Triggering a Marx Generator

Triggering a Marx Generator

This month, we discuss the triggering of Marx generators, or at least how APELC triggers them.

Many years ago, we published a couple of papers at the 2001 IEEE Pulsed Power Conference on the subject of triggering Marx generators including

• Sub-Nanosecond Jitter Operation of Marx Generators
• The Gatling Marx Generator System
• Spark Gap Triggering with Photoconductive Switches

Back in those early days, we were working to develop our Gatling Marx generator concept, and one of the necessary requirements was an extremely low jitter between the launched pulses.  This effort forced us to take a hard look at how we were triggering and study which approach would work best for us.  We reduced the various methods for triggering spark gaps to three, including using an embedded trigatron, laser triggering and the field distortion method.

The Physics of Spark Gap Breakdown

We began with the physics, providing an illustration of the breakdown process, shown below.  We argued that the closure of a spark gap is a statistical process.  However, the breakdown process is sequential. The spark gap is closed by overvolting. Initially, the spark gap voltage is set just below the statistical breakdown level, V_SB. The time for the breakdown process to occur is dependent on four events: (1) the statistical time delay for the appearance of a free electron, t_sd, which may be reduced to zero with the application of a UV source; (2) the streamer formation time, tsf, which is inversely proportional to the electric field; (3) the channel heating time, t_ch, which is also inversely proportional to the electric field; and (4), the trigger pulse risetime, t_r.

We then bring in the various triggering methods.

Trigatron Triggering

This method, shown below, is the simplest.  We simply embed an insulated pin inside the ground electrode.  This pin is then driven by an external pulse, which is typically a pulsed trigger generator, such as APELC’s thyratron-based trigger unit.  The general rule of thumb is that the trigger pulse should be equal in amplitude to the voltage across the main gap, and the pulse should be fast.

Another concept is that the gap length between the primary electrodes and the gap length between the pin and ground should be approximately 2/3 to 1/3.

Along these lines, we have developed two approaches to triggering.  For general triggering we try to drive the trigger pin with a very energetic pulse, with enough voltage to form an arc between the pin and ground, and with enough energy to push this arc out into the primary gap, while producing a lot of photons.  The arc extension effectively shortens the primary gap; thereby increasing the electric field strength of the primary gap, while generating the intense UV, promoting disassociation of electrons in the gas. Alternatively, if we are targeting a very low jitter, we would want a trigger pulse with a pulse magnitude greater than the voltage of the primary gap, and with a rise time that is approximately less than 10X the desired jitter.  In this case, we are working to promote two simultaneous arcs – one forming between the high-side primary electrode to the trigger pin, and a second arc forming between the trigger pin and ground.

Laser Triggering

The laser-triggered spark gap is shown below.  Like the trigatron method, there are at least two methods for laser triggering spark gaps.  One method “boils” electrons of the cathode surface.  A second method creates a plasma near the midpoint of the gap.

APELC is most accustomed to the first method.  In this case, we introduce an energetic laser pulse through a center hole in the anode, focusing the energy onto the opposing cathode.  The typical cathode material is tungsten, which has a very low work function.  When the pulse strikes the cathode, a plasma is created off the cathode.  And much like the trigatron method, the resulting plasma shortens the primary gap; thereby increasing the electric field strength, while also creating a lot of UV photons.

Less known to APELC is the idea of creating plasma points, maybe even short channels, between the primary electrodes.  In this scenario, arcs begin forming from both primary electrodes and can result in low jitter triggering.

Field Distortion Methods

The final method is the field distorting geometry.  And as usual (to this discussion), there are at least two approaches.  APELC has used some variant of the geometry shown below.    In this case, we bias a centered tungsten wire to ½ the voltage of the primary gap.  This essentially makes the wire transparent to the gap, or better, the wire does not affect the electric field lines.  By injecting a fast pulse from the trigger generator, the potential of the tungsten wire changes and can dramatically distort the electric field lines in the gap.  If a negative voltage pulse is applied, simultaneous arcs might form both primary electrodes, resulting in very low jitter.  From APELC’s experience, jitter values are approximately 10x smaller than the rise time of the trigger pulse.

There are variants to the field distortion method, and probably more industry standard.  The more “traditional” geometry is shown below, and commonly referred to as a midplane switch.  Much like APELC’s method, the midplane switch is biased to ½ the voltage across the primary electrodes.  However, the mechanics of the midplane is much different.  In this case, the centered “washer is very thick, allowing for an insulated pin to be inserted from the outer diameter to the center, and a secondary pin is included, and connected to the washer.  In essence, we have a tiny spark gap within the primary gap.

Under steady state conditions, the centered midplane does not disturb the field lines between the primary electrodes; however, with the application of the trigger pulse, an initial arc forms from the insulated trigger pin to the receiving pin, resulting in a healthy amount of UV to aid in the breakdown of the primary gap.  Once this arc forms, the entire midplane assumes the potential of the trigger pulse, and now the electric field of the primary gap is dramatically distorted.  These two events, the UV and the distorted field, result in the closure of the primary gap.

 

What method is chosen?

The desired application drives the choice.  We most often use the trigatron method for our “everyday” use.  Rarely do we employ the laser-triggering method.  And only when we are targeting very high performance do we use the field distortion method.  Again, the applications most often drive the design decisions.