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

Re: Avalon



Hi Chris,

More comments interspersed.

Original Poster: "Christopher Boden" <chrisboden-at-hotmail-dot-com>
>Reactor design? Alright...now's where one of you
>Mathemagictians get to do your thing.

Greg Leyh is a great fan of 3 phase DC coils. I believe he is
one of the leading experts on TC charging reactor design. He
also has great knowledge on big coils (as does Bill Wysock
and the rest of the high power gang), esp. when it comes
to DC charging problems (diodes, etc). You may wish to
contact him. He should be on this List. Greg, would you
care to comment ? However, remember, that all the big
guys do this for a living and most of them are probably
kept quite busy by their "ordinary" customers (not only
designing and building the equipment, but constantly
being on the move from place to place).

>>achieve. Go for 3 phase 660V setup.

>Can we GET 660V in the US? 3 phase is easy, and
>120-240-277(high freq mostly for industrial lighting)-and
>480 are easy...but 660 is new to me.

Hmm, I think so. U.S. power experts out there, what do
you say? As far as I remember there is even another
stage out there. Somewhere in the 950V+ (I donīt
remember the exact voltage) range.

>>it well. The G-forces on a the big (necessary) disk are almost
>>unimaginably high. Take material growth (due to heat) into
>>account.

>I am thinging balanced and turned aluminum disks, 12" dia
>and 2"thick at the center tapering to 3/4" at the edges.

A 12" disk wonīt cut it at this power level. Remember, your
electrodes must fly by very fast at this power level. A 250kVA
arc can be streeeeeeeetched quite a bit. Have a look at Bill
Wysockīs incredible RSG or Gregīs RSG (on each website
respectively). These are already very massive units and they
are running quite a bit less power than you are planning. Bill
even went to the trouble to introduce nitrogen into the (sealed)
gap. Also read up on their failure reports. These guys KNOW
what they are doing, and even though they have lots of
experience and knowledged people, they did have failure
problems, which cut back on time and took another slice out
of the money cake. I am not sure if tapering will be of help. It
should reduce windage losses, but at the same time you will
be weakening the disk area with the most stress (where the
electrodes are). Also Al grows quite a bit, when heated up.

The G-forces go up rapidly as speed and diameter increase.
Your 100-150 gram heavy tungsten electrode and holder are
going to be excerting forces of MANY kgīs on the disk. This
would call for an extremely well guarded RSG "compartment".
A normal brick wall will do almost nothing to stop such a
grenade. You will need bullet proof material, followed by a
heavy (several inches) steel enclosure. The space between
both should be packed with something which turns speed into
heat (something similar to the stuff used in the vests that bomb
squads use). Maybe even a liquid. However, that would need
expansion space as you canīt compress liquids and of course
it would require a liquid tight housing.

>Drilled and tapped for 1/2" electrodes with dual set screws
>for each electrode. Possibly turned concentric grooves in the
>disks for cooling.

Although not having done any mathematical work, my gut feel
tells me you will need at least 1-1.5" diameter, possibly even
up to 2" diameter tungsten electrodes for this kind of power,
which also demand constant attention and a periodic
replacement.

>The HV systems will be housed in a seperate room, a secure
>environment with incredible ventillation. It will have keyed access,
>and PIR motion sensors, the door will not lock to the inside, and
>the coil will shutdown if there is movement in the room.

"I" would discourage the use of electronic sensors such as motion
sensors. Why? A sensor needs to be reliable and withstand all
sorts of torture (be it mechanical, electrical or electromagnetic).
An electronic sensor has electronics, which can fail. I would rely
on electromechanical means. For example, you canīt open the
door w/o switching off the main power relays (make them
redundant) and you have to manually re-apply power once the
door is closed AND locked. Also, at this power level, the wiring
tends to jump around quite a bit due to the surge current you
will be seeing, which might let the motion sensor switch off the
power, even when there is no danger, Of course, this is better
than not having it shut off, when someone is inside. At my dadīs
plasma lab, I remember the crew having to reroute the 1/2 MW
480V 3 phase input power wiring because it had been routed
incorrectly and the movement of the wire caused the walls and
paint to crack and crumble away.

A few members on this List have stated the problem of structural
design, which is one aspect, that I hadnīt even given thought to.
Of course, this should be addressed first. As I said, designing and
building the grounding plane is going to be one of the larger and
more tedious jobs. Also, remember, if companies donate you
something that usually costs $$$ to buy, they will want something
back in return. These means you will paying them back for quite
some time, be it in terms of publicity, site usage, or even doing the
actual experimental work for them, etc. Although this shouldnīt be
a problem, it should be addressed. Of course, the facility will also
require a bit of money to keep it in shape, once it has been
completed. It might be a good idea to roughly calculate, what all
this is going to cost, then figure on doubling that number JUST to
see what it will cost you in the end. I donīt wish to put you down.
I just want to address some of the thoughts which pass through
my mind ;o))

Of course, I wish you lots of luck with your project ;o)) and I will
try to be of assistance, where I can.

Coiler greets from Germany,
Reinhard