Flyback. LTspice. Questions!

Hi group, apologies for jumping straight in with a question but I've been f ollowing this group for a while, learning a bit, and now I'm stuck!

I'm trying to quickly come up with a wide input power supply, 60-280VDC. Tw o low power 12V outputs, one common with the -ve input and one with the +ve .

I have thrown a circuit together on the bench with LTC3805-5, an off the sh elf Wuerth transformer (using their DCM flyback selector tool) and amazingl y it didn't immediately blow up. However it won't work over about 200V inp ut.

I wonder if I'm running out of minimum duty cycle (it is 6% minimum on this IC) and saturating the transformer, hitting the current limit.

The circuit drawn in LTspice is pasted below, and the transformer is this o ne:

formatting link

My question is, how do I visualise the timing of the gate signal properly? And is there an obvious basic flaw with the design anyway - quite possible , this is my first attempt at an SMPS. Any comments welcome [braces]

Version 4 SHEET 1 1216 680 WIRE 160 -448 48 -448 WIRE 256 -448 160 -448 WIRE 480 -448 320 -448 WIRE 160 -432 160 -448 WIRE 48 -416 48 -448 WIRE 480 -416 480 -448 WIRE 48 -304 48 -336 WIRE 160 -304 160 -368 WIRE 160 -304 48 -304 WIRE 416 -304 160 -304 WIRE 480 -304 480 -336 WIRE 480 -304 416 -304 WIRE 448 -240 320 -240 WIRE 1040 -240 528 -240 WIRE 720 -160 592 -160 WIRE 944 -160 784 -160 WIRE 976 -160 944 -160 WIRE 1040 -160 1040 -240 WIRE 1040 -160 976 -160 WIRE 1136 -160 1040 -160 WIRE 48 -128 -256 -128 WIRE 224 -128 48 -128 WIRE 416 -128 416 -304 WIRE 416 -128 224 -128 WIRE 480 -128 416 -128 WIRE 592 -128 592 -160 WIRE 48 -112 48 -128 WIRE 976 -112 976 -160 WIRE 1136 -112 1136 -160 WIRE 1040 -96 1040 -160 WIRE 320 -64 320 -240 WIRE 224 -48 144 -48 WIRE -256 -32 -256 -128 WIRE -64 -32 -80 -32 WIRE 48 -32 16 -32 WIRE 144 -32 144 -48 WIRE 592 -32 592 -48 WIRE -80 0 -80 -32 WIRE 224 32 224 -48 WIRE 320 32 320 0 WIRE 320 32 224 32 WIRE 144 48 144 32 WIRE 480 48 480 -48 WIRE 976 64 976 -48 WIRE 1040 64 1040 -32 WIRE 1136 64 1136 -32 WIRE 224 80 224 32 WIRE -256 96 -256 48 WIRE 944 96 944 -160 WIRE 944 96 656 96 WIRE 48 128 48 -32 WIRE 80 128 48 128 WIRE 432 128 368 128 WIRE 656 208 656 96 WIRE -64 208 -80 208 WIRE 80 208 0 208 WIRE 384 208 368 208 WIRE 480 208 480 144 WIRE 480 208 464 208 WIRE -16 288 -80 288 WIRE 80 288 64 288 WIRE 384 288 368 288 WIRE 480 288 480 208 WIRE 480 288 464 288 WIRE 480 304 480 288 WIRE 656 304 656 288 WIRE 656 304 592 304 WIRE 656 320 656 304 WIRE -64 368 -80 368 WIRE 80 368 0 368 WIRE 416 368 368 368 WIRE 480 400 480 384 WIRE 224 416 224 400 WIRE 656 416 656 400 WIRE 416 432 416 368 WIRE 592 432 592 304 WIRE 592 432 416 432 FLAG -256 96 0 FLAG -80 368 0 FLAG -80 288 0 FLAG -80 208 0 FLAG 480 400 0 FLAG 656 416 0 FLAG 224 416 0 FLAG -80 0 0 FLAG 144 48 0 FLAG 592 -32 0 FLAG 976 64 0 FLAG 1040 64 0 FLAG 1136 64 0 SYMBOL PowerProducts\\LTC3805-5 224 240 R0 SYMATTR InstName U1 SYMBOL voltage -256 -48 R0 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V1 SYMATTR Value 60 SYMBOL cap 0 352 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 WINDOW 39 60 32 VTop 2 SYMATTR InstName C1 SYMATTR Value 470p SYMATTR SpiceLine Rser=20K SYMBOL res 80 272 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R1 SYMATTR Value 112K SYMBOL cap 0 192 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C2

SYMBOL nmos 432 48 R0 SYMATTR InstName Q1 SYMATTR Value IRFP4668 SYMBOL res 464 288 R0 SYMATTR InstName R2 SYMATTR Value 0.068R SYMBOL res 480 192 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R3 SYMATTR Value 1.33K SYMBOL res 368 304 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 0 56 VBottom 2 SYMATTR InstName R4 SYMATTR Value 3K SYMBOL res 640 192 R0 SYMATTR InstName R5 SYMATTR Value 270k SYMBOL res 640 304 R0 SYMATTR InstName R6 SYMATTR Value 20k SYMBOL ind2 464 -144 R0 SYMATTR InstName L2

SYMATTR Type ind SYMBOL ind2 576 -32 M180 WINDOW 0 36 80 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName L3

SYMATTR Type ind SYMBOL res 32 -128 R0 SYMATTR InstName R7 SYMATTR Value 780k SYMBOL res 32 -48 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R8 SYMATTR Value 20k SYMBOL res 208 -144 R0 SYMATTR InstName R9 SYMATTR Value 4k7 SYMBOL cap 128 -32 R0 SYMATTR InstName C4

SYMBOL diode 304 -64 R0 SYMATTR InstName D2 SYMBOL res 544 -256 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R10 SYMATTR Value 10R SYMBOL schottky 720 -144 R270 WINDOW 0 32 32 VTop 2 WINDOW 3 0 32 VBottom 2 SYMATTR InstName D3 SYMATTR Value MBR20100CT SYMATTR Description Diode SYMATTR Type diode SYMBOL cap 960 -112 R0 SYMATTR InstName C5

SYMBOL cap 1024 -96 R0 SYMATTR InstName C6

SYMBOL ind2 496 -320 R180 WINDOW 0 36 80 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName L4

SYMATTR Type ind SYMBOL schottky 320 -464 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName D1 SYMATTR Value MBR20100CT SYMATTR Description Diode SYMATTR Type diode SYMBOL cap 144 -432 R0 SYMATTR InstName C3

SYMBOL res 32 -432 R0 SYMATTR InstName R11 SYMATTR Value 1000R SYMBOL res 1120 -128 R0 SYMATTR InstName R12 SYMATTR Value 500R TEXT 736 416 Left 2 !.tran 100m startup TEXT 480 -176 Left 2 !K1 L2 L3 L4 1 TEXT 56 -464 Left 2 ;AUX 12V TEXT 1056 -176 Left 2 ;12V OUT TEXT -256 -144 Left 2 ;60-280V IN

Reply to
Ev
Loading thread data ...

following this group for a while, learning a bit, and now I'm stuck!

Two low power 12V outputs, one common with the -ve input and one with the + ve.

OK I known next to nothing about DC-DC converters, but why not a buck? or why flyback?

George H.

shelf Wuerth transformer (using their DCM flyback selector tool) and amazin gly it didn't immediately blow up. However it won't work over about 200V i nput.

is IC) and saturating the transformer, hitting the current limit.

one:

formatting link

? And is there an obvious basic flaw with the design anyway - quite possib le, this is my first attempt at an SMPS. Any comments welcome [braces]

Reply to
George Herold

I took a look at your model; looks like you are running into minimum ON tim e at around Vin=140V... I ran the model up to Vin=600V just to see and it runs fine in LTspice land but if you look at the voltage at the drain o f the FET you can see it no longer is pulled all the way to ground during t he ON time. I guess in the sim you still get the right amount of V-us on t he primary as the drain temporarily dips to (Vin - 60V) or so.

In general nothing else in the model looks bad at first glance.

You definitely want the drain to go all the way to ground during the ON tim e, otherwise you are playing games with the FET turn on time, Cgd, and the IC minimum ON time. Hard to predict your true max Vin in this case. BTW s ince this is your first flyback, the drain voltage is a nice waveform to lo ok at in general since you can see exactly when the magnetizing current goe s to zero.

Two things I would try are picking a controller with pulse skip mode, and c hanging your turns ratio.

Reply to
sea moss

Hi George, good question. As I say I'm not an expert but these were my reasons for why not a buck converter:

1) I need two outputs, a buck only gives one, so then I'd need another DC-DC converter to get the other rail. 2) The switch arrangement on flyback seems simpler - one N-channel FET to ground via a sense resistor - all ground referenced. 3) the voltage ratio is 23:1 for a buck circuit, worst case. That seems excessive, needing a lot of inductance. 4) if the switch fails on, a lot of volts goes through some small low voltage components which might explode dramatically.

So, flyback seems "more right", but maybe it's jumping in at the deep end, for a beginner.

Reply to
Ev

Hey thank you very much for taking the time to look at it.

Great tip, I can see what you mean. It's encouraging that the model shows that happening right around the point where the real circuit stops working.

OK. Yes I figured more ratio would help. I was hoping to choose something off the shelf but maybe that's not going to happen. I'll look into both of those. thanks again.

Evan

Reply to
Ev

For protos, I usually power things from random laptop bricks and use an inverting buck for the negative supply.

Flybacks are fairly notorious for putting out hash on the input, output, ground wires, through the air, and anywhere else they can think of. ;)

What's the power source?

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 
https://hobbs-eo.com
Reply to
Phil Hobbs

Rob just designed a +24 to +-15 buck for me, that uses a DRQ-series dual inductor, one buck winding and the opposite winding rectified to get the other supply. That usually has terrible cross-regulation, but he did the inverting side as the main switcher, and rectified the + output off the side winding. Apparently the duty cycles work right that way, and cross regulation is pretty good.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

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

How did he avoid getting 39 volts on the cross-regulated side? Did he hang the secondary on the +24 input?

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

No, the +15 output winding has one side grounded, the other end diode rectified into a cap. He built it and it works. Apparently the second winding makes 39V p-p and one phase of that half-wave rectifies to 15. I could do the math but I'd rather fry shrimp and drink beer.

I don't have it here at home, but I can post it on Monday.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics
Reply to
John Larkin

Hey, it's obvious. In the inverting buck, one end of the inductor is ground. The other end switches from +24 to -15. So the extra winding must switch from -24 to +15.

If the switcher is non-synchronous, the catch diode voltage drop on the buck side cancels the rectifier diode drop on the +15 side. Pretty slick.

I think the beer helped.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics
Reply to
John Larkin

What switcher chip is it, and does it work with the +15 loaded more heavily than the -15?

Thanks

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

It's an ancient Simple Switcher, LM2576 I think. I was concerned about that loading situation but Rob says the unequal-loaded outputs work about like you'd want them to. That would take more thinking to understand. I'm powering a mess of THS6022s, all making sinewave outputs, so my power supply loads will be well balanced.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics
Reply to
John Larkin

Cool.

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

Reminds me of some little low power DC-DC converter modules I made way back for powering +-12V MC1488 RS-232 drivers from a 5V only system:

formatting link

piglet

Reply to
piglet

That's a nice topology. It's like Rob's idea: do the negative side as the primary output first. Flyback ICs usually assume positive side operation, which requires a 3-winding transformer to get equal + and - flyback outputs.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics
Reply to
John Larkin

OK, but I need to make a small run of these once I've got it working, like 10 off. And then hopefully progress it towards something manufacturable.

Good point!

PV panels, a variable quantity of, hence 40-280VDC

Thanks!

Reply to
Ev

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.