Interesting Twist On Cockcroft-Walton Pump

Cute. Higher efficiency, fewer parts. (Of course a transformer would be even fewer....) ;)

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net

I know :-( ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

=A0 =A0 =A0 =A0...Jim Thompson

ld

Basically a C-W multiplier driven in bridge. Cute. I've seen that several places before, I think in the long-defunct 'Electronics', for one. It's got a name, but I can't remember what.

Thanks for the reminder!

-- Cheers, James Arthur

d

=A0 =A0 =A0 =A0...Jim Thompson

ould

Sorta like a Marx generator, but low-voltage. Cute.

Mark L. Fergerson

d

=A0 =A0 =A0 =A0...Jim Thompson

ould

Is it like a 'digital' full wave C-W multiplier? As shown here,

I couldn't help building one of these on a white proto board a while ago. Lotsa 10uF 100V caps in stock. I built the half wave version. You get a lot of ripple coming through the caps if you make the wrong number of legs.

George H.

We used a Class D stereo amp chip to drive a transformer, and then the multiplier once.

The twist is not on the Cockroft-Walton, it is on the pump driving it.

I have seen this circuit before. I have also seen a multiplier driven by the pins of a counter chip. Picking a few outputs of a CD4060 that was needed for other reasons can give you N*Vdd for N diodes.

There is also this circuit:

Vcc->!----+---------!!----------------------------------- \ ! / G2 Vcc->!-----+--------------- ! \ ! ! ! ! ! ! !!--Vcc =3D=3D=3D !!- ! +-----+----!! ! G2---!! ! ! !!--+-------- !!------+-+--- ! ! ! !!----+--+-!

I know it as the Schenkel multiplier. Here's a reference:

M. Schenkel, "Eine neue Schaltung fur die Erzeugerung hoher Gleichspannungen", Electrotechnische Zeitschrift, Vol.40, No.28, 1919, pp333-334

Regards, Jeroen Belleman

^^^^^^^^^^^ Erzeugung.

Fingers running too fast. Sigh.

Jeroen Belleman

Bunch of things I'd do different, though... the 1N4001 is overkill for a milliamp pump (maybe a signal diode or fast-recovery rectifier instead), and it's awkward to use 1 kHz; we can boost that to 50 kHz and drop the capacitors from

4.7uf to 0.1 uF ceramics. Then there's the terrible conduction-angle for the square wave drive into a capacitive load: the original Cockcroft- Walton design used sinewave drive, of course... one could limit both driver outputs with a bit of a choke (ferrite beads); at 'sustain' it won't have much effect, but it keeps the high-frequency current to the power rails limited during that fast ramp-up at turnon time.

Efficiency (yeah, I know, not really important here) would benefit from Schottky rectifiers, and power drive would be better with something differential that has higher output current (NE556 would get two orders of magnitude in output current drive, if your power filtering can stand it). Regulated +5 often means an unregulated +9V is available, I'd try to take power from the

+9V (but that's not an option with 74HC04 drive).

und

=A0 =A0 =A0 =A0 =A0...Jim Thompson

would

ator

Yeah, kind of like that.

Here's a variation on Jim's, a x6 multiplier with the advantage that only C6 sees the full output voltage. Caps C1-C5 only see 2*Vdd; ditto D1-5. That makes for cheaper caps and diodes, or higher output voltages, take your pick.

.__.-Vdd _| |_ 0 C1 C3 C5 clk>-+----||---+-------||------+-------||------. | | | | | Vdd | | | | -+- | | | | | D1 | D2 D3 | D4 D5 | D6 | '-|>|--+--|>|--+--|>|--+--|>|--+--|>|--+--|>|--+----> Vo | | | | | | | --- C6 | |\ C2 | C4 | --- '-----| >o---||---+-------||------' | |/ =3D=3D=3D

For f(in) =3D 25kHz, Vdd =3D 9v, and C1-C6 =3D 10nF, this circuit will driv= e a 10M load to 50V.

-- Cheers, James Arthur

und

=A0 =A0 =A0 =A0 =A0...Jim Thompson

would

Yep. It's fun multiplying voltages--it always feels like magic somehow.

-- Cheers, James Arthur

found

=A0 =A0 =A0 =A0 =A0...Jim Thompson

er would

at

erator

.

C5

=A0 =A0 =A0 =A0 |

=A0 =A0 =A0 |

=A0 =A0 =A0 |

=A0 | =A0D6

=A0 | =A0 =A0 =A0 =A0 =A0 =A0 =A0 |

=A0 | =A0 =A0 =A0 =A0 =A0 =A0 =A0--- C6

=A0 =A0 =A0 =A0 =A0 =A0---

=A0 |

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =3D=3D=3D

ive

Ohh, that's nice. No reason not to drive it from 'say' a 20V sine wave and get a bit more voltage out of it.

George H.

Was curious so did a SPICE emulation and got a VERY crappy "multiplier". So what is wrong? Version 4 SHEET 1 888 680 WIRE 560 64 336 64 WIRE 128 80 -80 80 WIRE 144 80 128 80 WIRE 272 96 272 80 WIRE 560 96 560 64 WIRE -80 128 -80 80 WIRE 208 144 208 80 WIRE 336 144 336 64 WIRE 128 208 128 80 WIRE 144 208 128 208 WIRE 272 208 272 144 WIRE -80 224 -80 208 WIRE 560 240 560 176 WIRE 208 272 208 208 WIRE 336 272 336 144 WIRE 272 336 272 272 WIRE 144 352 144 336 WIRE 208 384 208 336 WIRE 304 384 208 384 WIRE 400 384 304 384 WIRE 304 480 304 464 FLAG 272 96 0 FLAG 144 352 0 FLAG -80 224 0 FLAG 304 480 0 FLAG 400 384 OUT IOPIN 400 384 Out FLAG 560 240 0 SYMBOL diode 208 96 R270 WINDOW 0 32 32 VTop 0 WINDOW 3 65 105 VBottom 0 SYMATTR InstName D1 SYMATTR Value MURS120 SYMBOL diode 272 128 R90 WINDOW 0 0 32 VBottom 0 WINDOW 3 -5 125 VTop 0 SYMATTR InstName D2 SYMATTR Value MURS120 SYMBOL diode 208 224 R270 WINDOW 0 32 32 VTop 0 WINDOW 3 44 125 VBottom 0 SYMATTR InstName D3 SYMATTR Value MURS120 SYMBOL diode 272 256 R90 WINDOW 0 0 32 VBottom 0 WINDOW 3 -7 130 VTop 0 SYMATTR InstName D4 SYMATTR Value MURS120 SYMBOL diode 208 352 R270 WINDOW 0 32 32 VTop 0 WINDOW 3 33 130 VBottom 0 SYMATTR InstName D5 SYMATTR Value MURS120 SYMBOL cap 208 64 R90 WINDOW 0 0 32 VBottom 0 WINDOW 3 -25 83 VTop 0 SYMATTR InstName C1 SYMATTR Value 4.7µ SYMBOL cap 336 128 R90 WINDOW 0 0 32 VBottom 0 WINDOW 3 4 -42 VTop 0 SYMATTR InstName C2 SYMATTR Value 4.7µ SYMBOL cap 208 192 R90 WINDOW 0 0 32 VBottom 0 WINDOW 3 -26 82 VTop 0 SYMATTR InstName C3 SYMATTR Value 4.7µ SYMBOL cap 336 256 R90 WINDOW 0 0 32 VBottom 0 WINDOW 3 -2 -41 VTop 0 SYMATTR InstName C4 SYMATTR Value 4.7µ SYMBOL cap 208 320 R90 WINDOW 0 0 32 VBottom 0 WINDOW 3 -25 80 VTop 0 SYMATTR InstName C5 SYMATTR Value 4.7µ SYMBOL Misc\\signal -80 112 R0 WINDOW 3 -121 125 Left 0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR Value PULSE(0 5 0 10n 10n 1m 2m 10000) SYMATTR InstName V1 SYMBOL res 288 368 R0 SYMATTR InstName R1 SYMATTR Value 20K SYMBOL Misc\\signal 560 80 R0 WINDOW 3 -121 125 Left 0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR Value PULSE(5 5 0 10n 10n 1m 2m 10000) SYMATTR InstName V2 TEXT -128 376 Left 0 !.tran 0 100m 0 1m TEXT -224 424 Left 0 !.ic V(N004)=-4.0 V(N006)=-8.0 V(OUT)=-7.5V TEXT -224 464 Left 0 ;IC set to appx final

found

=A0 =A0 =A0 =A0 =A0...Jim Thompson

er would

at

erator

.

C5

=A0 =A0 =A0 =A0 |

=A0 =A0 =A0 |

=A0 =A0 =A0 |

=A0 | =A0D6

=A0 | =A0 =A0 =A0 =A0 =A0 =A0 =A0 |

=A0 | =A0 =A0 =A0 =A0 =A0 =A0 =A0--- C6

=A0 =A0 =A0 =A0 =A0 =A0---

=A0 |

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =3D=3D=3D

ive

The capacitors C1 and C2 need to have a larger value for a given output current. This eats into the cost savings that there may be.

I found

f

=A0 =A0 =A0 =A0 =A0 =A0...Jim Thompson

rmer would

that

or

enerator

on.

=A0C5

=A0 =A0 =A0 =A0 =A0 |

=A0 =A0 =A0 =A0 |

=A0 =A0 =A0 =A0 |

5 =A0 | =A0D6 o

=A0 | =A0 =A0 =A0 =A0 =A0 =A0 =A0 |

=A0 | =A0 =A0 =A0 =A0 =A0 =A0 =A0--- C6

=A0 =A0 =A0 =A0 =A0 =A0 =A0---

=A0 |

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =3D=3D=3D

drive

Or better yet make it 10 diodes long and drive it with 100V.

[snip]

V2 pulse description is wrong... you have it as a constant +5V, no pulsing ;-) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

I found

f

=A0 =A0 =A0 =A0 =A0 =A0...Jim Thompson

rmer would

that

or

enerator

on.

=A0C5

=A0 =A0 =A0 =A0 =A0 |

=A0 =A0 =A0 =A0 |

=A0 =A0 =A0 =A0 |

5 =A0 | =A0D6 o

=A0 | =A0 =A0 =A0 =A0 =A0 =A0 =A0 |

=A0 | =A0 =A0 =A0 =A0 =A0 =A0 =A0--- C6

=A0 =A0 =A0 =A0 =A0 =A0 =A0---

=A0 |

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =3D=3D=3D

drive

uF are usually cheap!, 'volts' not so much. But yeah, parts cost isn't that big a deal.

I made zero effort to optimize this, other than minimizing my time spent on it.

-- Cheers, James Arthur

When i originally set it up, a probe there showed V2 as being opposite in phase to V1. But will re-check.. Thanks.