Re: Marx generator (fwd)

["High Voltage list" <hvlist@xxxxxxxxxx>]

Original poster: Steven Roys <sroys@xxxxxxxxxx>



---------- Forwarded message ----------
Date: Tue, 22 Nov 2005 16:22:59 -0700
From: D.C. Cox <resonance@xxxxxxxxxx>
To: High Voltage list <hvlist@xxxxxxxxxx>
Subject: Re: Marx generator (fwd)

In the real world 76 kV/In. is too high (ideal) value.

Use 26.5 kV/cm for accurate results.  This equates to 67.31 kV/Inch.

This is also very useful for measuring the peak potential of a resonance 
(Tesla) transformer when operated in the 1 shot/10 sec pulse mode.  Peak 
potential spark length is independent of waveform (not rms but peak). This 
assumes rounded electrodes of a suitable dia. for the task at hand.

Ref: High Voltage Engineering by M.S. Naidu & V. Kamaraju   McGraw-Hill 
1995.

Dr. Resonance
>
>
>
> ---------- Forwarded message ----------
> Date: Mon, 21 Nov 2005 14:26:32 -0800
> From: Jim Lux <jimlux@xxxxxxxxxxxxx>
> To: High Voltage list <hvlist@xxxxxxxxxx>
> Subject: Re: Marx generator (fwd)
>
> At 10:55 AM 11/21/2005, High Voltage list wrote:
>>Date: Fri, 18 Nov 2005 13:11:52 +0100
>>From: Herwig Roscher <herwig.roscher@xxxxxx>
>>To: hvlist@xxxxxxxxxx
>>Subject: Marx generator
>>
>>
>>Q1:
>>The distance, a voltage of 1 kV can jump, usually is said to be
>>0.013". For Marx generators I found values of about 0.04" as well.
>
> 0.013" would be 76 kV/inch, which is approximately the "uniform field"
> breakdown voltage.
>
> Your gap will be far from a uniform field.
>
>>Other Marx builders claim:
>>
>>0.052"          Rapp
>>0.045"          Steve Ward
>>0.052"          Kronjaeger
>>0.050"          Mike Gilchrist
> These are roughly 3 times the uniform field gap, implying a strongly
> non-uniform field (e.g. a needle gap or close to it).
>
>
>>My system jumps only 0.027"/kV. Why? Is this a question of the form
>>of the electrodes?
>
> Exactly that.. what are you using for gap electrodes?
>
>
>
>>Q2:
>>Other Marx builders get:
>>
>>1.75"/Joule             Rapp
>>1.92"/Joule             Steve Ward
>>1.83"/Joule             Mike Gilchrist
>>
>>I only get 0.9"/Joule. Again: Why?
>
> Spark length is dependent on a lot of things other than energy.
>
> Voltage for one
> Waveform for another
> Shape of the test gap for a third
>
> You might have significant series inductance or resistance.
>
> If all your gaps aren't firing together, you might have a lower voltage,
> longer duration pulse.
>
> What sort of gap are you using for your HV gap?  Many experimenters use
> what is essentially a point/plane gap (i.e. a sharp wire end against a
> ground plane).
>
>
>
>>Q3:
>>During and after completion of the bank loading there is no sizzling
>>noise. Therefor I assume, that there is little or no corona at 20 kV.
>
> Good assumption..
> You can also look at it in the dark.
>
>>Is it possible, that at ignition of the bank there is too much corona
>>at 500 kV?
>
> Sure, anything is possible
>
>>What are the secrets of Marx tuning? Any help would be appreciated
>>very much.
>
> Did you remember to sacrifice a goat while facing precisely 14 degrees 
> west
> of true north?
> <grin>
>
> But really, these things are highly idiosyncratic.  Seemingly small 
> changes
> in things like gaps can have huge effects on the output.  Getting all the
> gaps to fire together is real important, so you want a good optical path
> from the first gap (which can be slightly smaller, so it fires first) to
> all the rest. The strong UV from the spark on the first gap helps 
> breakdown
> the rest of the gaps.
> Using decent gap electrodes helps.  One of my Marxes used the "twist of
> wire in a loop" electrodes and it was very erratic.  When I went to
> smoothly rounded drawer pulls, it worked a LOT better and more
> consistently.  I think it has more to do with the uniformity of the 
> fields.
>
> Good luck..
>
>
>
>
>