I'm forwarding this for Roger Bourne, as it seems I somehow managed to lose
his original posting.
-----Original Message-----
From: Roger Bourne [mailto:rbourne_at_imrr.usyd.edu.au]
Sent: Thursday, July 27, 2006 12:19 AM
To: Richard.Shoemaker_at_Colorado.edu
Subject: ???AMMRL: SUMMARY Magnet Dissection (message lost????)
Dear Colleagues,
Many thanks to all who responded to my request for advice on dissection
of a decommissioned magnet for educational display. Judging from the
enthusiasm of the responses I suspect that the interest goes beyond
education and includes a certain amount of vengeance. When I get started
I’ll buy a few slabs of beer and invite you all to Sydney for a magnet
cutup party. Everyone can have a turn with the cutoff saw so there’ll be
no need for pushing and shoving.
Some examples of undressed magnets can be found at these URLs:
VXR
http://nmr.chem.umn.edu/cutaway.html.
JEOL Delta-GSX 270MHz
http://www.jeolusa.com/nmr/mag_view/magnet_destruction.html
Varian R2D2 300MHz
http://www.chem.purdue.edu/nmrfac/Facility%20Spectrometers/anatomy%20of%20a%
20magnet.htm
Varian 7.05 Tesla
http://www.chem.fsu.edu/facilities/cutaway_nmr_magnet.asp
Thanks especially to Karl Vermillion (NMR Lab, NCAUR, USDA) who
proffered the following detailed description of his experience.
… As far as tips or tricks, there really weren't any - just a lot of old
fashioned grunt work. We looked into several "easy" methods of cutting
the magnet open: acetylene torch (too messy for a display item), water
jet (not good with voids), electrical discharge machining or EDM (no
cutters wide/tall enough for our magnet), laser (specular reflections).
In the end we fell back on the time honored tradition of "the hardest
way is the best way" and used several dozen 6" cut-off blades with an
air-powered grinder.
The outer can took by far the longest. I believe it took the machinist
about 10 hours to get through that one can (3/16" stainless steel). Most
all the other cans were 1/4" aluminum and cut like butter in comparison.
They might have taken an hour each. Initially there were a million
little razor sharp burrs left from the cuts but they cleaned up nicely
with a file/edging tool.
We snipped the mylar insulation to within a couple of inches of the can
edge and then folded it back into the vacuum area until we were done
cutting. Then it was trimmed up neatly with scissors.
The only advice I can give is that you should make sure to start with a
large initial opening--near to a 180 degree opening. As you peel away
each successive layer the size of the opening and the angle it subtends
will decrease. By the time we got to the shim form (#14 in my display)
deep in the display, the grinder was too big to fit into the space even
with a small disk. Eventually I cut the opening for the shim form using
a Dremel tool (a tiny hand held grinder) and a 1" disk.
My magnet was missing it's bore tube (54mm ID), and a piece of thin
walled 2.25" copper pipe from a plumbing supply shop made a wonderful
substitute (#13 in my display).
I made a dust cover for the display from scratch-resistant polycarbonate
sheeting. It is possible to heat form that material. We clamped the
sheet to a table with about a foot or so overhanging and heated the edge
with a torch. When it softened, we bent the sheet in a 90 degree angle.
With a carefully placed 2nd bend there was a "C" shaped piece that fit
around the magnet, resting on the bottom flange between the bolts and
the side of the outer can. A piece glued to the top gave it stability.
It works well for the dust and it keeps curious fingers out of the display.
I had a hard time finding nice looking numbers to put in the display for
labels. Eventually I found the ones I used in a woodworking shop. They
were sold for people making their own clocks.
All in all we (the machine shop and I) spent a year and a half trying to
find clever easy ways to cut the magnet in half, and in the end we just
did it brute force in a week. It was sort of an adventure as I had no
idea what we were going to find as we removed each layer. I'm sure that
all magnets are different enough that there will be a few wrinkles as
you cut into yours.
The following are Karl’s notes for each of the numbered labels on the
display magnet:
1) Outer can – The outside of this 3/16” thick stainless steel can is at
room temperature (~300K).
2) “Super” insulation – Approximately 150 layers of aluminized Mylar
reduce the heat radiated from the outer can to the cooler interior. This
area is under high vacuum (<10-6 torr) to reduce heat transfer by
conduction and convection.
3) Outer nitrogen can – Made from aluminum, the nitrogen chamber outer
can is cooled to 77K.
4) Liquid nitrogen chamber – ~50 liters of liquid nitrogen fill this
space, keeping it a balmy 77K. Should this chamber ever run out of
liquid nitrogen the helium boil-off rate would dramatically increase,
and the pressures in areas #2 and #7 would be permanently raised. Both
liquid nitrogen and liquid helium boil-off rates would stay high even
after this chamber was refilled with liquid nitrogen. The boil-off would
not return to the previous value until the magnet was disassembled and
reinstalled. The liquid nitrogen is refilled weekly.
5) Liquid nitrogen fill port – The rubber hose above is hooked to an
external liquid nitrogen Dewar, and liquid nitrogen pours into the
nitrogen chamber via this opening. Water vapor from the atmosphere can
form ice inside the fill ports. A blocked fill port is a very bad thing.
Three blocked fill ports turn the magnet into a bomb.
6) Inner nitrogen can – The inner can of the nitrogen chamber is made of
aluminum.
7) High vacuum area – This area has a pressure of <10-6 torr to reduce
heat transfer by convection and conduction.
8) Radiation shield – Intercepts radiated heat from the liquid nitrogen
can. The shield has an approximate temperature of 20K.
9) Liquid helium can – Up to 30 liters of liquid helium fill this
chamber and keep the temperature at 4.7K. The liquid helium level should
never fall below 30 percent or there will be danger of a quench during
the fill. The liquid helium is refilled approximately every 2 months.
10) Superconducting solenoid (main coil) – Many miles of a coaxial wire
are wound in this solenoid. The inner superconductive core of the wire
is made of a niobium-titanium alloy, while the outer jacket is made of
copper. Copper acts as an insulator when the Nb/Ti wire is
superconductive. In the event of a quench (Nb/Ti wire not
superconducting) the copper has a much lower resistance than the Nb/Ti
alloy. In this case the copper would act as the preferred path to
transfer the energy to the quench resistors (#19) where it will be
harmlessly dissipated (with luck).
11) Superconducting solenoid (end coils) – The magnetic field at the
ends of a solenoid are half the strength of the center of that solenoid.
To help make the magnetic field more uniform a “doughnut” of
superconductive wire is added to each end of the main coil, boosting the
field in that region. This effect helps give a more homogeneous field at
the center of the main coil (sample region). Only one of the end coils
is visible here.
12) Superconducting shims – There are superconducting shims in the
magnet that are adjusted at installation to provide the most homogeneous
magnetic field possible. There may be as few as 3 (Z1, X, Y) or as many
as 9 superconducting shims (Z1, Z2, Z3, X, Y, XY, XZ, YZ, X2-Y2) in any
magnet. The proton linewidth of a H2O/D2O sample is typically 100 Hz
using only the superconducting shims. Linewidth of the same sample after
adjustment of the room-temperature shims (installed later, as they are
not part of the magnet proper) may be less than 1 Hz. There is a Z0 coil
that is part of the supercon shim set but is not a “shim” in the proper
sense of the word. It is only used to make fine adjustments to the
magnetic field strength during installation.
13) Bore tube – This copper tube is all that separates the outside world
(and your sample) from the high vacuum areas inside the magnet (areas #2
and #7).
14) Shim form – This aluminum form provides a solid support for the
superconducting shims and it acts as a baffle to prevent sloshing of the
liquid helium during fills. There is a narrow slot near the bottom of
the shim form where liquid helium may enter.
15) Liquid helium splash guard – Prevents spray during the liquid helium
fills from hitting the superconducting shims.
16) Magnet form – A heavy aluminum form around which the superconducting
wire is wrapped.
17) Helium filling funnel – When filling the magnet with liquid helium,
the transfer wand should extend downward through the left stack to
within 1” of the top of the funnel.
18) Liquid helium sensor – Made of superconductive wire, it has a
resistance of 3.2 ohm/cm at temperatures above 4.7K. Any part immersed
in liquid helium has zero resistance. By knowing the total length of the
sensor and the measured resistance, you can determine the height of
liquid helium still in the can.
19) Quench resistors – These absorb the energy of the magnet if part of
the (normally) superconductive coils were to become non-superconductive
(known as a quench). Damage may occur to the magnet if it quenches, and
it is much better to power a magnet down if the vacuum has degraded or
cryogens are not available.
20) Charging plug – At installation, this plug is connected to power
supply leads inserted through the top of the right stack. The power
supply is used to power up the field and adjust the superconductive
shims. This plug is out of it’s normal position because it is broken.
Ice can form on top of the plug which prevents the pins from the power
supply leads from being inserted. Apparently someone applied too much
force trying to force the pins through the ice and the plug broke,
thereby rendering this magnet into very expensive junk.
(text from Karl Vermillion, NMR Lab, NCAUR, USDA)
/ /
--
__________________________________________________
Dr Roger M Bourne
Computer Systems Officer
Faculty of Medicine
University of Sydney
Royal North Shore Hospital
St Leonards
Australia
Tel: (02) 9926 6195
Fax: (02) 9926 5220
Mob: 0414 597 589
__________________________________________________
Received on Fri Jul 28 2006 - 14:59:49 MST