Flyback Efficiency -- Looking for suggestions

I done messed up, and have managed to make a 50% efficient switching converter.

I may have to just grin and use it (look! it's a space heater, too!), but if there's a quick solution I'm open to suggestions.

It's a flyback, going from 3.7V (1-cell LiPo) to 200V. I'm using off-the- shelf transformers.

It was looking a lot better in simulation than in real life, until I added what seems to be a realistic amount of capacitance to the secondary.

So -- anyone see anything I can do with this? Suggestions welcome. Thanks in advance.

Version 4 SHEET 1 4556 1460 WIRE 1808 -352 1696 -352 WIRE 2512 -352 1872 -352 WIRE 2560 -352 2512 -352 WIRE 2944 -352 2560 -352 WIRE 3168 -352 2944 -352 WIRE -144 -304 -240 -304 WIRE 48 -304 -64 -304 WIRE 176 -304 48 -304 WIRE 1520 -304 176 -304 WIRE 2144 -304 1520 -304 WIRE 2416 -304 2320 -304 WIRE 2560 -304 2560 -352 WIRE 2560 -304 2480 -304 WIRE 1520 -272 1520 -304 WIRE 1696 -272 1696 -352 WIRE 2144 -272 2144 -304 WIRE 2320 -272 2320 -304 WIRE 48 -256 48 -304 WIRE 2560 -256 2560 -304 WIRE 3168 -224 3168 -352 WIRE 1712 -192 1696 -192 WIRE 2336 -192 2320 -192 WIRE 48 -160 48 -192 WIRE 1392 -160 1328 -160 WIRE 1520 -160 1520 -192 WIRE 1520 -160 1456 -160 WIRE 1712 -160 1712 -192 WIRE 2016 -160 1952 -160 WIRE 2144 -160 2144 -192 WIRE 2144 -160 2080 -160 WIRE 2336 -160 2336 -192 WIRE 2560 -160 2560 -192 WIRE 1328 -64 1328 -160 WIRE 1728 -64 1328 -64 WIRE 1952 -64 1952 -160 WIRE 1952 -64 1728 -64 WIRE 2352 -64 1952 -64 WIRE 2512 -64 2352 -64 WIRE 192 -48 64 -48 WIRE 256 -48 192 -48 WIRE 416 -48 256 -48 WIRE 672 -48 416 -48 WIRE 896 -48 672 -48 WIRE 1136 -48 896 -48 WIRE 1872 -48 1136 -48 WIRE 2512 -48 2512 -64 WIRE 3168 -48 3168 -144 WIRE 416 -32 416 -48 WIRE 256 -16 256 -48 WIRE 1136 -16 1136 -48 WIRE 1520 -16 1520 -160 WIRE 1872 -16 1872 -48 WIRE 2144 -16 2144 -160 WIRE 672 0 672 -48 WIRE 2352 0 2352 -64 WIRE -240 32 -240 -304 WIRE 64 32 64 -48 WIRE 1040 32 1008 32 WIRE 1296 32 1232 32 WIRE 1776 32 1744 32 WIRE 2032 32 1968 32 WIRE 3120 32 3072 32 WIRE 2512 48 2512 32 WIRE 256 64 256 48 WIRE 416 64 416 48 WIRE 512 64 416 64 WIRE 896 64 896 -48 WIRE 896 64 832 64 WIRE 1008 64 1008 32 WIRE 1040 64 1008 64 WIRE 1296 64 1296 32 WIRE 1296 64 1232 64 WIRE 1472 64 1296 64 WIRE 1744 64 1744 32 WIRE 1776 64 1744 64 WIRE 2032 64 2032 32 WIRE 2032 64 1968 64 WIRE 2096 64 2032 64 WIRE 3072 80 3072 32 WIRE 3168 96 3168 48 WIRE 2352 112 2352 64 WIRE 1136 128 1136 112 WIRE -240 144 -240 112 WIRE 64 144 64 112 WIRE 2512 144 2512 112 WIRE 336 160 288 160 WIRE 512 160 400 160 WIRE 928 160 832 160 WIRE 1008 160 1008 64 WIRE 1008 160 928 160 WIRE 1744 160 1744 64 WIRE 1744 160 1008 160 WIRE 288 192 288 160 WIRE 1872 192 1872 112 WIRE 3072 192 3072 160 WIRE 2496 208 2400 208 WIRE 1520 240 1520 80 WIRE 2144 240 2144 80 WIRE 2144 240 1520 240 WIRE 2400 240 2400 208 WIRE 512 256 64 256 WIRE 960 256 832 256 WIRE 1088 256 960 256 WIRE 1184 256 1088 256 WIRE 1408 256 1264 256 WIRE 1520 256 1520 240 WIRE 1520 256 1408 256 WIRE 3488 272 3440 272 WIRE 3488 288 3488 272 WIRE 64 304 64 256 WIRE 2944 304 2944 -352 WIRE 240 352 224 352 WIRE 352 352 320 352 WIRE 480 352 352 352 WIRE 512 352 480 352 WIRE 960 352 832 352 WIRE 1792 352 960 352 WIRE 1904 352 1792 352 WIRE 1952 352 1904 352 WIRE 2096 352 2032 352 WIRE 2400 352 2400 320 WIRE 1408 384 1408 256 WIRE 352 400 352 352 WIRE 3488 400 3488 368 WIRE 896 416 896 64 WIRE 1088 416 1088 256 WIRE 1520 416 1520 256 WIRE 1792 416 1792 352 WIRE 1904 432 1904 352 WIRE 2048 432 1904 432 WIRE 2192 432 2128 432 WIRE 2240 432 2192 432 WIRE 2288 432 2240 432 WIRE 2400 432 2368 432 WIRE 2544 432 2464 432 WIRE 64 464 64 384 WIRE 416 464 416 64 WIRE 672 464 672 416 WIRE 2496 464 2496 208 WIRE 2544 480 2544 432 WIRE 2544 480 2528 480 WIRE 3264 480 2544 480 WIRE 3376 480 3344 480 WIRE 3392 480 3376 480 WIRE 352 496 352 464 WIRE 1408 496 1408 464 WIRE 1408 496 1296 496 WIRE 2240 496 2240 432 WIRE 2464 496 2240 496 WIRE 1408 512 1408 496 WIRE 2800 512 2528 512 WIRE 2944 512 2944 384 WIRE 2944 512 2800 512 WIRE 3392 512 3392 480 WIRE 896 528 896 480 WIRE 1088 528 1088 480 WIRE 1520 528 1520 496 WIRE 2944 528 2944 512 WIRE 2800 544 2800 512 WIRE 1792 560 1792 496 WIRE 2496 560 2496 528 WIRE 416 592 416 544 WIRE 1408 608 1408 576 WIRE 2800 640 2800 608 WIRE 2944 640 2944 608 WIRE 3392 640 3392 592 FLAG -240 144 0 FLAG 672 464 0 FLAG 1520 528 0 FLAG 896 528 0 FLAG 64 464 0 FLAG 928 160 Vg1 FLAG 2560 -160 0 FLAG 3168 96 0 FLAG 3072 192 0 FLAG 48 -160 0 FLAG 256 64 0 FLAG 2352 112 0 FLAG 2512 -352 Vout FLAG 1728 -64 Vsnub FLAG 176 -304 Vin FLAG 480 352 Vc FLAG 960 352 Vfbx FLAG 1136 128 0 FLAG 1872 192 0 FLAG 192 -48 Vgs FLAG 416 592 0 FLAG 2512 144 0 FLAG 3392 640 0 FLAG 2496 560 0 FLAG 3440 272 Vvg FLAG 3488 400 0 FLAG 2944 640 0 FLAG 2192 432 Vvcmd FLAG 2096 352 Vvg FLAG 224 352 Vvg FLAG 2400 352 0 FLAG 1792 560 0 FLAG 352 496 0 FLAG 64 144 0 FLAG 3376 480 Vcmd FLAG 1296 496 VHV_Curr FLAG 2336 -160 0 FLAG 1712 -160 0 FLAG 1408 608 0 FLAG 1088 528 0 FLAG 960 256 Vsense FLAG 288 192 0 FLAG 2800 640 0 SYMBOL voltage -240 16 R0 WINDOW 123 0 0 Left 2 WINDOW 39 24 124 Left 2 SYMATTR SpiceLine Rser=10m SYMATTR InstName V1 SYMATTR Value PWL(0 0 1m 3.2) SYMBOL PowerProducts\\LT3757A 672 208 R0 SYMATTR InstName U1 SYMBOL cap 880 416 R0 SYMATTR InstName C1 SYMATTR Value 4.7? SYMBOL res 1504 400 R0 SYMATTR InstName R1 SYMATTR Value 3m SYMBOL nmos 1472 -16 R0 SYMATTR InstName M1 SYMATTR Value BSZ067N06LS3 SYMBOL ind2 1504 -176 M180 WINDOW 0 36 80 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName L1 SYMATTR Value 2.4? SYMATTR SpiceLine Rser=12.5m Rpar=20 SYMATTR Type ind SYMBOL res 2144 416 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R2 SYMATTR Value 100k SYMBOL res 48 288 R0 SYMATTR InstName R4 SYMATTR Value 270k SYMBOL cap 2544 -256 R0 SYMATTR InstName C4 SYMATTR Value 1? SYMBOL res 3152 -240 R0 SYMATTR InstName R11 SYMATTR Value 2.80k SYMBOL nmos 3120 -48 R0 SYMATTR InstName M3 SYMATTR Value FDS2734 SYMBOL voltage 3072 64 R0 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V2 SYMATTR Value PULSE(0 12 4m 1u 1u 500u 1m 5) SYMBOL cap 32 -256 R0 SYMATTR InstName C7 SYMATTR Value 2000? SYMATTR SpiceLine Rser=14m SYMBOL cap 240 -16 R0 SYMATTR InstName C9 SYMATTR Value 10? SYMBOL ind2 1712 -288 M0 SYMATTR InstName L2 SYMATTR Value 1310? SYMATTR Type ind SYMATTR SpiceLine Rser=3 Cpar=2n SYMBOL diode 1808 -336 R270 WINDOW 0 32 32 VTop 2 WINDOW 3 0 32 VBottom 2 SYMATTR InstName D1 SYMATTR Value RFN2L6S SYMBOL cap 2336 0 R0 SYMATTR InstName C3 SYMATTR Value 10? SYMBOL nmos 2096 -16 R0 SYMATTR InstName M2 SYMATTR Value BSZ067N06LS3 SYMBOL ind2 2128 -176 M180 WINDOW 0 36 80 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName L3 SYMATTR Value 2.4? SYMATTR SpiceLine Rser=12.5m Rpar=20 SYMATTR Type ind SYMBOL ind2 2336 -288 M0 SYMATTR InstName L4 SYMATTR Value 1310? SYMATTR Type ind SYMATTR SpiceLine Rser=3 Cpar=2n SYMBOL diode 2416 -288 R270 WINDOW 0 32 32 VTop 2 WINDOW 3 0 32 VBottom 2 SYMATTR InstName D4 SYMATTR Value RFN2L6S SYMBOL schottky 1456 -176 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName D5 SYMATTR Value UPSC600 SYMATTR Description Diode SYMATTR Type diode SYMBOL schottky 2080 -176 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName D6 SYMATTR Value UPSC600 SYMATTR Description Diode SYMATTR Type diode SYMBOL ZXGD3001E6 1136 48 R0 SYMATTR InstName U2 SYMBOL ZXGD3001E6 1872 48 R0 SYMATTR InstName U3 SYMBOL res 400 -48 R0 SYMATTR InstName R8 SYMATTR Value 301k SYMBOL res 400 448 R0 SYMATTR InstName R10 SYMATTR Value 100k SYMBOL zener 2528 112 R180 WINDOW 0 24 64 Left 2 WINDOW 3 24 0 Left 2 SYMATTR InstName D7 SYMATTR Value EDZV30B SYMATTR Description Diode SYMATTR Type diode SYMBOL res 2496 -64 R0 SYMATTR InstName R13 SYMATTR Value 100 SYMBOL voltage 3392 496 R0 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V3 SYMATTR Value PWL(0 0 1.5m 0 3m 4) SYMBOL Opamps\\UniversalOpamp2 2496 496 M0 SYMATTR InstName U5 SYMBOL voltage 3488 272 R0 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V4 SYMATTR Value 1.6 SYMBOL res 2928 512 R0 SYMATTR InstName R14 SYMATTR Value 25.5k SYMBOL res 2928 288 R0 SYMATTR InstName R15 SYMATTR Value 1.24meg SYMBOL res 3360 464 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R17 SYMATTR Value 25.5k SYMBOL res 2048 336 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R3 SYMATTR Value 4.99k SYMBOL res 224 368 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 0 56 VBottom 2 SYMATTR InstName R6 SYMATTR Value 100k SYMBOL voltage 2400 224 R0 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V5 SYMATTR Value 5 SYMBOL res 1776 400 R0 SYMATTR InstName R12 SYMATTR Value 174meg SYMBOL cap 336 400 R0 SYMATTR InstName C10 SYMATTR Value 33p SYMBOL voltage 64 16 R0 WINDOW 123 0 0 Left 2 WINDOW 39 24 124 Left 2 SYMATTR SpiceLine Rser=10 SYMATTR InstName V6 SYMATTR Value PWL(0 0 100u 0 101u 12) SYMBOL res 2384 416 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R5 SYMATTR Value 100k SYMBOL cap 2464 416 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C2 SYMATTR Value 10n SYMBOL cap 1392 512 R0 SYMATTR InstName C5 SYMATTR Value 100n SYMBOL res 1392 368 R0 SYMATTR InstName R7 SYMATTR Value 1k SYMBOL res 1280 240 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R9 SYMATTR Value 100 SYMBOL cap 1072 416 R0 SYMATTR InstName C6 SYMATTR Value 1n SYMBOL cap 400 144 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C8 SYMATTR Value 33n SYMBOL cap 2784 544 R0 SYMATTR InstName C11 SYMATTR Value 330p SYMBOL ind -160 -288 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 5 56 VBottom 2 SYMATTR InstName L5 SYMATTR Value 220n TEXT -208 560 Left 2 !.tran 10m startup TEXT 1312 -240 Left 2 !K1 L1 L2 0.985 TEXT 1296 -272 Left 2 ;Wurth 760871113 TEXT 1936 -240 Left 2 !K2 L3 L4 0.985 TEXT 1912 -272 Left 2 ;Wurth 760871113 TEXT -208 592 Left 2 !.lib ZXGD3001E6.spicemodel.lib TEXT -200 624 Left 2 !.ic V(Vsnub)=30

--
Tim Wescott 
Wescott Design Services 
 Click to see the full signature
Reply to
Tim Wescott
Loading thread data ...

I managed 70% on my 24-to-1400 volt supply. The transformer gets hot.

It complains about ZXGD3001 being unknown.

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin

I'd be happy at 70 or 75%, I think.

Crap. Thank you. This is why I didn't post before. Here are images of TWO FILES, the model first, then the footprint:

***** Model: ZXGD3001E6.spicemodel.lib ***** * *Zetex ZXGD3001E6 Spice Model v2.0 Last Revised 17/09/08
  • .SUBCKT ZXGD3001E6 1 2 3 4 5 6
*pins Vcc, In1, Gnd, Sink, In2, Source Q1 1 2 6 ZXGD3001N Q2 3 5 4 ZXGD3001P
  • .MODEL ZXGD3001N NPN IS=9E-13 BF=990 NF=1 VAF=25 IKF=3.8 ISE=8E-14 NE=1.35
  • BR=410 NR=1 VAR=8 IKR=1.25 ISC=8e-14 NC=1.35 RE=0.0117 RB=0.1 RC=0.0081
  • CJE=168E-12 VJE=0.7 MJE=0.38 CJC=61E-12 VJC=0.52 MJC=0.31 TF=0.5E-9 TR=1.7e-9
  • XTB=1.4
  • .MODEL ZXGD3001P PNP IS=5.5E-13 NF=1 BF=650 VAF=20 ISE=1.9E-13
  • IKF=2.5 NE=1.53 BR=72 VAR=4.1 ISC=7E-14 NC=1.2 IKR=0.25 RC=0.010
  • RB=0.15 RE=0.006 QUASIMOD=1 RCO=0.7 GAMMA=1.7E-9 CJC=57E-12 MJC=0.35
  • VJC=0.53 CJE=168E-12 MJE=0.54 VJE=0.95 TF=0.42E-9 TR=8.4E-9 TRC1=0.005
  • TRB1=0.005 TRE1=0.005 XTB=1.4
  • .ENDS ZXGD3001E6
* *$ *
  • (c) 2008 Diodes Incorporated
*
  • The copyright in these models and the designs embodied belong
  • to Diodes Incorporated (" Diodes "). They are supplied
  • free of charge by Diodes for the purpose of research and design
  • and may be used or copied intact (including this notice) for
  • that purpose only. All other rights are reserved. The models
  • are believed accurate but no condition or warranty as to their
  • merchantability or fitness for purpose is given and no liability
  • in respect of any use is accepted by Diodes Incorporated, its distributors
  • or agents.
*
  • Diodes Incorporated, 1566 N. Dallas Parkway, Suite 850, Dallas, TX
75248, USA

***** Footprint: ZXGD3001E6.asy *****

Version 4 SymbolType CELL LINE Normal -96 64 -96 -64 LINE Normal 96 64 -96 64 LINE Normal 96 -63 96 64 LINE Normal 96 -64 96 -63 LINE Normal -96 -64 96 -64 WINDOW 0 32 -80 Left 2 WINDOW 1 32 80 Left 2 SYMATTR Prefix X SYMATTR SpiceModel ZXGD3001E6 PIN 0 -64 TOP 8 PINATTR PinName VCC PINATTR SpiceOrder 1 PIN -96 -16 LEFT 8 PINATTR PinName IN1 PINATTR SpiceOrder 2 PIN -96 16 LEFT 8 PINATTR PinName IN2 PINATTR SpiceOrder 5 PIN 0 64 BOTTOM 8 PINATTR PinName GND PINATTR SpiceOrder 3 PIN 96 16 RIGHT 8 PINATTR PinName SINK PINATTR SpiceOrder 4 PIN 96 -16 RIGHT 8 PINATTR PinName SOURCE PINATTR SpiceOrder 6

--
Tim Wescott 
Wescott Design Services 
 Click to see the full signature
Reply to
Tim Wescott

That's a pretty weird part. It seems to be just two transistors inside, poorly specified. Do you really need it?

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin

Dunno, but Joerg suggested it. They drive lots of current.

--
Tim Wescott 
Wescott Design Services 
 Click to see the full signature
Reply to
Tim Wescott

I can't get that to run.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

-
.

The transformer doesn't seem to have been designed for the job

formatting link

The specification doesn't cover the self-resonant frequency. A 54:1 step-up is fairly fierce. Baxandall's class-D oscillator seems to have been design ed to cope with similar sorts of step-up ratios in 12V to 1000V photomultip lier supplies, and it's generally difficult to run them at much over 50kHz, unless you organise the secondary winding into banks, but Baxandall typica lly got better than 90% efficiency, and Jim Williams did as well with the s ame circuit.

The 7608711113 does seem to lend itself to being hooked-up as centre-tapped primary (windings 2 and 3 - with pins 3 and 4 tied together) but even if y ou put the other two winding in series you'd have to put a Cockroft-Walton voltage multiplier on the output to get your 200V.

156:6 is is only 26, so doubling it doesn't get you to 54:1 but the Baxanda ll gives a pi/2 peak output - an extra 57% - which would get you home.

formatting link

With only 3.2V primary drive, the ultra-simple class-D oscillator I show on my web-site would work

formatting link
r1.htm

It doesn't have an obvious mechanism for pulse-width modulating the drive t o allow feedback control of the output voltage - that is do-able with a sep arate centre-tapped gate-drive winding, where you pull down the centre-tap to turn on the "on" MOSFet sometime after cross-over, and turn it off again sometime before cross-over, but the implementation is left as an exercise for the reader.

Obviously, the MOSFET gate being driven can't be protected by diodes to the substrate.

Using a second 7608711113 as the gate-driver would be a fun option, but get ting a proper transformer wound would probably lead to a better solution.

--
Bill Sloman, Sydney
Reply to
bill.sloman

A lot of, I don't know, rookie mistakes..? My read:

- LT3757 is a peak current mode controller, much like the classic UC3842 and such. Easy to use, lots of protection/monitoring, internal gate drive...

...So why the external gate drive? And such oversized transistors? Shouldn't be needing either.

The outboard error amp seems like it's just inviting trouble. Does this need to be variable?

I can go through this in more detail, would you like to Skype or e-mail or anything?

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
 Click to see the full signature
Reply to
Tim Williams

It's just a complementary pair with some bizarre attached specs (like 1ns rise/fall times but no logic level input). They kind of imply it's not BJTs, yet cannot escape the obvious truth of what it is...

There was also a ZXTC series, which is specified sanely. I don't know which they prefer to sell, or which is more common.

If you like, you can toss a pair of lone parts around, like ZTX651/751, or not even from Diodes/Zetex, e.g. PBSS303NX/PX.

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
 Click to see the full signature
Reply to
Tim Williams

I played with it a little, by driving the fets directly from the LTC part. The gate rise/fall looked fine.

53% simulated. The snubber dumps over 2 watts. R1 gets hot too.
--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

The snubber was dissipating wisps of power until (blush) I realized that I hadn't modeled the coil's parallel capacitance.

--
www.wescottdesign.com
Reply to
Tim Wescott

The snubber is supposed to gobble the energy stored in the leakage inductance. Is K=0.985 realistic?

Capacitance in the fet(s) and the transformer are the enemy of efficiency in a boost or flyback converter. One test is to run it open-loop and see how much voltage you can untimately get. A voltage doubler or C-W multiplier after the transformer can help. I'm getting

70%, 24 volts to 1400, 7 watts out, with just a cheap center-tapped inductor as the transformer.

formatting link

formatting link

The transformer is the hottest part, around 60C; haven't figured out why. The fet is now a Fairchild FDD7N20, a good compromise between capacitance and Rds-on at this power level.

I got some bigger Wurth dual inductors, samples made in China and Fed-exed from Germany, but they are slightly worse.

What parts are getting hot on yours?

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

They sure do. One of my spring projects is where a client asked if my

2015 design can deliver about 10x the peak power. If I hadn't put in those ZXGD3001 it would not be able to do it.

Can't study your switcher right now because of a very full plate of work but my suggestion would be to see what gets hot. Heat is about the only method of escape for dissipated energy. Except for smoke :-)

--
Regards, Joerg 

http://www.analogconsultants.com/
Reply to
Joerg

Where did you get parallel output capacitance of 2000pF from?

If SRF is 100KHz, do you think you can hard-switch PWM at 50KHz without chaotic interactions?

RL

Reply to
legg

And EMI!

I don't think that Tim needs the gate drivers. The LTC chip looks fine alone in the sim.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

I cannot get it to run. No PWM at all. Have not had time to dig into it

Cheers

Klaus

Reply to
klaus.kragelund

Dropbox files, lib etc:

formatting link

formatting link

formatting link

Reply to
klaus.kragelund

A wild-ass guess after looking at oscilloscope traces.

Well, I _am_ thinking I done messed up.

--
Tim Wescott 
Wescott Design Services 
 Click to see the full signature
Reply to
Tim Wescott

I simplified the power supplies and the load, changed the many ? things to u, and got rid of the gate drivers.

It runs but isn't very stable. Lots of runt switching pulses.

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin

If it's the LT3757 it is quite staunch. I never use the internal LDO because that's not enough swing for real hard switching. Sometimes it is better to use the A version. Has a deficiency fixed that could hurt with fast load changes. That change was triggered by yours truly. LTC is a great company, they really listen.

With really big FETs there comes a limit where you need those ZXGD drivers or similar. Especially if the FETs can't be heatsinked well and a few hundred milliwatts in transition loss avoidance counts. That will now save a client of mine from having to do a board spin. I bridge those drivers with a resistor anyhow to retain full swing. If not needed one can then simply leave the driver off. But I mostly don't, they are cheap.

--
Regards, Joerg 

http://www.analogconsultants.com/
Reply to
Joerg

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.