Back to Ideas Page                    .                       Back to Bi-Polar TC Page

1/8.  Dielectric Strength or Breakdown Voltages of Some Gases

These ratings are all relative to nitrogen (1).

*Dielectric gases used in research and industry.  Source:  83rd Edition of the CRC Handbook of Chemistry and Physics, page 15-34

2/8.  Some Sources              

There's a new finding in dielectric gas research:  When you blend different dielectric gases, the sum of their electrical discharge voltages at their respective partial pressures can exceed the electrical discharge voltage of each individual gas at the same temperature and pressure as that of the mixture - synergy.  Patent abstract 4,162,227 describes this. 

3/8.  Dielectric gas mixtures with polar components

L G Christophorou, D R James and R A Mathis  Health & Safety Res. Div., Oak Ridge Nat. Lab., Oak Ridge, TN, USA

Abstract. Dielectric gas mixtures comprising mainly one electron-attaching component and one dipolar component have been investigated. It has been found that polar electron-slowing-down components effect a sharp increase in the breakdown voltage, Vs, with small percentages of electron-attaching additives. The effect of electron-dipole scattering on Vs for multi-component gas dielectrics is assessed, especially in combination with indirect electron scattering via negative ion states. The results demonstrate further the beneficial effect of large electron scattering cross-sections at sub-excitation energies on Vs and suggest that a careful combination of gases slowing down electrons via dipole scattering and via negative ion states can effect large Vs values. This and earlier studies suggest that a number of dielectric gas mixtures containing one or two electron-attaching components from c-C4F8, 2-C4F8, SF6 and a dipolar component from CHF3, CH2F2 or 1,1,1-CH3CF3 with or without N2 are excellent candidates for large-scale testing for possible eventual industrial adoption.

4/8.  Nitrogen breakdown voltage vs. Pressure for small spherical electrodes

5/8.  Marx gas - Febetron (field emission beta ray device)

The spark gap or "Marx" gas used by the Febetron is pressurized Nitrogen (or air which is 78% nitrogen). The reason for using a pressurized gas is that it enables the manipulation of the breakdown voltage of the spark gaps simply by adjusting the pressure of the gas in the module column chamber.  The Febetron uses the principle of a Marx generator to produce high energy impulses.  http://hibp.ecse.rpi.edu/~leij/febetron/febetron.html   

6/8.  Three types of Nitrogen

1.1 This specification covers three types of nitrogen used as an electrical insulating material in electrical equipment:

1.1.1 Type I obtained from the air by liquefaction processes and dried,

1.1.2 Type II obtained from the air by liquefaction processes, deoxidized with hydrogen over a platinum catalyst, and dried, and

1.1.3 Type III obtained from the air by liquefaction processes and if necessary deoxidized by suitable means.

Note 1—The fact that metal containers are filled with materials meeting this specification does not exclude the possibility that the materials might become contaminated with unlisted impurities.

7/8.  Measuring Dielectric Strength

The dielectric strength is measured using U.S. standard ASTM D149 which uses symmetrical electrodes.  In the U.K. standard BS2918, one electrode is a plane and the other is a rod with the axis normal to the plane.

Because the dielectric strength (breakdown voltage) of gases strongly depends on the electrode geometry and surface condition and the gas pressure, it's generally accepted to present the data for a particular gas as a fraction of the dielelectric strength of either nitrogen or sulfur hexafluoride measured at the same conditions.  The table I posted is relative to nitrogen being = 1.00

The CRC Handbook doesn't specify which electrode pair is used but here's the possibilities:

TF-2-50 Type 1 electrodes - per ASTM D149 - 2" diameter electrode with edges rounded to 0.25" radius - for use with the 750-2-D149 or lower voltage units

TF-1-50 Type 2 electrodes - per ASTM D149 - 1" diameter electrode with edges rounded to 0.125" radius - for use with the 750-2-D149 or lower voltage units

TF-.25-50 Type 3 electrodes - per ASTM D149 - 1Ú4" diameter electrode with edges rounded to 0.0313" radius - for use with the 750-2-D149 or lower voltage units

TF-2-75 Type 1 electrodes - per ASTM D149 - 2" diameter electrode with edges rounded to 0.25" radius, 75kV rated - for use with the 775-5-D149 units and 7100-5-D149 units

TF-1-75 Type 2 electrodes - per ASTM D149 - 1" diameter electrode with edges rounded to 0.125" radius, 75kV rated - for use with the 775-5-D149 units and 7100-5-D149 units

TF-.25-75 Type 3 electrodes - per ASTM D149 - 1Ú4" diameter electrode with edges rounded to 0.0313" radius, 75kV rated - for use with the 775-5-D149 units and 7100-5-D149 units

8/8.  Tesla list discussions on gases

1.  Very interesting.  Tetrachloromethane is another name for Carbon Tetrachloride is it not?  Good di-electric - shame it is a nasty carcinogen.  On contact with hot metal (electrodes) it may produce phosgene gas, very lethal.

2.  Most of the halogenated hydrocarbons make fairly decent insulators. Partly because of their density, partly because they have electronegative ions.

3.  You'll note hexafluoroethane (perfluoroethane) and octafluoropropane (aka perfluoropropane, Halocarbon H218) on the list, which are essentially standard alkenes with all the H replaced by F.

4.  Likewise, most of the refrigerants work fairly well.  Dichlorodifluoromethane (R-12) for instance. There are regulatory issues if you're buying them, and the price is high enough that you wouldn't want to vent them willy-nilly, greenhouse gas problems notwithstanding (for what it's worth SF6 is a monster greenhouse gas, although there's some legitimate dispute about whether it's being rated fairly).

5.  R-134a is tetrafluoroethane, readily available, and probably works as well as most of the other halogenated hydrocarbons.

6.  Hi John, thanks for the table. However, there's apparently more to quenching than can be gleaned by comparing relative dielectric strengths. Although dielectric strength will provide an indication of the voltage standoff, this parameter does not indicate how quickly a previously conducting spark gap will recover its dielectric strength (i.e., how well it will quench).  For example, although hydrogen has only about half the dielectric strength of nitrogen, it recovers more quickly from a plasma state to a non-conductive state, giving it superior quenching ability. Hydrogen's small molecules have a higher molecular speed so that heat can be removed more quickly from a recovering gap. Hydrogen was used for high performance multiple-gap switches for spark radio, and it's sometimes even used today for special high-power high rep-rate spark gaps. Hydrogen and deuterium are also the fastest "fill" gases available for high speed thyratron switches, the low pressure evolutionary cousins of spark gaps.

7.  Has anyone tried a DeLaval-type nozzle on a blown/compressed-air gap for Mach++ airflow velocities? Shock wave (sonic boom) quenching?

8.  Don't need a DeLaval nozzle for supersonic flow in a gap, just make the gap the narrowest part, so that's where the supersonic flow is.  The standard Marx blast gap works this way.

9.  Original poster: Bert H.

Obviously the spark gap must be completely immersed in a pressurized pure hydrogen atmosphere to avoid other "interesting" effects. The combustion hazard is eliminated by thoroughly purging the chamber of residual oxygen prior to firing (possibly via an N2 purge). While typical recovery times for common dielectric gases such as air, nitrogen, argon, and SF6 is of the order of 10 milliseconds, hydrogen recovers about one order of magnitude more quickly - about 1 millisecond, and under the right circumstances, recovery times can be pushed down to as little as 100 usec even when switching multi kilo-joule energy levels.

For example, in the April, 1991 issue (volume 38, Issue 4) of IEEE Transactions on Electron Devices, Stuart Moran and Leonard W. Hardesty from the Naval Surface Warfare Center reviewed their work using a pressurized, un-blown hydrogen triggered spark gap (trigatron) with copper-tungsten (Elkonite) main and trigger electrodes. They were easily able to achieve recovery times of ~1 millisecond, and were able to reduce this to 100 usec without forcing gas through the gap. Best results were obtained by undervolting the gap (using ~50% of the self-breakdown voltage) combined with high trigger voltages. Pressures were varied from atmospheric up to 7 MPa (~1000 psig). Although the power levels were MUCH higher than those used in TC's - they were able to achieve 100 usec recovery in a trigatron that handled a 50 kV 170 kA current pulse (12.5 kJ!) by using a trigatron pressurized to 400 psig of H2 - without using flowing gas.

More recent work by a team at the Prather Air Force Research Laboratory (IEEE 12th Pulsed Power Conference, 1999) confirms similar results using a sealed pressurized H2 trigatron. The device can switched 70 Joule pulses at 600 Hz (>40 kW!) with no gas replenishment over a 1 year period.

Looks like hydrogen (or deuterium) is the king for rapid quenching... :^)  Best regards, -- Bert --
 


 Back to Ideas page                     To Top of Page                 Back to Bi-Polar TC Page