Magnetic Quenching of Spark Gaps
From: Greg Leyh[SMTP:lod-at-pacbell-dot-net]
Sent: Sunday, September 21, 1997 5:58 AM
To: Tesla List
Subject: Re: Magnetic Quenching of Spark Gaps
Tim Raney wrote:
> Magnetic quenching was (and is) used to "blow out" DC arcs in heavy-duty
> switches used to switch inductive loads. Same principle with the spark gap
> in the Tesla coil tank circuit. Tesla used this method in his work (Tesla,
> 1894) and details are available in the older electrical engineering texts.
But isn't the desired mode of quenching _very_ different for circuit breakers
and rotary gaps?
In a CB, the objective is to _immediately_ quench any arc as soon as it forms,
in order to reduce the pitting of the electrodes.
A stationary magnetic field around the arc is suitable for this, since the free
ions and electrons are rapidly moving, and F = qvB will tend to scatter them.
But in a TC rotary, the goal is to provide an undisturbed, low impedance path
for the primary current _during the entire 1st beat envelope_, and then somehow
clear the lingering free electrons and ions afterward, before the 2nd beat
envelope can develop an EMF across the gap.
Since the lingering free electrons and ions after the envelope have little net
velocity, it would seem that a DC or permanent magnetic field could not provide
significant clearing action, and would only serve to disturb the arc during the
1st beat envelope, when high currents are present.
Therefore, if magnetic quenching in a TC rotary is to be effective, the magnetic
field must be pulsed, either by permanent magnets that ride on the rotor, or by
a pulsed, off-axis electromagnet surrounding the gap area.
> For magnetic quenching to be effective, the magnetic field must be
> perpendicular to the spark gap axis.
This is true for clearing a high current arc. However, in order to clear a
relatively stationary cloud of charged particles from the gap area, the axis
of the pulsed solenoid should be parallel to, but offset from, the axis of
the gap electrodes. This arrangement would send the offending particles on
a trajectory 90 degrees from the gap axis.
> I think there is more work to do along these lines.