[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Preventing a REALLY expensive mistake.

Original poster: "Jim Lux by way of Terry Fritz <twftesla-at-uswest-dot-net>" <jimlux-at-earthlink-dot-net>

Tesla list wrote:
> Original poster: "Christopher Boden by way of Terry Fritz
<twftesla-at-uswest-dot-net>" <chrisboden-at-hotmail-dot-com>
> >Original poster: "Jim Lux by way of Terry Fritz <twftesla-at-uswest-dot-net>"
> ><jimlux-at-earthlink-dot-net>
> >
> > >
> > > 1. The building is fed with 1-P 240VAC, and will be upgraded to 3-P
> >480VAC
> > > (for the elevator of doom), should I keep the 1P service as well? or
> >just
> > > have the landlord upgrade from one to the other?
> >
> >Presumably, when they bring in the 480, they provide local transforming
> >down to standard 240/120?  Otherwise you'll have to do it yourself.
> >
> >If it doesn't cost anything, I'd leave the existing service in.
> Cool.
> >
> > >
> > > 2. Will a system of 10 seperate 8-10' ground rods driven vertically
> >around
> > > the buildings exterior and all laced together for a massive ground
> >system be
> > > an good enough protection for the computers and other electronics? Is
> >there
> > > a better way?
> >Not necessarily...
> >The key is to keep your electronics grounds separate from the power grounds
> >(hence the prevalence of those orange Isolated Ground receptacles) until
> >you get back to the common panel ground.
> Ok, at the common panel ground I can bring them TOGETHER? THIS is a part
> that had me wondering because even if I run them to completely seperate
> grounds outside, won't they connect to each other THROUGH the ground? But
> since this is all happening IN the Earth wouldn't the voltage be so low as
> to not be a concern?

Sure, the resistance is low (you hope), but the physical distance means
there is a significant inductance, and for any sort of transient (i.e. a
spark discharge), the inductance is what is going to determine the voltage.

Example: Say you had a 10 meter piece of 1/2" copper pipe connecting the
bottom of your coil and where the spark is going to hit.  You have
(foolishly) used the far end (i.e. the place the spark hits) as the ground
for your computer.  The resistance is going to be a few milliohms, and even
if your spark is 1000Amps pk current, you think that it's only going to be
a few volts... no problem.

But wait, that's not what really happens.  That 10 meters has an inductance
of around 10 microhenries.  The rise time of that tesla discharge is, say 1
microsecond (actually, it's a lot faster), corresponding, roughly to 3 MHz.
 At 3 MHz, the impedance of 10 uH is 188.4 ohms, so that 1000 Amp discharge
is going to make a voltage of 188 kV!!!  As you say, WHOA!

Such is the subtlety of trying to shield and isolate high current fast

Moral of the story: Think carefully about all the current paths!

> Ok...-dot-coming from an audio background I know that a Ground-Loop is one of 2
> things.
> 1. Some sort of connection that will result in injecting a nasty 60hz hum
> into your mix that is a pain to get rid of. This is why we buy things like
> rubber washers for rack gear, direct-boxes and a million other goodies with
> Ground Lift switches...etc. But noone ever explained to me EXACTLY just what
> a Ground Loop actually IS, lol. Help?

Any loop of a conductor that intercepts a changing magnetic field will have
a current induced in it.  If there is any resistance in the loop (as there
always is) there will also be a voltage across those resistances.  If one
of the resistances happens to be the input impedance of an amplifier,
you'll be amplifying that induced voltage.

> 2 The act of flipping (pitch axis) a plane, on (or very close to) the
> ground. This happens when you catch the landing gear on a small power line
> while landing, etc..

actually, around the yaw axis for a ground loop. In a tail dragger, it is
most common when you get the CG of the aircraft outside the landing gear.
> >
> >Run the sensitive loads off an isolation transformer.
> Where do I get one big enough? Can I just run the HVL on an Iso-Transformer?
Iso transformers are expensive. minimize the load running on the iso.