Flybuck Cockroft-Walton

Hi, All,

This is for a flexible photoreceiver board intended for use with PIN diodes, silicon and InGaAs APDs in both linear and SPAD mode, and MPPCs. We've been doing a fair number of proof-of-concept systems for technology licensing, and it would be nice to have a few dozen front end boards that can be used for everything from nanowatt PIN diode detectors to an MPPC front end that will go from photon counting to about 3 mA anode current in one range.

The bias generator is this C-W thing (.asc appended below) followed by an op amp controlled HV transistor voltage regulator. It's driven by the switch pin of a TPS560430F 1 MHz sync buck which also generates a

-6V rail from 24V, so the input waveform has an 80% duty cycle. Unlike a lot of sync bucks, this one actually does keep PWMing right down to zero load. (I got bitten by a lying datasheet last time I tried doing this--now I never use a new switcher without getting the eval board.)

Using pop options, it'll make anything from near zero down to -250V. Silicon APDs often need 220V or thereabouts, MPPCs are often around 50-70.

Doing one fancy board like this means that we don't have to pay to build N versions of more or less the same technology--we pull one out of the bin, mod it for the application, and away we go.

This is the high-voltage version, which needs to have its short-circuit current limited to a safe value. (There's also a LND150 in series with the amp output, as well as the 600 mA cycle-by-cycle current limit of the buck, so this is belt-and-suspenders.)

The main bit of ugliness is ringing caused by the inter-winding capacitance of the transformer, but it looks like it'll work pretty well.

Cheers

Phil Hobbs

------------- Version 4 SHEET 1 1868 692 WIRE -512 80 -608 80 WIRE -512 96 -512 80 WIRE -208 112 -336 112 WIRE -176 112 -208 112 WIRE -80 112 -176 112 WIRE 16 112 0 112 WIRE 144 112 96 112 WIRE 272 112 208 112 WIRE 416 112 272 112 WIRE 608 112 480 112 WIRE 752 112 608 112 WIRE 960 112 816 112 WIRE -208 160 -208 112 WIRE -336 176 -336 112 WIRE -608 208 -608 80 WIRE 96 240 64 240 WIRE 176 240 160 240 WIRE 272 240 272 112 WIRE 272 240 240 240 WIRE 304 240 272 240 WIRE 448 240 432 240 WIRE 464 240 448 240 WIRE 608 240 608 112 WIRE 608 240 592 240 WIRE 656 240 608 240 WIRE 800 240 784 240 WIRE 816 240 800 240 WIRE 960 240 960 112 WIRE 960 240 944 240 WIRE 1008 240 960 240 WIRE 1152 240 1136 240 WIRE 1200 240 1152 240 WIRE 1312 240 1280 240 WIRE 1440 240 1312 240 WIRE 1520 240 1440 240 WIRE 1568 240 1520 240 WIRE -336 256 -336 240 WIRE -208 256 -208 240 WIRE -208 256 -336 256 WIRE -208 272 -208 256 WIRE -512 288 -512 176 WIRE -336 288 -336 256 WIRE 64 304 64 240 WIRE 1312 304 1312 240 WIRE 1440 304 1440 240 WIRE 1568 304 1568 240 WIRE -608 320 -608 288 WIRE -512 368 -512 352 WIRE -336 368 -336 352 WIRE -336 368 -512 368 WIRE -208 368 -208 352 WIRE -208 368 -336 368 WIRE -208 384 -208 368 WIRE 320 384 272 384 WIRE 448 384 448 240 WIRE 448 384 384 384 WIRE 624 384 448 384 WIRE 800 384 800 240 WIRE 800 384 688 384 WIRE 944 384 800 384 WIRE 1152 384 1152 240 WIRE 1152 384 1008 384 WIRE -336 400 -336 368 WIRE 1440 416 1440 368 WIRE 1568 416 1568 384 WIRE 272 464 272 384 WIRE 1312 464 1312 384 WIRE 1264 480 1088 480 WIRE 1088 496 1088 480 WIRE -336 512 -336 464 WIRE -208 512 -208 464 WIRE 1264 560 1264 528 WIRE 1312 576 1312 544 WIRE 1088 592 1088 576 FLAG 64 304 0 FLAG 1568 416 0 FLAG 1440 416 0 FLAG -208 512 0 FLAG 272 464 0 FLAG 272 112 M0 FLAG 608 112 M1 FLAG 960 112 M2 FLAG 448 240 S0 FLAG 800 240 S1 FLAG 1152 240 S2 FLAG -176 112 top FLAG 1312 576 0 FLAG 1264 560 0 FLAG 1088 592 0 FLAG 1520 240 out FLAG -608 320 0 FLAG -336 512 0 SYMBOL voltage -608 192 R0 WINDOW 3 -446 -167 Left 2 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR Value PULSE(0 30 1n 10n 10n 790n 1000n) SYMATTR InstName V2 SYMBOL res 112 96 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R1 SYMATTR Value 1m SYMBOL cap 1456 304 M0 SYMATTR InstName C7 SYMATTR Value 2.2n SYMBOL cap 208 96 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C4 SYMATTR Value {Cm} SYMBOL ind -96 128 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 5 56 VBottom 2 SYMATTR InstName L1 SYMATTR Value {Lr} SYMBOL ind2 -224 480 M180 WINDOW 0 -46 60 Left 2 WINDOW 3 -65 29 Left 2 SYMATTR InstName L3

SYMATTR Type ind SYMATTR SpiceLine Rser=.8 SYMBOL ind2 -224 368 M180 WINDOW 0 -47 78 Left 2 WINDOW 3 -74 27 Left 2 SYMATTR InstName L4

SYMATTR Type ind SYMATTR SpiceLine Rser=.8 SYMBOL ind2 -224 256 M180 WINDOW 0 -45 88 Left 2 WINDOW 3 -61 33 Left 2 SYMATTR InstName L5

SYMATTR Type ind SYMATTR SpiceLine Rser=.8 SYMBOL cap -528 288 R0 WINDOW 3 -55 103 Left 2 SYMATTR InstName C9 SYMATTR Value {10*Cm} SYMBOL cap 480 96 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C2 SYMATTR Value {Cm} SYMBOL cap 384 368 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 39 32 VTop 2 SYMATTR InstName C1 SYMATTR Value {Cm*10} SYMBOL cap 816 96 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C3 SYMATTR Value {Cm} SYMBOL cap 688 368 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 42 25 VTop 2 SYMATTR InstName C5 SYMATTR Value {Cm*10} SYMBOL cap 1008 368 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 37 32 VTop 2 SYMATTR InstName C10 SYMATTR Value {Cm*10} SYMBOL res 1584 288 M0 SYMATTR InstName R5 SYMATTR Value 200k SYMBOL res -528 80 R0 SYMATTR InstName R2 SYMATTR Value 2.5 SYMBOL sw 1312 560 M180 SYMATTR InstName S1 SYMATTR Value finger SYMBOL res 1328 288 M0 WINDOW 0 51 27 Left 2 WINDOW 3 57 59 Left 2 SYMATTR InstName R6 SYMATTR Value 2k SYMBOL voltage 1088 480 R0 WINDOW 3 -438 76 Left 2 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR Value PULSE(0 2 0.6m 10n 10n 1) SYMATTR InstName V3 SYMBOL ind 1184 256 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 -2 56 VBottom 2 SYMATTR InstName L6

SYMBOL diode 368 224 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 39 34 VTop 2 SYMATTR InstName D1 SYMATTR Value BAS70 SYMBOL diode 432 224 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 63 39 VTop 2 SYMATTR InstName D3 SYMATTR Value BAS70 SYMBOL diode 528 224 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 38 23 VTop 2 SYMATTR InstName D4 SYMATTR Value BAS70 SYMBOL diode 592 224 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 63 15 VTop 2 SYMATTR InstName D5 SYMATTR Value BAS70 SYMBOL diode 720 224 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 34 32 VTop 2 SYMATTR InstName D6 SYMATTR Value BAS70 SYMBOL diode 784 224 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 58 32 VTop 2 SYMATTR InstName D7 SYMATTR Value BAS70 SYMBOL diode 880 224 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 40 20 VTop 2 SYMATTR InstName D8 SYMATTR Value BAS70 SYMBOL diode 944 224 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 60 0 VTop 2 SYMATTR InstName D9 SYMATTR Value BAS70 SYMBOL diode 240 224 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 59 28 VTop 2 SYMATTR InstName D2 SYMATTR Value BAS70 SYMBOL diode 160 224 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 36 32 VTop 2 SYMATTR InstName D10 SYMATTR Value BAS70 SYMBOL diode 1072 224 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 38 30 VTop 2 SYMATTR InstName D11 SYMATTR Value BAS70 SYMBOL diode 1136 224 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 64 33 VTop 2 SYMATTR InstName D12 SYMATTR Value BAS70 SYMBOL cap -320 176 M0 SYMATTR InstName C8 SYMATTR Value {Cw} SYMBOL cap -320 288 M0 SYMATTR InstName C6 SYMATTR Value {Cw} SYMBOL cap -320 400 M0 SYMATTR InstName C11 SYMATTR Value {Cw} TEXT 264 8 Left 2 !.tran 0.7m TEXT 400 456 Left 2 !.param Cm=270pF\n;.step param Cm 200p 600p 100p TEXT 520 -64 Left 2 !;\n.param Lr=10u\n;.step param Lr 3.3u 23u 2u TEXT -120 424 Left 2 !k1 L3 L4 L5 0.9987 TEXT 1208 632 Left 2 !.model finger sw(Vt=1 Vh=-0.5) TEXT 1008 -80 Left 2 ;QL02 Flexible Photoreceiver:\nCockroft-Walton Supply for APDs\n \nPhil Hobbs,\nElectroOptical Innovations LLC\n \nRev

1, December 3rd, 2019 TEXT -72 624 Left 2 !.lib \\electronics\\SpiceModels\\PHParts.lib TEXT -128 472 Left 2 !;\n.param Cw 18p\n;.step param Cw list 1f 10p 20p TEXT 1464 472 Left 2 ;Short circuit current\n is less than 14 mA,\nnot counting the \nbuck's current limit TEXT -304 56 Left 2 ;PE2004NL TEXT 712 600 Left 2 ;Short-circuit applied at \nt = 600 us TEXT -880 440 Left 2 ;80% duty cycle rectangle wave\nmimics +24 -> -6V switcher\n(24 / [24+6] = 0.8). LINE Normal -176 464 -176 160 2 LINE Normal -176 464 -160 464 2 LINE Normal -176 160 -176 464 LINE Normal -176 160 -176 160 LINE Normal -176 160 -176 160 LINE Normal -160 464 -160 160
--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs
Loading thread data ...

The usual solution to interwinding capacitance in the transformer is the Baxandall class-D resonant inverter.

formatting link

This circuit was intended to generate higher voltages, but it's the same basic idea.

The proposition is much the same - control the voltage output by controlling the make-to-space ration of the low voltage drive to the step-up inverter.

The nice part of my circuit is the feed inductor for the resonant inverter (L5 plus the bead L14) smooths out the PWM waveform before it get to the inverter, so no spikes.

--
Bill Sloman, Sydney
Reply to
Bill Sloman

[...]

  1. Fails on every wrapped line

  2. missing electronics\SpiceModels\PHParts.lib. Is this available on your web site?

  1. every inductor value appears as "µ"

  2. Personal Preference : I like to name every node usuing the ref des of the attached components. That way I don't get node names like NXXXX that change every time you revise the circuit.
Reply to
Steve Wilson

Right, sorry about that--I just needed the BAS70 model, and forgot I'd been lazy and put in my main library. A fixed version is at

formatting link

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

This is a nice dual 400V diode in SOT23, 5 cents each:

DIODES INC MMBD5004S-7 ROHS DUAL SERIES KJB

Nice fast recovery, too. Some HV diodes aren't.

This does 1200 volts, with a DRQ127 dual inductor and four dual diodes.

formatting link

It's an autotransformer flyback, whatever that's officially called.

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

Thanks. That is the slickest way I have seen to get model statements and long lines into a newsgroup posting.

It works because they don't appear in the newsgroup posting!

Now I can use all my favorite 1N and 2N parts from LTspice IV and have it run in XVII. That's Great!

Reply to
Steve Wilson

50 ns, not bad at all for 400V. I had this 1 MHz sync buck handy, so I wanted to stay with Schottkys for this one.

Dunno. Useful topology though. My C-W will be sharing a 25x40 mm shield can with the sync buck. (It might all fit in 25 mm square, we'll see.)

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

I suppose I've shown you my high-voltage-optocoupler totem-pole regulator.

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

Phil Hobbs wrote in news: snipped-for-privacy@electrooptical.net:

Gap the core. 1 mil transformer tape layer. Helps at zero crossing too.

Note that I did not say to get or use a factory "gapped core". They are usually very unreliable for consistency of the gap and the gap is only the center (pot cores). However, gapping it yourself at 1 mil (for starts) will make the oscillator happy. You might find it very happy operating around 56kHz too.

Reply to
DecadentLinuxUserNumeroUno

Yup. This one needs sub-nanovolt noise density though--the bias supply noise gets differentiated just like the bootstrap/TIA voltage noise.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

Too much like work. The AC comes from the switch of a 1-MHz sync buck regulator.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

Phil Hobbs wrote in news:qs829l$oqs$ snipped-for-privacy@dont-email.me:

I thought that since you were so concerned with inter-winding capacitance that you were winding your own transformers.

Hard to imagine an off the shelf product exhibiting ringing, since that would be a product/sales fail.

Adding a few layers of tape on top of the secondary before winding the primary onto it will reduce it too.

Reply to
DecadentLinuxUserNumeroUno

Can you fill up a large capacitor and shut off the C-W while you are making measurements? The linear regulator can hold the bias voltage constant while the cap discharges. Presumably you would need this only on the low current ranges where noise is critical, and not on the higher current ranges where it would be harder to keep the capacitor voltage high enough.

Reply to
Steve Wilson

Ringing comes from the way you drive the transformer.

Gapping the core can sometimes help, but isn't any kind of universal cure-all, and mostly won't do much at all.

If you haven't got the inductance specified from a factory-ground gapped core-pair (in inductance per root turn), you probably didn't put it together right - the ungapped mating faces have to be clean

It can reduce the inter-winding capacitance - which is calculable and measurable - but this doesn't usually make much difference.

--
Bill Sloman, Sydney
Reply to
Bill Sloman

PA2004NL, right?

Have you measured its interwinding capacitances yet? The leakage inductance is very low, 0.2%, so it must be trifilar wound, not good.

--
 Thanks, 
    - Win
Reply to
Winfield Hill

I've used it before at 2 MHz, and it worked fine. The 18 pF is a guess.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

It isn't hard to get the switching junk below the amp noise floor once you've done it a few times. A board-level shield, good ground stitching, and careful placement of vias (if both sides are populated) is enough, at least at the sorts of currents I use (a few amps at most).

And part of the 'Flexible Photoreceiver' shtick is that it can run at several mA (at lower voltage, obviously, so the photodiode doesn't turn to lava).

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

I've never wound a transformer. Coils for RF, lots.

Lots of designs ring in discontinuous conduction mode (DCM).

The transformer is a Pulse PA2004NL, which is only 7 x 9 x 4 mm.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

Tell us about L1, a series resonator?

--
 Thanks, 
    - Win
Reply to
Winfield Hill

It improves the output voltage a bit and reduces the peak current at start-up. It does a lot more with a more symmetrical drive waveform, but 80% duty cycle I can get for free, so oh well.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

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.