The voltage needed to jump a gap can be estimated by Paschen's Law https://en.wikipedia.org/wiki/Paschen%27s_law An arc exhibits a negative slope of resistance. The equation is something like V=3DK0+K1*I, K0 is a constant term which depends on the geometry of th= e electrodes. K0 is positive and fairly low compared to the breakdown voltage - tens to hundreds of volts is common. K1 is negative and represents the fact that higher current flow reduces the voltage drop (although the total drop will always be the proper sign for the power input to the arc to be positive, at least in steady-state conditions). This happens because higher current flow raises the arc plasma to a higher temperature, which then allows a wider plasma channel and more ionization within it. A good rule of thumb in air at 1 atm pressure is that 300V is the minimum DC or peak AC voltage for jumping any air gap, no matter how small (until you get within nanometers and tunneling can happen) and that 30V DC is the minimum voltage for sustaining an arc, no matter how small. Note that an arc may be struck by less than 300V by creating plasma via vaporizing metal by touching the electrodes together momentarily and then pulling them apart. Arc welders work this way and typically work with about 80V open circuit voltage for DC welding. I have also witnessed accidents where a plasma jet (which cut things quite effectively!) was created and sustained without high voltage by striking and then drawing-apart electrodes, as long as the available current was very high, especially during the striking event. One such accident I saw was caused by an improperly-spec'd fuse opening with a very large short circuit current going through it. This was delivered by some large lead acid batteries along with an array of electrolytic caps. Sean On Tue, Aug 9, 2016 at 6:18 PM, Sean Breheny wrote: > 50kV is high enough that corona discharge will be an issue. You will > likely want to pot the diode stack or submerge it in oil to prevent > excessive loading during the 50kV pulse. > > The kind of thing you are suggesting doing is common in starting xenon ar= c > lamps. I think they usually have a separate starting electrode or they > float one supply on top of the other (for example, putting an HV pulse > transformer in series with the high current DC source and then applying a= n > HV pulse to the primary of that transformer). > > > > On Tue, Aug 9, 2016 at 4:05 PM, David VanHorn wrote= : > >> Yes.. I am half enabled there. Email works but no web or file access. >> Yahoo >> is 'working on it'..... Yeah right. >> >> On Aug 9, 2016 9:57 AM, wrote: >> >> > > Arc resistance will depend on many things. I have seen 60ish ohms o= n >> a >> > gap >> > > that 50kv would jump in air. Your mileage will vary but that may wo= rk >> > for a >> > > rough estimate. >> > > >> > > I am looking for spice models that are moderately accurate. >> > >> > There have been a number of people on the LTSpice list modelling spark >> > gaps of various sorts. There have been references to papers that have >> been >> > found along the way from which models have been generated. >> > >> > >> > >> > -- >> > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive >> > View/change your membership options at >> > http://mailman.mit.edu/mailman/listinfo/piclist >> > >> -- >> http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive >> View/change your membership options at >> http://mailman.mit.edu/mailman/listinfo/piclist >> > > --=20 http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .