Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!zaphod.mps.ohio-state.edu!sdd.hp.com!decwrl!sgi!rpw3@rigden.wpd.sgi.com From: rpw3@rigden.wpd.sgi.com (Rob Warnock) Newsgroups: sci.electronics Subject: Re: PC Power ON/OFF Scheme Message-ID: <63914@sgi.sgi.com> Date: 12 Jul 90 08:13:41 GMT References: <334@uncmed.med.unc.edu> Sender: rpw3@rigden.wpd.sgi.com Reply-To: rpw3@sgi.com (Rob Warnock) Distribution: usa Organization: Silicon Graphics, Inc., Mountain View, CA Lines: 52 In article <334@uncmed.med.unc.edu> uchuck@uncmed.med.unc.edu () writes: +--------------- | ... a DPDT main relay with the PC and its peripherals on the moving | contacts and the power line on the normally open contacts, I had the idea | of connecting the normally closed contacts together and then to ground in | the hopes of preventing any lightning hits getting to the PC. | | Now, I begin to wonder... What is the effect of shorting the power | cord(s) of the PC and associated equipment together and to ground, even | under normal (read operator induced) conditions???? Can anyone think why | this might be a problem?? | Chuck Bennett +--------------- CAUTION! You could actually create quite a safety hazard! DPDT relays have been known (depending on the brand) to have fault modes (or even normal operation!) in which one pole switches before the other. It all depends on the spacing between the contacts, how much the contacts "stick" when opening (due to micro-welds due to arcing when closing), the stiffness of the piece of fiber/plastic which mechanically ties them together, etc., etc. Now if you simply short the normally-closed contacts, you're pretty safe against an out-of-order switching, since until both contacts go to "closed" the other end is open. But if you also ground this shunt, then an out-of-order switching can momentarily connect the "hot" wire to "ground" (NOT "neutral"!), thus creating a dangerous "ground fault", exactly the opposite of what a safe system would do. In fact, this is the very situation "ground-fault interrupters" are designed to prevent. [Why is it dangerous? Well, all "grounds" aren't the same, and if you happen to have one hand on a "hot" ground and another on a "cold" ground...] And even without grounding the normally-closed strap, there's another relay fault mode to worry about: transient bridging. Again, depending on the relay, it's load capacity, history, etc., it can be possible for the normally-open and normally-closed "stationary" contacts to be briefly connected together through the "movable" contacts (some audio relays are even deliberately des- igned this way, to avoid switching "clicks"), which in the wiring connection you propose would briefly (and then possibly permanently!) short the "hot" power line to "neutral". Heat, light, (hopefully) blown fuse/breaker, and (hopefully no) fire. For this reason, the UL (or CSA or VDE) would probably never approve the power connection you suggest above... ;-} ;-} [Hmmm... There just *may* be special-order (and exPENsive!) relays which could safely be used in this case, but a garden-variety DPDT mains relay isn't it.] -Rob ----- Rob Warnock, MS-9U/510 rpw3@sgi.com rpw3@pei.com Silicon Graphics, Inc. (415)335-1673 Protocol Engines, Inc. 2011 N. Shoreline Blvd. Mountain View, CA 94039-7311