-=> MIKE ROSS wrote to WAYNE CHIRNSIDE <=-

 MR> I remember using a 6 amp scr on 30,000 uF at 40 volts and the scr never
 MR> even got warm, as it should be BTW. Remember an scr is supposed to
 MR> behave as a dead short or switch once triggered.

A surge can easily destroy a semiconductor junction without the
package ever getting warm. Hell just pick up some static off
your T.V. screen and pick up some CMOS and see what happens to it.
Hell I was swapping hard drives about 6 years ago and fried a Seagate
when my elbow inadvertently touched the monitor screen...
the hard drive never even got warm but nor did it ever work
again either.

 MR> How long does it take to discharge 80,000 uF at 50 volts into .1 ohms?
 MR> Let's see... using the approximation t=CV/I, gives about 8
 MR> milliseconds.

 MR> Now translating that into power dissipation at say .7 volts across the
 MR> scr that's just shy of 3 watts averaged out over 1 second. Of course
 MR> the approximation assumes a constant current, so in reality it would be
 MR> an exponential decay and thus even less power dissipation.

 WC> Anyway I'm not a fan of stressing solid state components by
 WC> continually exceeding their ratings, leads to reliablity issues.

 MR> You should appreciate that semiconductor ratings are usually meant for
 MR> steady continuous duty at a conservative level. However,
"time" is a
 MR> factor that is not often given in the specs and "time" is very
 MR> important when dealing with very brief pulses. Semiconductors can and
 MR> do withstand really huge current surges if the timing is kept very
 MR> brief.

 I'm really rather well aware of this having pulsed many 
 a LED into the ampere ranges for very short pulses when I
 was working with light transmission communications devices.
 Something like 99.999 to .001 duty cycles.
 You can even easily destroy a power mosfet designed to handle
 many amperes this way.
 I've read some rather obscure literature on the damage
 caused to bipolar transistors and SCR's by short spikes which
 while not immediately evident do severely impact the MTBF
 of such devices.

 WC> I don't "believe" I *know*, both from literature and from personal
 WC> experience. You still have to get those milliamperes to do the
 WC> damage and that takes low skin resistance, higher voltage or both.
 WC> I've checked my typical dry skin resistence and it's in the 50K
 WC> range and at that you're not even going to feel 50 volts.

 MR> I don't doubt you believe you know... well, in all evidence to the
 MR> contrary you should since you're still alive here to reply about it!

Yes and I've been bit by upwards of 20KV off a color T.V. and lesser
voltages many a time.
Onetime installing a new higher amperage sevice to a mobile home
my partner installed the safety ground to the previously floating
lines I was connecting up top right off the street transformer
despite my specifically telling him to not do anything until
I was done doing so.  Unlimited current.
I was sweating so skin resistence was very low, I was standing on an 
insulated mat and using two sockets one in each hand to fasten
the clamps when when the moron ignoring me attached the ground and 
my arm came into contact with that 2 gauge ground wire :-(
I carried the imprint of that 2 gauge stranded wire burned into my
arm for a couple of months and had a few choice words for the
idiot.

 WC> BTW the reason that AC is so much more dangerous is there's a
 WC> relaxation cycle the heart passes through which is especially
 WC> subject to electrically induced fibrillation and the standard
 WC> 60 CPS AC current standard assures that that WILL occure in a very few
 WC> cycles of AC current.

 WC> I rewound an external winding on a salvaged T.V. flyback and drove it
 WC> with a  very stiff 12 volt source, 12 amps draw, through a push-pull
 WC> oscillator and got a wicked belt just from the low turns primary when
 WC> disconecting the  source...
 WC> now the secondary threw  a wicked spark and would ignite
 WC> cardboard in about 1 second.
 WC> Now that puppy was really deadly.

 MR> One should always worry when a table top experiment starts to exhibit
 MR> the phenomena of St.Elmo's fire. 

Well yes and I had intended the device for a few home physics 
experiments however never got round to enclosing it in the 
plexiglass case I'd intended for it so knowing my nature I
threw the beast out. That project only took an hour to slap together,
8 turns centrer tapped on the flyback driven by stiff push-pull
oscillator with the 2 drive transistors mounted on a large heat sink.
That puppy hit me worse disconnecting the power supply from the
primary worse than the color T.V. did off the secondary because
while the voltage was significantly lower it was still quite
high enough to push much more current though me than the 
other H.V. incident.
 

--- MultiMail/MS-DOS v0.42