Richard, INhomogenous fields? Now that you mention it, I've always assumed homogenous fields for this.
Theoretically, electric fields between two points vary inversely to the square of the distance between the points. Between two line charges (infinitely long), the field varies inversely to the distance. And the field between two plane charges (infinitely large) varies to the strength of the dielectric. I don't really know what two identical geometries Paschen used, but I've always assumed pointed. Any other combination of surfaces will have to be empirically studied. Paschens's law that I posted below only works to a point <bad pun> and what exactly those boundries are, I don't know. But, for instance, assume a p = 0, and see what you get. My opinion is that this equation was empirically derived with specific apparatus. Change that apparatus and you basically invalidate the given equation below. So the fact that you're using your own lab results as a datum to cross-checking the equation is a good thing. Somewhere I remember someone quoted as saying "When theory and experience clash, go with experience ..." Regards, Doug Richard Steele wrote: > > Doug, > > Great, at last some formulas, I take it that "flat opposable surfaces" > are refered to as Inhomogeneous fields? > Is the formula applicable to point to plane surfaces as well, i.e PCB > via or component pad to track rather than track to track? > What effect does the length of the opposable surfaces have, with surface > mount components the pads are usually oblong? > > Do you have a formula for "point to point electrodes" (homogeneous > fields)? Our testing results have given between 3 to 10thou for 1kV @ > 50Hz. Our next set of tests will be for a 10/700 pulse. > > Regards > > Richard Steele > Fujitsu Telecommunications Europe limited > > Douglas Mckean wrote: > > > > Richard, > > > > I'd be more than willing to tell you what I know. > > As such, it will be fairly brief ... <grin> > > > > Gaps vary greatly as you probably know according to the geometry > > of the two opposable surfaces under test. It's all a function of > > Paschen's Law for the breakdown of *** uniform *** gaps. > > Flat opposable surfaces normal to each other will give differing > > values than say two pointed opposable geometries. Further, a > > point opposed to a line will give yet another set of values. > > > > Paschen's Law can be stated as > > > > V(kV)= 24.2Sh +6.1(Sh)^0.5 > > > > where: > > > > V is the breakdown voltage in KV > > S=(293p)/760T > > h is electrode spacing in cm. > > p is pressure mm of mercury > > T is temperature in degrees Kelvin > > > > Now, I use from physics the fact that 3MV breaks down 1 meter of air > > (STP). > > > > Using the above equation at STP, we have 30KV/cm, which is a pretty > > good crosscheck. And thus, 3KV @ 1mm or approx 40 mils. Put in a > > factor of 2 for safety and you have 3KV for 2 mm or 80 mils. Boards > > are made out of FR4 which has a Dk of approx 4. Since this shows the > > strength of the material, we divide the 2 mm by 4 and we get 0.5 mmm > > for inner layer seperation. Not bad for the 0.4 mm seperation > > specified. > > I specified 0.5 mm for the designers. > > > > That's my 2.5 cents. > > > > Regards, Doug
