the goodness of human-readable files!

Cheers

Phil Hobbs

Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 USA +1 845 480 2058 hobbs at electrooptical dot net http://electrooptical.net

Vote

P

Phil Hobbs 13 years ago

The 2 ohms is the initial resistance of the polyswitch in the input protection network. Quite a lot of computer wall warts have a big cap at the output, leading to crackling sounds when you plug them into the back of the machine. Anyway, their current limits will be up in the 5A range, which is way above what my little switcher will allow through on a steady state basis.

Cheers

Phil Hobbs

Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 USA +1 845 480 2058 hobbs at electrooptical dot net http://electrooptical.net

Vote

Flyback vs half-bridge

Join the Discussion

Didn't find your answer?