----- Original Message ----- From: "Nate Duehr" To: Sent: Tuesday, September 10, 2002 2:11 AM Subject: Re: [EE]:surge suppressing (MOV) > Hey Jim, > > A little OT, but... kinda right for this thread... > > On our radio gear around here we use a three-phased approach to > lightning protection from the power line... and surprisingly most of the > surges we've seen are not from the RF side or the tower, but from the AC > mains... That can be that way for a variety of reasons - a) induced lightning from a distance into power line, b) lightning nearby "depresses" (or elevates) the ground potential and the 'line', of course, is still at whatever'potential' (common mode-wise) WRT to "ground" from where is was sourced (called "Ground Potential Rise"). c) and line-transients on the lines themselves from ???? Depending on line 'faults' - there could be all kinds of strange transients/currents in neutrals. Once again, when the line is off your property - you have little control over it and what it may carry! > > Heavy-duty AC line filter (isolation 1:1 transformer type). > Right BEFORE that (outside) we put gas tube type protectors. > Right after that we put MOV's. Gas tubes - like TRL tubes (transmit-receive limiters) in RADAR transmitters - usually followed by limiter diodes preceeding the mixer 'crystal' (really a diode!)! > > If the "proof's in the pudding" it seems that the gas tubes can handle > the huge spikes but take a while to react, while the MOV's mop up Yup. Gas must first ionize (gas particles must collide, start some free electron flow from what I remember), then the resulting plasma conducts *nicely*. Plasma is like a wire - but formed of ionized gas (e.g. lightning is a plasma). Most shops seem to find what combo of grounding, supressors and filters works out best - eventually! > anything that makes it through the transformer pretty well. > > Lightning burns up a lot of MOV's in this configuration but rarely have > we seen any of the expensive equipment behind the MOV's dead due to > surges. Better the MOV's than gear! I used MOVs prior to an AC line filter in one application of a UHF repeater at a water tower - and of course, I used two separate fuses in fuse holders for both line and neutral (just to be safe). > > Of course, the usual precautions for having a single-point grounding > system, polyphasers on the RF stuff, and everything on one panel are > also done... as well as having the tower properly bonded to its grounds > (usually cadweld) and all that part done "right". > > Any other suggestions? Cell sites use a couple of "internal ground loops" in their "shelters" - one running near ceiling and down near ground - these then run via large guage cable to wide copper straps that surround the building (ground ring) buried in the ground outside and 'cad-welded' to ground stakes pounded into the ground every so often. They seem to take lightning hits (when I was with the AT&T WS carrier here we had 200+ cell sites and a paging sytem that notified ops as well as engineering when sites went out of service for any length of time), and aside from the T-span sometimes needing to re-synchronize for a minute or so they seem to do okay. There should be some good informative material on the web about this - "grounding kits for cellular base-station shelters" or some such nomenclature ... A PRACTICAL GUIDE TO GOOD EARTHING http://www.eso.qld.gov.au/publicat/men/menappe.pdf A quick little read on grounding gear expressing the need to properly bond everything using cadwelding (versus simple clamped connections): http://www.broadcast.net/pipermail/broadcast/2001-July/023934.html I came across this doing some research on this topic: Evaluation of the Electromagnetic Effects due to Direct Lightning to Nuclear Explosive Areas at Pantex Final Report http://infoserve.sandia.gov/cgi-bin/techlib/access-control.pl/1993/932517.pd f > > This seems to work pretty well out here on the mountaintops in the Wild > West. (Colorado) I understand ground conductivity can be a problem out there in the mountains. I know from the AM radio station antenna engineering studying I have done that the ground conductivity is in the range of 1 to 5 mS (milli-Seimens)/meter (mS = milli-mhos) in mountain areas versus 5000 mS/meter for sea water or 30 mS/meter for our good soil. How do you guys do "grounds in rock" ? Have you guys got any experience with tower "top-hats" that employ the principle of dissipating static charges so the electric potential never reaches the flashover point? Some of them look like frizzy things with a bunch of sharp points (creates a high potential at a point thereby ionizing the air mass locally). RF Jim > > Nate > > On Mon, 2002-09-09 at 19:37, Jim wrote: > > "their sucess depends on the nature of the noise you > > are trying to supress." > > > > Make that: "their success depends on the nature of the *spike* > > you are trying to 'arrest' or clamp." > > > > - and I'll agree with you. > > > > I have seen these installed on 3-phase 480 VAC lines coming into > > a radio station where they were there to attempt to 'clamp' > > lightning stikes to the AC service into the building - they > > had been hit some number of times before and appeared to > > have gotten hot enough to cause the solder (?) on their > > surfaces to flow downward due to gravity ... > > > > RF Jim > > > > > > > > ----- Original Message ----- > > From: "Richard Prosser" > > To: > > Sent: Monday, September 09, 2002 8:25 PM > > Subject: Re: [EE]:surge suppressing (MOV) > > > > > > > My experiance is that their sucess depends on the nature of the noise you > > > are trying to supress. They are graet for knocking the top of low energy > > > spikes etc. but must always have suitable fuse protection somewhere. (due > > > to the wearout mechanism & their tendancy to fail short circuit) > > > And they will not help if the mains swells above their conduction > > voltage - > > > they just blow the fuse (if you are lucky!) > > > > > > To be really effective in reducing spikes you may need to add a few metres > > > of cable as a series inductance before the MOV . - and keep the earth > > > leads/bonds as short as possible. > > > > > > Richard P > > > > > > > > > > > > >At 03:44 PM 9/9/02 -0600, you wrote: > > > >>Anyone had good luck...ie....not much burnout in MOV's for a 50A 120VAC > > > >>circuit? Or is MOV not the way to go...something better/cheaper/etc ? > > > > > > > >50A? Where are the transients coming from? > > > > > > > >MOVs are inexpensive and okay for very occasional transients that are > > > >within their joule ratings, but they do have an inherent wear out > > > >mechanism. > > > > > > Also they sometimes fail in shorted mode, with > > > attendant smoke and flames. If you have a 50A > > > circuit, these would be BIG flames. > > > > > > Reg > > > > > > > -- > > http://www.piclist.com hint: PICList Posts must start with ONE topic: > > [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads > > > > > > -- > http://www.piclist.com hint: The list server can filter out subtopics > (like ads or off topics) for you. See http://www.piclist.com/#topics > > -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.