On Sun, 15 Aug 2004 11:39:30 GMT, Dave Shrader
wrote:
Richard, I'm not going to try to out calculate you.
But, please tell the group what the junction temperature of any
semiconductor device is at transient thermal failure at 0.1 and 1.0
useconds.
That was covered in another posting.
Twenty years ago the USAF test data indicated that failure occurred at
0.5 microjoules!!! That's 300,000 times more sensitive than your numbers.
Current USAF lore holds, rightly, that it is the power supply that
fulfills the failure, junction penetration energy is insufficient to
accomplish this. Turn off the equipment, disconnect leads and NOTHING
HAPPENS! Same pulse = no damage. Damage is not found in the
miniscule power it is found in poor practices.
Shielding is rather simple to accomplish, and taking care of front
door and back door paths is not Rocket Surgery. It takes very little
imagination to withstand these pulses; unfortunately it takes even
less imagination to succumb. The stories of 1 idiot's plight make the
news, not the 1 million survivors' success.
Note, it is extremely difficult to really achieve a 0.001 ohm mechanical
interconnect.
For you perhaps, and certainly others who experience lightning's
catastrophic failure, but 1mOhm is no big deal - measuring it is
however. It is simple to enumerate poor examples, but that does not
pass as careful accomplishment being impossible or unlikely.
Secondly, the major electronic failure mode is from the coupling of the
magnetic field. A di/dt of 10^5/[1E-6] yields a value of 10^11
amperes/second. A one foot length of wire has a self inductance of
approximately one nanohenry [1*10^[-9]] and the di/dt impact is ... 100
volts peak from one end of the wire to the next.
You are pencil whipping yourself, Dave. You are off by at least 1
order of magnitude in inductance.
As current USAF lore to this matter reveals (and it is nigh on to 25
years stale), huge pulse coupled currents at low frequencies do a
****-poor job plain and simple.
Saying coupling and making it happen are two different things (per
current USAF lore) and trying to fit 1MHz into a 1 foot wire is loath
to considerably less potential. This is called overplaying your hand,
a direct strike is enough to argue.
Back in the olden days, twenty years ago, the action integral, that you
calculated was sufficient to burn a 0.5 inch diameter hole right through
titanium that was 0.1 inch thick. Note: the titanium alloy we used
melts at slightly below 2000 degree F.
I've already commented to this specifically.
The design issue is TRANSIENT THERMAL EFFECTS not average heating. At 1
microsecond the heat flow from the stressed area has not started.
I've already commented to this specifically.
The
USAF required adiabatic heating as the peak temperature for the starting
condition for the transient thermal analysis. Restated, all the energy
is converted to instantaneous heat and then the thermal stress analysis
would be performed under that constraint.
Conclusion, lightning is a highly stressful environment.
Stress also encompasses primitive fear and probably far more so than
the actuality of failure. This is called the probability of big
numbers. In a nation of 280 Million, any individual's once in a
million experience like lightning striking them or near them happens
280 times a year (5 times a week or more) - it misses the other
279,999,720 times tho'. Thus such hair raising stories inordinately
color the topic and are suitable for selling insurance and amulets.
Hence the advertising and $$$$$$ suggested retail cost for "peace of
mind." At that price, if only 0.01% of those 279,999,720 pushed their
credit cards across the counter, that makes for a very profitable
living second only to ENRON.
Moral: "Don't use insulated towers."
Risk: "How well can you reduce R?"
73's
Richard Clark, KB7QHC
|