Re: [TCML] comments on designing a lightning protection system

[Jim Lux <jimlux@xxxxxxxxxxxxx>]

-----Original Message-----
>From: FIFTYGUY@xxxxxxx
>Sent: Jun 9, 2008 10:42 PM
>To: tesla@xxxxxxxxxx
>Subject: Re: [TCML] comments on designing a lightning protection system
>
> 
> 
>In a message dated 6/9/08 8:00:01 P.M. Eastern Daylight Time,  
>mike8675309@xxxxxxxxx writes:
>
>>>What's the *best* possible  conductor to use  to
>>>channel lightning? Heavywall large-diameter silver pipes   braided into a 
>Litz cable?
>>
>>the best possible conductor is one  sized to carry the load while able
>>to provide a low impedance path to  the ground point.

>
>That sounds like the most "adequate" or  "economical", not the "best".

"good enough".. 

The goal in lightning protection is to prevent the building from burning down, and, in some cases, to reduce transient overvoltage damage (although that's actually determined more by other things).  One doesn't need superconducting downleads to do it.


>    And I haven't seen in any of the links below  where "the load" is 
>defined. How can you engineer a lightning protection  system if you don't define the 
>parameters of lightning? It would seem that a  truly professional installation 
>would include statistical data about  historical lightning in the area, and 
>what level of lightning the system would  be designed to protect against (to 
>what degree). How do the contracts  read?

NFPA 780 determines what's needed for a lightning protection system, not by analysis, but by specification of what measures need to be implemented. The "code" is determined by professionals in the business on the basis of historical experience (which would include the typical statistics of lightning)


>
>>From a canadian source here is one  example:
>>http://www.iaei.org/subscriber/magazine/00_c/canadian_code.htm
>>
>>[quote]  In general, lightning protection systems grounding is done using
>>copper  or aluminum cables and ground rods, 1/2 inch copper or>
>>copper-clad steel, 5/8 inch galvanized steel or copper plates.  Rods
>>should be spaced to allow down conductors to be run down a building  to>
>>the ground electrode as directly as possible. A metallic  underground>
>>water system may be used for both electrical and lightning  protection>
>>grounding. The lightning protection grounding should always  be
>>interconnected with a metallic water system if available outside  the>
>>building.
>
>
>Hmmm, that "metallic water system" is usually  "galvanized steel pipe"... ;)


And, in the U.S., the National Electrical Code does not allow the use of water systems as a sole grounding electrode.  It DOES require that all grounding electrodes be bonded together.

>
>>Lots of info here on grounding done for a 911 center in  Florida.
>>http://www.copper.org/applications/electrical/pq/casestudy/florida911.html
>
>
>Well, the "copper.org" people would certainly  advocate copper conductors... 
>Mentions that hollow braided copper cable.  Wonder if that stuff would be 
>suitable for monster TC primaries. Maybe a bit  hard to tap!

I once talked to a custom wire company about getting copper braid over a fiber/plastic out  (think of coax without a center conductor).  It's certainly doable, and not that expensive.  OTOH, regular old 2/0 stranded works pretty well.


>
>>Interesting look at grounding from a HAM  perspective.
>>http://www.astrosurf.com/luxorion/qsl-lightning-protection.htm
>
>
>I'm glad I'm not a HAM - protecting those  lightning-magnets looks like a 
>real pain!
>    Copper ribbon seems to be popular.


There's lots of ham literature out there with things that are just plain wrong, or out of date, or are tribal lore.  Sometimes it comes from old handbooks, etc.  This is particularly the case when talking about transient suppression, etc.

Just because G. Marconi did it in 1910 doesn't mean that it's still good advice....


For a review of more recent literature: http://home.earthlink.net/~w6rmk/lightning.htm
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla