From: Terry Pinnell
Subject: Re: CircuitMaker spice simulation setup help needed
Date: Wed, 15 Jan 2003 21:07:06 +0000
NNTP-Posting-Date: Wed, 15 Jan 2003 21:07:35 +0000 (UTC)
X-Newsreader: Forte Agent 1.92/32.572
"Smile, it can only get WORSE" wrote:
>You might be sorry you said that...
On Wed, 15 Jan 2003 09:53:38 -0500, you wrote:
>OK, I'm back Terry.
>I am attempting to simulate the attached cockcroft-walton voltage
>multiplier. The load resistors are chose to simulate a photomultiplier tube
>in moderate light (6 ua first dynode, 1 ua seconds dynode, 1/6 ua third
>dynode and 1/36 ua at D output....total current drawn is representative for
>pm tubes in moderate light.
>I have 2 problems....
>I can't even make it run without the load connected. Without a load, the
>caps charge and there should be 100v steps from A through D, with a total
>of 400 volts at D. My simulation shows the signal generator
>malfunctioning...it can't even maintain 100v peak at teh output terminals!
>I must be doing something wrong!
>All caps are 1 uf, but I've tried .05 and .15 uf...no difference. I chose a
>fast recovery diode, so there should not be a frequency response problem
>with the diodes.
>I'm at a loss here.
>Outputs go to mv whenever the load is connected! The values of the load
>resistors are typical for cockcroft walton based supplies driving
>photomultiplier tubes. The voltage increase might not be exactly 100v per
>multipleir stage when loaded, but these values should not load the circuit
>to the point of making it useless!
>I hope I didn't do something really stupid. Drop me a line if you can.
Thanks for your latest e-mail and attached circuit.
I'm no expert on multipliers in general or Cockcroft-Walton's in
particular. (But I recall an Ionizer I made about 20 years ago may
well have been Cockcroft based - must dust it off sometime.) So I had
to go scurrying for books and google. Here are some observations so
I couldn't square your circuit with my reading. For example, this
"The conventional drawing looks like a letter W
Each leg of the W is a rectifier diode, all pointing in the same
east-west direction i.e down cath to up anode: up cathode to down
anode; down cathode to up anode.
Across the top of the letter W, two caps
Across the bottom of the W, one cap.
Transformer connects directly to the top left of the W and via a cap
to the bottom left of the W
Output comes from the top right of the W and via a cap from the bottom
I drew 5 stages of that, converted into verticals and horizontals to
keep CM happy. The result simulates OK, as seen at
In the course of learning a little about these multipliers, I saw that
some regard Schenkel as superior. This seems borne out by
I'll post this to the group too for broader discussion.
Terry, West Sussex, UK