Hi all,
I've gotten a lot of great responses to my inquiry about shielded/
ultrashielded/etc. magnets, and I've pasted them below. Several
consensus opinions emerged, which I'll describe here. However, I'd
like people with experience maintaining more recent generations of
extra-/ultra-shielded and 2.2K pumped magnets to help me append this
thread by telling me about their cryogen use and field drift.
Consensus:
A) Everyone loves shielded magnets. Not only do they offer more
flexible siting, but they present less of a safety hazard by reducing
the likelihood of attracting stray objects.
B) A fraction of colleagues reported problems on startup, but they
did not appear related tho shielding.
C) People are generally impressed with the good homogeneity of modern
magnets, as measured by the cryoshim currents required to correct
inhomogeneities.
D) It is difficult to separate the effects of extra shielding from
the concomitant improvements in other areas of magnet technology. For
instance, helium boiloff from an old unshielded 500 will be lower
than that from a new ultrashielded, pumped 800, but eliding the
nature of the boiloff requires broader comparisons.
So, yes, shielded is better than unshielded, but I think we still
haven't answered some critical questions. For instance, when deciding
whether to purchase a normally shielded 700 and an ultrashielded 700,
how much is that 6" smaller 5-gauss footprint worth? Will the floor
need to be reinforced? Will it need more helium? Will its operating
costs be greater or less?
I'll do a little quantitative research on magnet specs and post the
findings to NMRWiki. I welcome anyone to contribute further comments
on the systems they know best.
Thanks.
- Josh
RESPONSES:
I have not seen any extra troubles for ultra shielded magnets.
***
I've had terrific experience with Bruker shielded magnets at 400/500
MHz (not a single quench in about 30 magnet-years), but my experience
with Magnex magnets at the same field strength is much less steller.
***
In the earlier year ('03), we have some troubles with the shielded
magnet. One of our 600 MHz magnet was charged for many times to
bring up to the field and eventually quenched without any warning.
Later on (after '05), we have been installed two shielded magnets
without any troubles whatsoever. One is Ultrashielded 600 MHz and
another one is Ultrashielded Plus 400 MHz. Both magnets have been
survived for twice relocations because of the lab renovations. One
thing I have noted is that the University Safety regulations have
been becoming tighter and tighter in recently years and it is
strictly required that 5-gauss lines not penetrating to public space
and other laboratory. To make an arrangement of the magnet location
to meet these requirement is really difficult without having the
shielded magnets. In fact, it has been become our policy that no un-
shielded magnets will be allowed to put in our laboratory to save the
space and to meet the safety requirements. This is especially true
for accepting the donation of old un-shielded magnets.
***
We got an 800 MHz actively shielded magnet from Magnex in 2004 (this
was just before Varian acquried (is that the right word?) Magnex to
make their magnets). (We had to get a shielded magnet since the
fringe fields from an unshielded magnet would have extended into
space outside the lab that we couldn't safely control.) The field of
the magnet is incredibly homogeneous (with the RT probe, all 40 shims
where under +/- 5000 DAC units, and all but 5 were under +/- 2000,
causing us to joke that we could probably run without the RT shim
coils), and the shielding is pretty phenomenal as well (our field
engineer does this test with a couple stainless steel bolts where he
puts them head to head vertically under the magnet so that they stick
together (something to do with induced eddy currents? I'm more of a
chemist than a physicist.), holds the top one, then moves radically
away from the center of the magnet; where the bottom bolt drops (eddy
current no longer sufficient to keep it attracted to the top one?) is
approximately the 5 gauss line, and for this magnet it's about 2 feet
outside the vacuum can!). Drawbacks of this magnet are it's size
(it's 10 tons and 10-12 feet high (just the magnet; with the legs,
the top of the magnet is about 15 feet off the ground) with a 6 foot
diameter), and it's thirst (fills every 2.5 months, 750 Lt. liq.
He). And stability-wise, we have had issues with the main switch
opening, which causes the magnet to start to run down at 4 MHz/hr.
(exciting times!); this has happened 3 times since we got the magnet
(last time 2.5 yrs. ago, so we REALLY hope we're done with that
foolishness). Magnex has sent a field engineer each time w/i 36 hrs.
to arrest the run down and re-energize the magnet. In our
discussions with Magnex, no mention has been made of the problems
with the switch being related to the fact that the magnet is shielded
(there was either a problem with the design of the switch or the
materials from which it was made, and both of these have been
improved). I'm pretty sure Magnex's/Varian's present version of the
actively shielded 800 is substantially smaller and less thirsty.
Sorry, this reply may have more info than you really asked for, hope
there's something useful for you somewhere in there.
***
We have one 800 US2 magnet. It was installed five years ago and we
haven't
had any trouble with it (except for the occasional problem with a
pump or
a switch). We're very happy with it so far. Certainly there's been no
indication that the magnet is unreliable.
But don't get me started on cryoprobes.
***
I'll find out soon because I'm replacing my 100% reliable, trusty, 1980s
vintage Varian R2D2 300 magnet with a Bruker Ultra-Shield-300
magnet. Of
course, since these R2D2 magnets have exceeded their servicable
lifetime, I
really had no choice. I'll await your summary, and should be adding
my new
experience to the pool sometime early next summer.
***
I currently have two shielded and two unshielded magnets, and have
experience of one ultrashield in a previous job. I am not aware of
any problems with unshielded magnets. I would not consider an
unshielded magnet now, they take up much more space and are more
sensitive to their surroundings.
The only issue that a manager should be aware of is how the helium
level meter is set up. I believe a shielded magnet is bulkier and so
occupies more of the bottom of the helium can. So it is not safe to
let the helium level drop as far as on an unshielded magnet. So if
the meter is linear with depth, there isn't much volume of liquid
helium at the bottom end. I believe that to avoid this problem, some
magnets are set up so that the level meter only measures the depth of
liquid above the magnet.
***
I've have 3 Bruker ultra shielded magnets (300, 400 & 500) and had no
issues with them. Only once in the last 5 years did our 400MHz flip out
a bit and the shimming changed but was a cryoshim problem and that could
happen to any magnet. The 300 and 400 are at least 8 years old (we got
them used) I think, hardly drift at all.
The magnets only need to be filled with He 2-3x a year and can go for 2
weeks without a N2 fill so the maintenance is minimal.
***
XXXYYY is getting a shielded magnet 700/750 wb here in our chem dept
and it
is really heavy. that is the biggest problem with these magnets. it
will
weigh about 11 to 12000 pounds.
anyway, floor loads need to be considered in a very big way. also, most
elevators can't haul them up (actually they worry more about the
brake for
the way down). their height and foot print are really nice though if
the
floor can take the weight.
***
we have a Bruker Ultrashield II 800 magnet - pumped. Drift is 3 Hz
and we have been very happy with it. It is now 5 years old and we had
no problems apart from 1 quench during installation. Last year a
light rail started running 200 yrds away (600VDC) and the disturbance
is minor. Originally an unshielded magnet was ordered, but the
shielded magnet has allowed us to put 2 more magnets in the same space.
***
I have experienced three Varian 400 systems two with Oxfords and one
Magnex. The supercon radial shims need to be touched up about once a
month. No other issues at all.
Oops, Josh, these were all shielded. They all are about 3 years old.
***
We have 7.05 Tesla (300 MHz) non shielded magnet and
had problem in the past due to movement of Steel
cupboard swiftly on first floor !. As the movement of
cupboard was exactly above magnet centre line, the
field moved along with it. "The field moved along the
cupboard for some distance and when the cupboard went
out of range, the field tried to come back to original
field shape, in the process a transient magnetic field
was generated generating voltage in the coil, which
cancelled the original field, hence the magnet almost
went to half the field, loosing the lock instantly !"
As explained to us.
The magnet was discharged completely with back to back
diodes with heat sink and recharged. Non shield / un shield
magnet always have danger if heavy magnetic material moves
in the magnetic field region. Shielded magnet has very
little region out side the magnet dewar, which can
influence the magnet. 5 Gauss line is as near as one to two
feet out side circumference and about 2 to 3 feet above in
the centre. Where as our centre magnetic field line went
well above 10 feet causing the problem!
***
We have Bruker Shielded 300, 400 and 600 MHz magnets. We have not
had any issues with any of them. The 400 and 600 have been in
operation since about 2000. The 400 was de-energized and moved in
about 2004 with no issues. The 300 was purchased used in 2006 and
moved to this site and energized here with no problems. We also have
a very old Oxford 500 non-shielded magnet and have had no issues
with that system. I am happy with all of the magnets at this location.
***
We have an 800 US2 installed last fall. We are very happy with it.
Everything is within spec and we have had no problems yet. We simply
wish it used less He.......
Seriously, as a pumped magnet it uses a lot of He (~200 liters every
5 Weeks), although it is within spec. But that is because of the
higher field and not because it is shielded.
***
We have several shielded magnets and several non-shielded magnets
and I
wouldn't say there is any difference in reliability whatsoever. Every
magnet has its own personality so if you didn't have many you might
develop
a prejudice.
***
we had a Bruker 400 Ultrashield installed a few months ago a noticed
that N2 consumption was ~3 times above specs. Bruker sent us a
'blowing gauge' (I don't know the actual name of it) and confirmed
today that the blowing rate is ~3 times higher than normal. A magnet
specialist will be scheduled to try to find the problem. The
instrument has worked well otherwise.
***
We have a Bruker 800 MHz US2, Ultra stabilized, Ultra shielded magnet
since early 2006 and no problem with it.
We also got a Bruker 600MHz "Ultra shielded Plus" magnet installed in
April 2008. It has the nitrogen liquefier. No need to fill N2 every
week! It's great and we don't have any problem with it either.
Both are pretty stable. No complain.
***
We have two shielded Varian magnets, a 500 and 600 that were
installed during the summer of 2006. The 500 when it was installed
was quite unstable and more or less a pain in the rear. We ended up
yelling loud enough that Varian replaced the magnet after about 18
months. The 600 had what I would consider to be unjustifiable
sensitivity to changes in atmospheric pressue. We had to put a
manostat on the magnet for a couple of months. When it finally did
settle down, we haven't had a problem since. Were the problems
attributable to shielded magnets vs. the older unshielded design?
Hard to say exactly, but I've had 20+ different supercons over the
years and these were the only two that we've had that kind of
problems with.
***
unfortunately we have only a few of the newer shielded magnets, most
of our
spectrometers still have the old "traditional" magnets.
We DO have an "ultrashield" 500 MHz widebore from Bruker, vintage ca.
2000,
in our research lab, plus an "ultrashield" 500 MHz standard bore from
Bruker
(ca. 2003) in our routine NMR lab.
So far there were NONE AT ALL issues specific to shielded magnets,
except for
the advantage of being able to place them closer to other NMR
magnets, and
having much less to worry about magnetic attraction (for metal
toolboxes,
all sorts of dewars and gas cylinders, etc. pp.). In additon, NMR
experiments
also seem to be better "shielded" from adverse effects from the outside.
In the near future, we are planning to buy a few more MHz (another 500
and 800 MHz standard bore magnets). Needless to say that we are
specifying
shielded magnets - the ease of site planning and everyday handling is
definitively
worth the little extra price over traditional magnets (if still
available at all).
***
We have an Ultrashield(tm) 600MHz/89mm that hasn't had any
problems for almost three years. Its on solids 99+% of the
time so we may not notice small instabilities.
I do wish that they left a charging rod with the systems.
***
It will be useful to separate pumped from non-pumped systems when you
collate this stuff....
***
We have a 500 MHz Bruker ultrashielded magnet, installed in March,
2002. It has been extremely reliable and completely trouble free to
date. It has virtually no field drift and has had no increase in
helium consumption and has much better resolution than the specs,
although that seems to be normal for Bruker.
Needless to say, we are very happy with it!
***
I would say that another factor would be the space where the
new magnet would be put. In my case I had an unshielded 300
and we were putting in a 500 into a 30'x20' room, so a
unshielded 500 wouldn't have worked well.
We have had our Inova 500 since June 03 and all has been fine;
I have had to reset Z0 just recently so now I have a 499.75 instead
of the original 499.76 .... at least I have a 500 =)
The shielded type definitely go thru LHe faster.
***
AFAIK, to make a shielded magnet, they essentially start with a
magnet of higher field strength and the shielding reduces the field
to the target level. So I guess if you're getting a shielded magnet
near the cutting edge, it will only be as reliable as a magnet with
that higher, original field. (A shielded 900 might be an unshielded
950 for example.) I think a more useful measure of reliability is if
the magnet in question is a part that has been in production for many
years or if it is new and possibly "buggy" technology. Don't be an
early adopter!
We had an unshielded 300 (Bruker) for about 14 years that never gave
us any trouble.
Now we have a shielded 600 (Oxford) and ultrashielded 400 (Magnex/
Varian) that we have had for about 2 years and they too are fine.
The drift rate on the 600 is so low as to be difficult to measure.
***
My observations are purely anecdotal.
We have a mix of shielded and older unshielded magnets here.
The 11.7T unshielded magnet has by far the largest stray field of any
of the magnets we have. The console and shim stack for this magnet
were recently upgraded and the performance of this updated system is
remarkable. By contrast, the previous shim stack was quite old, and
people believed that the poor performance of the system was due to
the aged magnet. The upgrade proved that the magnet was just fine.
We have also had pretty good luck with our shielded magnets. The
"newest" magnets are vintage 2003; one is a 16.4T (700MHz) magnet and
the other is a 14.1T (600MHz). In both cases, the magnets have
performed very well.
My thoughts:
The benefits of improved active shielding is a mixed blessing. It
means that the space requirements for siting a system are greatly
relaxed because we no longer have to be so concerned about the rather
large stray magnetic field. However, we still have to deal with all
of the other site requirements such as clean and stable power,
cooling, ferrous building materials, ventilation, weight capacity, RF-
fields and ergonomics. In the past we had the "luxury" of using a
large space to site one system due to the stray field effects. Now
with the newest shielded magnets, the decreased stray field size
permits us to squeeze more and more systems into smaller and smaller
spaces. We are then prone to compromise on the other restrictions,
and thus we may end up sacrificing the ultimate system performance.
As for reliability: I think that the magnet technology is quite
mature. I have no doubt that any commercial magnet up to 16.4T
(shielded or not) will be completely reliable.
I have no direct experience with pumped magnets. The requirement for
2K superconductors introduces a whole different set of technical
hurdles. At this time, I'd rather not have the added headache of
dealing with such a system (of course, assuming that I had the funds
to even consider such a system).
If you are considering a new system, go for a shielded magnet. But
you MUST push as hard as you can for as much space as possible. And
make sure that all of the infrastructure needs are met. We are just
now renovating space for a new system, and I can't believe how tight
it is going to be. I had to fight tooth and nail just to get the
space I was ultimately granted. It will work, but I don't look
forward to helium fills.
***
Our latest acquisition, now 3 years old, has a Bruker 600 MHz
ultrashielded narrow (52 mm) bore magnet. A few months ago it
'dropped' the Z2 cryoshim requiring that shim be recharged. The
process was expensive since by then my University had implemented its
PhD + 'vendor' option despite my having been trained by Bruker at
their Swiss magnet facility (usual grumble, but more just puzzled).
I do not see any connection between the Z2 cryoshim fault and the
shielding but it may correlate with other replies that you get. If so
then it starts to get interesting.
Shielding is already a relatively mature technology. It certainly
makes such magnets easier to work with, especially in an 'open shop'
environment.
***
I currently have 7 magnets, (3 shielded and 4 un-shielded) ranging
from 200 to 700 MHz. I have not noticed any correlation between
shielding and reliability. However, I would point out that my
comments relate only to persistent 4.2K magnets. I don't know about
actively pumped magnets, but I would expect them to be inherently
less reliable (more moving parts = more bits to go wrong). The other
point I should make is that the shielded magnets are inevitably(?)
younger than the un-shielded ones, so maybe you need to ask me again
in say another ten years :-) In any case, I don't think there is a
choice here for many of us. There is simply no way I could have
fitted a 700 magnet into our existing facility if it wasn't shielded,
and the chances of me getting a custom-built facility to house a new
NMR machine are tiny, so it's shielded magnets or nothing.
***
Not the answer you have requested but possibly relevant.....all our
magnets are on the second floor and we find the new style magnet stand
on our ultrashield magnet does not damp vibrations as well as the old
style black air-filled 'donuts' that the old style magnet stands had.
Even adding the new active damping system did not perform better than
the donuts and so our new ultrashield magnet gives spectra with more T1
noise than an old Oxford magnet of early 1990s vintage and hefty stray
field.
***
My lab has 4 magnets of various builds (wide bore, narrow bore, 300,
400, non-shielded, shielded). Here are my opinions on shielded magnets.
- I don't see them as more prone to "falling out of bed" type quenches
- They are less likely to pick up a stray chair or trashcan from the
lab then others, which means they are less likely to see the violent
crashing interactions that that would cause
- I cant remember the exact stat's but I believe the drift rate is
higher on my newer shielded magnet then my non-shielded magnets, but
not alarmingly so.
- I believe the zone of homogeneity is smaller on our shielded
magnet vs our non-shielded magnets. We routinely use 16mm probes and
finding the sweet spot on our shielded magnet was substantially more
difficult then our other magnets. I don't believe that you would
notice the difference with a 10mm or smaller probe though.
*******
*******
ORIGINAL POST:
Hi all,
In some of my discussions with people regarding new magnet
technology, I've heard anecdotal reports of tradeoffs between
shielding strength and magnet reliability.
If you have a shielded/ultrashielded/ultrastabilized/premiumshielded/
compact/etc... magnet, could you please tell me a little about how
much or how little trouble it has caused you? How happy are you with
it/them? Do you think they're more prone to problems than older, less-
shielded magnets? I think it will be useful to collect your anecdotal
reports and see if there's really such a trend.
Thanks.
- Josh
Josh Kurutz, Ph.D.
Technical Director, Biomolecular NMR Facility
University of Chicago
Gordon Center for Integrative Science, room W123C
929 E. 57th St.
Chicago, IL 60637
Office: (773) 834-9805
Spectrometer Room: (773) 702-4052
Cell: (773) 315-5732
Fax: (208) 978-2599
nmr.bsd.uchicago.edu
homepage.mac.com/jkurutz
www.joshkurutz.com
Received on Thu Feb 26 2009 - 10:29:46 MST