[TowerTalk] Pipe with water

[Jim Lux]

At 08:51 PM 7/12/2006, Red wrote:
>Hi, Gang;
>
>I scribbled a few calculations concerning the temperature rise if a
>copper pipe filled with water conducts a lightning strike.
>
>I assumed a 0.5" OD pipe with 0.050" wall thickness and I assumed a
>30,000 A current for a duration of 1 millisecond.  That makes a worse
>than average strike.

<snip>
looks basically sound to me.  This is actually not a lot different from the 
water resistors used in HV pulse power applications.  I've built various 
sizes as things like charging resistors for a marx generator, as well as 
the load dump.

http://home.earthlink.net/~jimlux/hv/rwater.htm has some similar 
calculations to yours.

I would suspect that electromagnetic forces might have a bigger effect. The 
fast pulse at high current can put pretty high forces on the pipe.  It's a 
real design issue for bus bars in electrical substations for instance: you 
don't want things breaking during a fault.  Heck, even the regular 60 Hz 
current at several thousand amps puts a pretty impressive force on the bus 
bars.

F = 2E-7 * I1*I2/r

Put, say, 10 kA on the two buss bars, and have them 50 cm apart, and you're 
looking at 1000 N (about 200 pounds).  Put a 100 kA transient from a fault 
into the system, and you're looking at 100kN.  (and therein lies the secret 
of shrinking a quarter to smaller than dime size...)

Jim