Mail*Link(r) SMTP FWD>RE>FW: DES:FAB:ASSY: Breakdown Welcome to the interesting world of high-voltage electrical/electronics --------------------- Norm started with: HV breakdown is a current topic here as well. In a basic double sided application that has a continuos 7KVAC with frequencies up to 30 MHz, is there any advantage to using the more expensive polyimide laminate in place of FR4? Is there any truth to the rumor that polyimide is more 'arc tolerant' because of it's higher Tg? Norm Dill [log in to unmask] ------------------- Then Lou Hart responded with: Date: 8/13/96 12:04 PM From: [log in to unmask] Response to Norm Dill's message...Years back I spent a week at the Biddle School in Plymouth Meeting, PA, supposedly learning how to detect partial discharges (arcs that do not bridge conductive electrodes) in insulators. Partial discharge (sometimes called, erroneously, "corona") is what is responsible for long term destruction of insulation, in many cases. Most resistant materials are inorganics, less resistant are natural organics (paper), and least resistant are synthetic organics (epoxy encapsulants, for example). Lou Hart Compunetics -------------------- And I'd like to add the following: Arc Resistance is the ability of an insulating material not to develop an conducting path (carbon or organic material) due to a plasma like arc between a pair of electrodes. Per the standard tests, the time is reported in seconds. For most epoxy's, the arc resistance is in the range of 120-180 seconds. To my knowledge, polyimides are similar. Partial Discharges -- generally, partial discharges are a "limited charge" local breakdown of a gas due to electrical stress. Partial discharges are generally a precursor to corona. Partial discharges are the electrical breakdown of a gas and follows Paschen's Law, and generally occur in a gas void in a solid/liquid dielectric. For convenience, a sketch should simplify the condition CCCCCCCCCC_____ dddddddddd __V1 ddd ddd ddd ddd V2 ddd ddd ____ dddddddddd __V3 CCCCCCCCCC V1 and V3 are the voltages across a dielectric "d" between a conductor and the "top" (V1) and the "bottom" (V3) of a void , which is the hole (blank space) in the middle of the "d's". V2 is the voltage across the void itself. Now what happens: With continuous dc (direct current) voltages, some very small current will flow due to the "seriesed" insulation resistances. The voltage drops for V1, 2 and 3 will be porportional to the current flow and their individual resistances. At some applied voltage (or as a function of time), the voltage across the "void" V2 will exceed the Paschen minimum for the gas contained in the void and an "arc" will occur between the top and bottom of the void, what happens is the "top" and "bottom" surfaces of the void form capacitor electrodes, when the partial discharge occurs, it is the discharge of the capacitance of the void (from one side of the void to the other) that occurs. Repeated partial discharges will degrade the dielectric and ultimately cause "treeing" in the dielectric, then a short (conductive filament growth) and ultimate failure. For most "small" voids in dielectrics, the partial discharges are generally less than 100 pC (pico Coulombs), more frequently about 10. Voids are worse with ac (alternating currents). With ac, we no longer have a simple resistive voltage divider. Instead we have three capacitors in series. Let's assume all three capacitors have the save cross-sectional area, and have capacitances of C1, 2, and 3 corresponding to V1, 2 and 3. Assume the dielectric thickness's are equal and the gas dielectric constant Dk = 1, and the solid Dk = 5. Then capacitors C1 and C3 are each 5X the capacitance of C2 just due to the Dk of the dielectrics -- all else being equal. The capacitive reactance (the ac electrical resistance of a capacitor) is inversely porportional to capacitance. (The foumula is Xc = 1 / ( 2 * pi * f * C), where f = frequency, and c = capacitance in Farads). What this means is that due to the "seriesing" of the the capacitors, there will be 5X as much voltage across C2 as there is across each of the other capacitors (C1 and C3) --- talk about a stress concentrator. Now Paschen's Law comes into effect -- and you've got zappooo, increased partial discharge breakdown. Yes, there are some effects due to variations of gas breakdown as a function of frequency. More details will be avoided at this time, but the above is the basis for the caution statement in the A-600 E on the acceptance of "measles" for "high voltage" applications, meaning at STP >240 Vac, rms or 340 Vdc or peak pulsed amplitude. Hope this is of interest and/or helps to understand the problems in "high-voltage" applications. Ralph Hersey [log in to unmask] ------------------ RFC822 Header Follows ------------------ Received: by quickmail.llnl.gov with SMTP;13 Aug 1996 12:03:29 -0700 Received: from ipc.org by simon.ipc.org via SMTP (940816.SGI.8.6.9/940406.SGI) id NAA05761; Tue, 13 Aug 1996 13:51:48 -0700 Resent-Date: Tue, 13 Aug 1996 13:51:48 -0700 Received: by ipc.org (Smail3.1.28.1 #2) id m0uqOH0-0000NOC; Tue, 13 Aug 96 13:33 CDT Resent-Sender: [log in to unmask] Old-Return-Path: <[log in to unmask]> Date: Tue, 13 Aug 1996 14:39:15 +0400 (EDT) From: <[log in to unmask]> Subject: Re: FW: DES:FAB:ASSY: Breakdown To: "Dill, Norm J" <[log in to unmask]> cc: "'IPC - Technet Input'" <[log in to unmask]> In-Reply-To: <[log in to unmask]> Message-ID: <[log in to unmask]> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Resent-Message-ID: <"xLHGv3.0.V9G.GfC4o"@ipc> Resent-From: [log in to unmask] X-Mailing-List: <[log in to unmask]> archive/latest/5729 X-Loop: [log in to unmask] Precedence: list Resent-Sender: [log in to unmask] *************************************************************************** * TechNet mail list is provided as a service by IPC using SmartList v3.05 * *************************************************************************** * To unsubscribe from this list at any time, send a message to: * * [log in to unmask] with <subject: unsubscribe> and no text. * ***************************************************************************