Spark gap electrode wear paper

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Lau, Gary by way of Terry Fritz <twftesla-at-uswest-dot-net>" <Gary.Lau-at-compaq-dot-com>

The recent thread on the collapsable secondaries reminded me to take a look
at the papers that I downloaded from the iop-dot-org web site before it
restricted access.  Pertinant to the more recent thread on SS electrode wear
was a very readable paper comparing spark gap electrode between brass,
aluminum, stainless steel, and elconite (a copper/tungsten alloy)
electrodes.  A surprising finding was that the discharge repetion rate
affected which material was superior.  The experimental study only compared
rep rates of 10HZ and 1KHz, well below and above what we typically see in
our applications, so it's difficult to know which rate is more applicable.
The gap discharged a 3 nF capacitor charged to 22.5kV.

Most remarkably, the authors concluded that of the 4 materials tested, BRASS
was judged to have "the best overall performance", and stainless steel the
worst!  Too bad that pure tungsten wasn't tested.  One of the factors cited
as affecting errosion is the resistivity of the materials, so this makes me
wonder if tungsten might be inferior to the presumably more conductive
elconite alloy.

I have posted the paper on my web site at:
http://people.ne.mediaone-dot-net/lau/tesla/electrodeerrosion.pdf 

I have also included the closing summary of the paper below.

Regards, Gary Lau
Waltham, MA USA

Electrode erosion and lifetime
performance of a high repetition rate,
triggered, corona-stabilized switch in
air - J M Koutsoubis and S J MacGregor
J. Phys. D: Appl. Phys. 33 (2000) 1093-1103. Printed in the UK

Conclusions
The electrode erosion mechanism of a sealed rod-plane
TCS switch was investigated and the device lifetime was
determined for electrodes made of elkonite, brass, stainless-steel
and aluminium. The switch was filled with air at a
pressure of 3.0 bar absolute, and operated at repetition rates
of 10 Hz and 1 kHz. The rod-plane geometry was selected
because it possesses the lowest lifetime of the TCS switch
designs. The erosion of the rod (anode) electrode of the
TCS switch was found to play a major role on the lifetime
of the device, and this lifetime was found to be inversely
proportional to the erosion rate of the anode electrode. This
dependence was found to be valid at both 10 Hz and 1 kHz
operation for all the materials tested. The erosion rate of the
trigger disc (cathode) electrode, was not found to have any
significant influence on the lifetime of the switch.
The anode electrode materials showed a different erosion
behaviour between the 10 Hz and 1 kHz tests. For the 10 Hz
tests, the performance of most of the tested materials was
closely associated to their thermo-physical properties, where
generally high melting temperature materials exhibited the
lowest erosion rates. This situation was almost inverted for
the 1 kHz tests. At this repetition rate, it is believed that
the erosion mechanism was predominantly influenced by the
melting temperature of metallic oxides formed on the rod tips.

Brass showed the best overall performance, with low an-ode
erosion rates and very good lifetime performance at both
10 Hz and 1 kHz tests. At a PRF of 10 Hz it showed a good
anode erosion rate, as well as switch lifetime. On the other
hand, at 1 kHz, brass outperformed all the tested materials
by exhibiting a very good anode erosion rate and a lifetime
of more than 2.11 10E6  shots. Elkonite exhibited the lowest
anode erosion rates and consequently the highest lifetime of
more than 7.13 10E5 shots at a PRF of 10 Hz, but displayed
the opposite behaviour in the 1 kHz tests. Aluminium dis-played
a very poor erosion rate and the lowest switch lifetime
at 10 Hz, while at 1 kHz it showed a good anode erosion rate
and switch lifetime. Lastly, stainless-steel exhibited poor
performance for both the 10 Hz and 1 kHz tests, and was
clearly the most inferior of the electrode material tested.

The results indicate that no single electrode material
exhibited superior performance at both the 10 Hz and 1 kHz
repetition rates. Elkonite was superior to the rest of the tested
materials at 10 Hz, but at a PRF of 1 kHz, the results showed
that in the rod-plane switch configuration it is not a promising
material. On the other hand, brass showed a surprisingly
good and consistent performance at both repetition rates.
This makes brass a good and low-cost, all-around anode
material for the given rod-plane switch geometry. The only
major disadvantage of brass is that it is highly susceptible
to corrosive gas by-products formed by the operation of the
TCS switch with air.

Finally, the operational consistency of the TCS switch
was found to be affected mainly from the surface conditions
of the rod electrode and deterioration of the filling gas.
Sparking modifies the electrode tip, causing a change in the
mode of the corona discharge, ranging from a steady glow
type to a fluctuating filamentary type. At the same time the
properties of the gas are altered by the production of ozone
and nitrogen oxides. These effects were found to introduce
inconsistent operation (misfiring/prefiring) periods during a
long test run and to degrade the performance of the switch.