This is pleasingly weird.
What's strange is that adding the two snubbers increases the LT spice sim speed radically, about 10:1.
There are bunch of topologies that charge inductors in parallel and discharge in series, helps take the stress off the switch(es) and don't need huge duty cycles. I have one like that in my filez that's like a Cuk, with a quasi-floating output posted at the end though haven't found a particular use for it yet..
High dv/dt stiffens the system, maybe.
Version 4 SHEET 1 2640 1316 WIRE 16 -768 -912 -768 WIRE 448 -768 16 -768 WIRE 672 -768 448 -768 WIRE 16 -624 16 -768 WIRE 448 -624 448 -768 WIRE 672 -400 672 -768 WIRE 224 -384 -624 -384 WIRE 448 -384 448 -544 WIRE 448 -384 304 -384 WIRE 512 -384 448 -384 WIRE 624 -384 576 -384 WIRE 16 -208 16 -544 WIRE 976 -208 16 -208 WIRE 1200 -208 976 -208 WIRE 1488 -208 1264 -208 WIRE 1728 -208 1488 -208 WIRE 2032 -208 1808 -208 WIRE 2432 -208 2032 -208 WIRE 2528 -208 2432 -208 WIRE -624 -80 -624 -384 WIRE 976 -64 976 -208 WIRE 1488 -64 1488 -208 WIRE 2432 64 2432 -208 WIRE -912 80 -912 -768 WIRE 976 80 976 0 WIRE 1488 80 1488 0 WIRE 2032 80 2032 -208 WIRE 16 288 16 -208 WIRE -624 368 -624 -16 WIRE -528 368 -624 368 WIRE -384 368 -448 368 WIRE -144 368 -384 368 WIRE -32 368 -80 368 WIRE 672 368 672 -304 WIRE 976 368 1488 80 WIRE 976 368 672 368 WIRE 1216 368 976 368 WIRE 1504 368 976 80 WIRE 1504 368 1280 368 WIRE 2032 368 2032 144 WIRE 2032 368 1504 368 WIRE 2432 368 2432 144 WIRE 2432 368 2032 368 WIRE 2512 368 2432 368 WIRE 2432 480 2432 368 WIRE -384 512 -384 368 WIRE 672 608 672 368 WIRE 2432 624 2432 560 WIRE -912 800 -912 160 WIRE -384 800 -384 592 WIRE -384 800 -912 800 WIRE 16 800 16 384 WIRE 16 800 -384 800 WIRE 672 800 672 688 WIRE 672 800 16 800 WIRE 16 880 16 800 WIRE -624 912 -624 368 WIRE -624 1120 -624 992 FLAG 16 880 0 FLAG -624 1120 0 FLAG 2432 624 0 FLAG 2528 -208 Out- IOPIN 2528 -208 Out FLAG 2512 368 Out+ IOPIN 2512 368 Out SYMBOL voltage -912 64 R0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName V1 SYMATTR Value 5 SYMBOL nmos -32 288 R0 SYMATTR InstName M1 SYMATTR Value Si3440DV SYMBOL cap 1264 -224 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 40 32 VTop 2 SYMATTR InstName C1 SYMATTR Value 0.47µ SYMBOL cap 1280 352 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C2 SYMATTR Value 0.47µ SYMBOL ind2 1712 -192 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 4 56 VBottom 2 SYMATTR InstName L3 SYMATTR Value 220µ SYMBOL schottky 960 -64 R0 SYMATTR InstName D2 SYMATTR Value RB168LAM150 SYMATTR Description Diode SYMATTR Type diode SYMBOL voltage -624 896 R0 WINDOW 3 52 50 Left 2 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR Value PULSE(0 5 0 10n 10n 0.0000122 0.0000142) SYMATTR InstName V2 SYMBOL cap 2016 80 R0 SYMATTR InstName C4 SYMATTR Value 4.7µ SYMBOL ind2 32 -528 R180 WINDOW 0 36 80 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName L1 SYMATTR Value 220µ SYMBOL ind2 688 704 R180 WINDOW 0 36 80 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName L2 SYMATTR Value 220µ SYMBOL res 2416 464 R0 SYMATTR InstName R1 SYMATTR Value 1Meg SYMBOL Digital\\inv -688 -16 R270 SYMATTR InstName A1 SYMATTR SpiceLine Vhigh=5V,Td=5e-9,Trise=100n SYMBOL pmos 624 -304 M180 SYMATTR InstName M2 SYMATTR Value QS8M51_P SYMBOL ind2 208 -368 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 4 56 VBottom 2 SYMATTR InstName L6 SYMATTR Value 2.2µ SYMBOL ind2 -544 384 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 4 56 VBottom 2 SYMATTR InstName L7 SYMATTR Value 2.2µ SYMBOL res -400 496 R0 SYMATTR InstName R3 SYMATTR Value 10k SYMBOL res 432 -640 R0 SYMATTR InstName R4 SYMATTR Value 10k SYMBOL FerriteBead 544 -384 R90 WINDOW 0 -16 0 VBottom 2 SYMATTR InstName L4 SYMATTR Value 1.542µ SYMATTR SpiceLine Ipk=2 Rser=0.045 Rpar=587 Cpar=1.216p mfg="Würth Elektronik" pn="74279218 WE-CBF 1206" SYMBOL res 2416 48 R0 SYMATTR InstName Rload SYMATTR Value 10k SYMBOL FerriteBead -112 368 R90 WINDOW 0 -16 0 VBottom 2 SYMATTR InstName L5 SYMATTR Value 1.542µ SYMATTR SpiceLine Ipk=2 Rser=0.045 Rpar=587 Cpar=1.216p mfg="Würth Elektronik" pn="74279218 WE-CBF 1206" SYMBOL schottky 1472 -64 R0 SYMATTR InstName D1 SYMATTR Value RB168LAM150 SYMATTR Description Diode SYMATTR Type diode TEXT -872 560 Left 2 !.tran 0.1 TEXT 200 -568 Left 2 !K1 L1 L6 0.99 TEXT 256 464 Left 2 !K2 L7 L2 0.99
I started simulating with a single 1:5 transformer, but couldn't find one for sale.
24 to 500 is sort of a black hole for flyback transformers. The "capacitor charging" flybacks are either the wrong ratio or too wimpy.The DRQ127 parts are cool and cheap (under a dollar) and multi-sourced and pick-and-place compatible, so it makes sense to use four of them.
Inductors charged in parallel and discharged in series does sound cool, but I'd expect that to need a lot of parts.
On 9/24/2023 1:25 PM, John Larkin wrote:
If you have transformers with a dual secondary you can make a boosting autotransformer-type topology by using the spare secondaries to couple fluxes, so they act like they're all wound on the same core, sort of like this somewhat silly example:
Version 4 SHEET 1 2652 1076 WIRE 880 -144 240 -144 WIRE 1040 -144 880 -144 WIRE 1280 -144 1104 -144 WIRE 1472 -144 1280 -144 WIRE 1664 -144 1536 -144 WIRE 1840 -144 1728 -144 WIRE 2272 -144 1840 -144 WIRE 2496 -144 2272 -144 WIRE 0 -80 -80 -80 WIRE 144 -80 80 -80 WIRE 624 -64 240 -64 WIRE 1280 -64 1280 -144 WIRE 288 -16 240 -16 WIRE 544 -16 368 -16 WIRE -80 16 -80 0 WIRE 144 16 144 0 WIRE 144 16 -80 16 WIRE -80 32 -80 16 WIRE 480 64 240 64 WIRE 1280 112 1280 0 WIRE 880 128 880 -144 WIRE 1840 128 1840 -144 WIRE -80 144 -80 112 WIRE 304 240 240 240 WIRE 480 240 480 64 WIRE 480 240 384 240 WIRE 544 320 544 -16 WIRE 544 320 240 320 WIRE 2272 320 2272 -144 WIRE 192 352 16 352 WIRE 1280 352 1280 176 WIRE 1280 352 768 352 WIRE 1840 352 1840 192 WIRE 1840 352 1280 352 WIRE 192 432 64 432 WIRE 624 448 624 -64 WIRE 624 448 240 448 WIRE 1840 496 1840 352 WIRE 1280 512 1280 352 WIRE 352 528 240 528 WIRE 1280 656 1280 576 WIRE 16 672 16 352 WIRE 352 672 352 528 WIRE 352 672 16 672 WIRE 480 704 480 240 WIRE 480 704 240 704 WIRE 544 784 544 320 WIRE 544 784 240 784 WIRE 880 784 880 192 WIRE 1280 784 1280 720 WIRE 1280 784 880 784 WIRE 1488 784 1280 784 WIRE 1680 784 1552 784 WIRE 1840 784 1840 560 WIRE 1840 784 1744 784 WIRE 2272 784 2272 400 WIRE 2272 784 1840 784 WIRE 64 800 64 432 WIRE 192 800 64 800 WIRE 192 880 16 880 WIRE 768 880 768 352 WIRE 768 880 240 880 WIRE 240 896 240 880 WIRE 1840 912 1840 784 WIRE 464 976 240 976 WIRE 16 1040 16 880 WIRE 464 1040 464 976 WIRE 464 1040 16 1040 FLAG 1840 912 0 FLAG -80 144 0 FLAG 2496 -144 ~3kVDC IOPIN 2496 -144 Out SYMBOL ind2 128 16 M180 WINDOW 0 36 80 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName L1 SYMATTR Value 10 SYMBOL ind2 256 -48 R180 WINDOW 0 36 80 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName L2 SYMATTR Value 10 SYMBOL ind2 256 80 R180 WINDOW 0 36 80 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName L3 SYMATTR Value 10 SYMBOL ind2 256 224 M0 SYMATTR InstName L4 SYMATTR Value 10 SYMBOL ind2 176 448 M180 WINDOW 0 36 80 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName L5 SYMATTR Value 10 SYMBOL ind2 256 544 R180 WINDOW 0 36 80 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName L6 SYMATTR Value 10 SYMBOL voltage -80 -96 R0 WINDOW 3 -219 77 Left 2 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR Value SINE(0 170 60) SYMATTR InstName V1 SYMBOL cap 1824 496 R0 SYMATTR InstName C1 SYMATTR Value 4.7µ SYMBOL cap 1824 128 R0 SYMATTR InstName C2 SYMATTR Value 4.7µ SYMBOL diode 1472 -128 R270 WINDOW 0 32 32 VTop 2 WINDOW 3 0 32 VBottom 2 SYMATTR InstName D1 SYMATTR Value VS-E5TX0812 SYMBOL res -96 16 R0 SYMATTR InstName R1 SYMATTR Value 1Meg SYMBOL res 384 -32 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R3 SYMATTR Value 1 SYMBOL res 96 -96 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R2 SYMATTR Value 0.0001 SYMBOL diode 1296 176 R180 WINDOW 0 24 64 Left 2 WINDOW 3 24 0 Left 2 SYMATTR InstName D3 SYMATTR Value VS-E5TX0812 SYMBOL cap 1104 -160 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C3 SYMATTR Value 4.7µ SYMBOL diode 1296 576 R180 WINDOW 0 24 64 Left 2 WINDOW 3 24 0 Left 2 SYMATTR InstName D4 SYMATTR Value VS-E5TX0812 SYMBOL diode 1552 768 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName D5 SYMATTR Value VS-E5TX0812 SYMBOL cap 896 192 R180 WINDOW 0 24 56 Left 2 WINDOW 3 24 8 Left 2 SYMATTR InstName C4 SYMATTR Value 4.7µ SYMBOL res 2256 304 R0 SYMATTR InstName R4 SYMATTR Value 470k SYMBOL ind2 256 688 M0 SYMATTR InstName L7 SYMATTR Value 10 SYMBOL ind2 176 896 M180 WINDOW 0 36 80 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName L8 SYMATTR Value 10 SYMBOL ind2 256 880 M0 SYMATTR InstName L9 SYMATTR Value 10 SYMBOL res 400 224 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R6 SYMATTR Value 1 SYMBOL diode 1664 -128 R270 WINDOW 0 32 32 VTop 2 WINDOW 3 0 32 VBottom 2 SYMATTR InstName D2 SYMATTR Value VS-E5TX0812 SYMBOL diode 1296 0 R180 WINDOW 0 24 64 Left 2 WINDOW 3 24 0 Left 2 SYMATTR InstName D6 SYMATTR Value VS-E5TX0812 SYMBOL diode 1296 720 R180 WINDOW 0 24 64 Left 2 WINDOW 3 24 0 Left 2 SYMATTR InstName D7 SYMATTR Value VS-E5TX0812 SYMBOL diode 1744 768 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName D8 SYMATTR Value VS-E5TX0812 TEXT -32 -144 Left 2 !K1 L1 L2 L3 0.99 TEXT -40 272 Left 2 !K2 L4 L5 L6 0.99 TEXT -296 -48 Left 2 !.tran 10 TEXT -168 752 Left 2 !K3 L7 L8 L9 0.99
I could also rectify each of my secondaries independently into DC, and stack those in series. But no big benefit, more parts.
The DRQ127's are 2 widings, 1:1. Handy parts.
What was your 3KV for? I'm powering a Pockels Cell driver. It's only a moderate number of KHz so I shouldn't need a lot of power.
It would be nice to spread-spectrum my supply. That LT chip is fixed
200 KHz. The customer can get whiney about EMI.
wouldn't plenty of off-line smps transformers be in that range run in reverse? like f.eks.
The latest version seems to allow open caps and floating parts.
Resistors in the snubbers are burning watts. I'm seeing 90% without the snubbers. They need tweaking. Their current virtue is to speed up the sim time. Saveral other part values are there to do that, and will be changed on the real thing.
At under 5 watts out, and megawatts available, I don't really need efficiency.
LM5156 is a similar controller, but does spread-spectrum.
It's for passing EMI tests.
When the pencil sketch shows the inductor a 2uH, but doesn't show it's parallel capacitance, nor a series ferrite bead (which may not add all that much inductance, but only has about 1pF of parallel capacitance) the EMI conformity isn't likely to be great.
Have you measured the inter-winding capacitance of those DRQ127s?
My guess is many tens of pF and if you simulate with that then your snubber requirements may be quite changed?
piglet
LTSpice certainly lets you simulate with parallel and inter-winding capacitances.
Winding capacitance is easy enough to measure - you just need to resonate the winding with it's parallel capacitance. For 1:1 transformers like this you can put the two winding in parallel and measure the resonant frequency of the combination. The inter-winding capacitance isn't excited so it's a clean measurement. A capacitance meter can give your the interwinding capacitance, if you keep the test frequency well below the self-resonant frequency.
I suppose if you excited two windings anti-parallel you would emphasis the interwinding capacitance, but I'd have to Spice it to get some feel for what you'd see.
It's curious that the data sheet doesn't give parallel and interwinding capacitances. Transformers are something of a cottage industry, and not all the people who make them know as much as they might.
Or use a capacitance meter.
I think those parts are bifalar wound so C between windings will be high and will vary between the many parts in the family. I don't have the 200 uH around but I'll measure a few others.
I plan to breadboard the supply of course, if the customer is serious about wanting us to mke the driver. It's fast and fun to work with their engineers and scientists but the business people are glacial.
Inter-winding capacitance is the capacitance between two windings and you measure it by setting up an alternating voltage difference between the two winding and measuring the capacitative current current flowing between them.
John Larkin has made the point that if the windings are bifilar wound, as they often are in wound 1:1 transformers, you can get the interwinding capacitance from the wire and insulator properies. It ends to be high.
If the windings are printed - which is what you'd expect in a mass-produced part these days - life gets a bit more complicated.
Not so commonly that I've ever seen it used.
The resonant frequency tells you about the parallel capacitance and series inductance of the winding your have excited, and any other winding that is closely coupled to that winding. The Q tells you are about the resistance of the winding and the resistance of any parallel coupled loop (including currents induced in the core). If you are trying to pose an an expert, you aren't doing all that well.
Since I couldn't find a suitable flyback transformer, and I only need
500 volts at 10 mA, a flyback/C-W multiplier makes sense.The MMBD5004S is a cute little dual HV diode in SOT-23. BAS21S is similar.
Not usually a great idea; at HV, the best switch is photo-SCR. It takes a while to turn off, which limits your pulse rate somewhat. Turn-ON and HV tolerance is spectacularly good, though. Triggered spark gaps (krytron, anyone?) work a treat, too.
OK, this has become sort of a background hobby, waiting on the customer to get serious.
The RCD snubber is nice. It softens the rise at the fet drain and dissipates less power than a simple RC that damps ringing about the same.
10 ms of sim runs in about 30 seconds on a wimpy laptop, which is tolerable.I just got four new monster Win11 machines which should be blinding fast with Spice. We're setting up one to look and behave sensibly (ie, like Win7 did) and will clone the other three when I'm happy with that.
We did manage to get the PADS pcb programs to install. That's always an adventure.
Windows keeps getting worse. It has become an ad+annoy vehicle for Microsoft. At boot time, the home screen is a full-screen ad for Edge that won't go away. Explorer drag-and-drop is insane. Screen capture is insane.
I can't make Irfanview the default image viewer. Another annoyance.
I will generate a single pulse, about 5us long, shaped like a half wave sinusoid, at 100kHz repetition frequency. That's why I was looking at flyback, since I can control the voltage delivered simply by adjusting the primary peak current. Power seems to be less than 5W, at 4.5kV peak, duty of let's say 25%. probably below 10mA
I have successfully used power-over-ethernet transformers (Pulse PA1137) for a 12V -> 200V nixie supply. Not sure how they handle 500V.
PA1137 is 1:8, PA1138 is 1:3.43
cu Michael
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.