At 08:05 AM 1/4/2007, Gary Schafer wrote:
>
> > > A common
> > >direct stike is 100 kA with a risetime of 8 us, correct?
> >
> > A more common waveform would be rise time of 2 microseconds, fall to
> > 50% in 50 microseconds. Typical average stroke is 20-30kA. 100kA
> > would be a big-un
>
>The industry standard for protection device comparisons is a waveform of
>8x20 microseconds. You want to see how much energy is let through with a
>given size strike with that wave form. If they don't tell you the let thru
>energy, look to someone else.
That's the waveform for something like a phone line (where the phone
line has flattened things out a bit), and you'd have no where near
the kA kinds of levels.
One should be very careful about specifications on transient
suppression. Most equipment can tolerate a pretty hefty spike,
voltage wise, with no damage, but, as Gary points out, it's the
energy that counts.
There isn't any "standard" for the actual suppression by the
way. There's just a spec for the test waveforms. (for instance UL
1449(?) just says if you test, you have to use a certain waveform and
quote your results in those terms).
And, in some cases, the waveforms may not reflect typical experience
in a ham station. For instance, a transient waveform for above
ground telephone lines that stretch for many tens of km (and would be
appropriate for designing the transient suppressor where your phone
line enters the house) would not necessarily be appropriate for the
transients you might see induced on your coax from a tri band Yagi on
a 65 ft tower.
Likewise the 2/50 waveform is typical of a surge induced near a
lightning stroke, but a more common surge might a switching transient
on the power line. These have lower peak voltage (say 4-5 times
Vrms) but MUCH longer durations (milliseconds). In my area
(relatively little lightning) the switching surges are a bigger issue
than lightning (although, the switching surge might be caused by
lightning.. lightning hits, trips a distribution breaker, load goes
away, big rise in voltage propagates back, breaker automatically
recloses, voltage comes back, load is now present, etc., another
transient propagates away...)
But this would be *power line* transients and not *antenna
feedline*. I think (off the top of my head) that feedline type
transients are almost always going to be the fast rise time sort of
thing. That would be either carried on the outside of the coax, or as
a differential mode signal as if the antenna had received the transient.
There's also a pretty big difference in the mechanics of the various
kinds of transient suppression (i.e. MOVs are very different from a
Vacuum spark gap, for instance). I think that in a lot of cases,
what you probably really want is more of a low pass filter than a
transient absorber. Standler recommends something like a fast
gas/vacuum switch at a few hundred volts as a clipper at the service
entrance surrounded by low pass filtering. An inductor on the line
side to limit the peak current through the switch so the di/dt
doesn't induce trouble elsewhere, and a LC on the output side to
squash the let through peak (which will have sharp edges) to a lower
voltage longer duration swell, which most electronics won't have any
trouble with.
Standler also believes that "point of use" suppression, particularly
with MOVs, might cause more trouble than it's worth, especially with
consumer grade gear. (fires, silent failures, etc.). His point is
that you suppress the big transient at the boundary (service
entrance) and then all you have to worry about is something induced
within your house, and, almost by definition, that's fairly small
(unless you have a hobby building Marx banks or Tesla coils...I have
several friends who have killed their Garage door openers (more than
once) that way)
Same thing for lightning impulses on the antenna. Suppress at the
entrance to the house, tie all the stuff inside together so it goes
up and down together, and let everything outside do whatever it wants.
Jim, W6RMK
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|