Many suggested I work with the supplier to plan the facility. I am and
will continue to do so.
Many pointed out my neglecting to mention ceiling height. I will make
sure it is at least 12 feet.
Many pointed out the benefits of actively shielded magnets. Charles Fry
cautioned that they may be more sensitive to moving metal objects within
their 5 G zone. I will recommend we get actively shielded magnets and
will be cautious about their 5 G zone.
Many cautioned that vibrations could be a problem especially on the third
floor. Roger Kautz made the point that motors on the roof around the
magnet(s) can be a problem. I'll look into the roof top equipment and we
will definitely be getting vibration dampening legs.
Several mentioned having good temperature stability for the NMR room. I
guess this will require special measures such as a separate temperature
control unit. I'm not sure what a Liebert is, but I'll look into it.
I particularly appreciated the detailed comments of Jessica Dion and Roger
Kautz.
Thanks again to all who took the time to reply.
Nathan Dubree
Genentech Inc.
**************************************************
> We are currently in the planning stages for a new building which is
> scheduled to contain up to two NMR instruments (300 and up to 600 MHz),
> probably Varian. What sort of requirements and concerns can you
> recommend for the room construction. Particularly with regards to the
> structure (dimensions and materials) and services (power, lights, HVAC,
> compressed air, etc.). Of particular interest would be NMR lab setup
> horror stories, so that these situations could be avoided.
>
> Schematically we've planned a 21' X 27' room on the 3rd floor of a large
> building. It is bound on two sides by lab space (people, small
> instruments, and chemicals), another side by offices, and the final
> side is an exterior wall with windows. Above it will be the roof and
> below it will be more working environment (labs and/or offices). This
> location cannot be easily changed. Because of the location we will use
> shielding and are considering active shielding. Is active shielding
> worth the extra cost? Does it hamper the performance of the instrument?
>
> I read through the archives and found some interesting discussions about
> fire sprinklers and florescent lights, but none that addressed my
> concerns directly.
>
> I will summarize and post responses if I get enough to be of interest.
>
> Nathan Dubree
> Genentech Inc.
**************************************************
You don't mention the ceiling heights. The only way to have people
on a floor below a 600 with 12' ceiling heights, is to use active
shielding. Otherwise, you must block off the area where the stray
field extends. That's my opinion anyway; the stray fields on a
600 are too large to allow uncontrolled movement directly below
the magnet. We are currently in the purchasing stage to get an
actively shielded 600; I certainly expect to see no performance issue
with respect to a normal magnet. Only thing I've heard about is that
actively shielded magnets might be more sensitive to movements of
metal objects (gas cylinders, cryo-tanks) at the 5 G line. That
wouldn't appear to be a problem with the size of the room you have,
but be careful you don't place the 600 magnet such that the 5 Gauss
line is extending outside the room into a hallway.
----------------------------------------------------------
Charles G. Fry, Ph.D. Tel: (608)262-3182
Director, MR Facility Fax: (608)262-0381
Chem. Dept., Univ. Wisconsin
Madison, WI 53706 USA email: fry@chem.wisc.edu
----------------------------------------------------------
**************************************************
The big issues are ceiling height in your room, large doors for
installations, and good room temperature stability. Most of the rest are
fairly obvious like not being too close to large pieces of moving metal
like elevators.
Jane Strouse
Dr. Jane Strouse
Dept. of Chemistry and Biochemistry
UCLA
Los Angeles, CA 90095-1569
(310)-825-9841 - voice
(310)-825-0393 - FAX
strousej@chem.ucla.edu
**************************************************
We are currently installing 400 and 500 systems in a similar
size area. A factor you have not taken into account is ceiling height. You
will probably need > 3.2 metres to install a 600. In my opinion actively
shielded magnets are excellent news for NMR. Our new 400 magnet (from
Bruker) has been installed in a space previously occupied by an AC250. We
plan to do LC-NMR on our new shielded 500 and we can put computers and LC
equipment much closer to the magnet than before.
Also check your compressed air. We recently had rusty water in ours and
this
got into probe, magnet and sample changer. Check your dewpoint with a
Draeger tube and consider putting a drier in line.
Good luck with your plans
Kind regards
Ian
Ian Whitcombe Registered address
Team Leader/Physical Methods Dept. Roche Products Limited
Roche Discovery Welwyn 40 Broadwater Road
* Phone : (+44) 01707 366546 Welwyn Garden City
* Fax No : (+44) 01707 366907 Hertfordshire AL7 3AY
* e-mail : ian.whitcombe@roche.com Registration Number 100674
**************************************************
I encourage you to pose these same questions to your Varian salesperson as
well. The sales and
installation staff will work closely with customers to evaluate sight
plans, and can be very helpful
in avoiding problems. Power, HVAC and compressed air requirements will
all vary depending on
whether you have immediate or future needs for solids, imaging, and/or
diffusion accessories. Sales
and installation staffs of the various vendors are very knowledgeable
about these requirements.
Your salesperson can also give advice on your room dimensions and
materials. Lastly, your Varian
salesperson has some interesting information about shielded-magnet
technology. Regarding shielding,
I highly recommend that you seek out the opinions of those who own
shielded magnets, especially
those close to train and truck ways.
Best regards,
Cindy Ridenour
Varian Applications
**************************************************
An actively-shielded 600 can be placed in the same space as an unshielded
400. You have greatly reduced ceiling height requirements and reduced
requirements for protection of personnel from stray fields. For example,
you can likely put two 600's in the space you described with active
shielding. I would certainly use active-shielding first, particularly if
you can eliminate the need for some other shielding of the room. I know
of very few cases where passive room shielding has been completely
successful.
Dr. Ronald A. Nieman, Director NMR Facility
Department of Chemistry & Biochemistry
Box 871604, Arizona State University
Tempe, AZ 85287-1604
**************************************************
Bruker provides a site package planning guide to answer these questions
for
you - I'm sure Varian does the same.
bob sato
=================
Robert K. Sato, Ph.D.
Research Fellow
Catalytica Inc.
430 Ferguson Drive
Mountain View, CA 94043-5272
**************************************************
This may seem like a trivial part of your overall installation, but I
would suggest a few things. First, choose the color and material of your
floor wisely. It's important to be able to see small bits of things on
the floor, such as slivers of metal, paper clips and staples, and screws,
and other dangerous objects that may wander toward the magnet on the
floor. You basically want a color that would be everyone's nightmare for
home - something where everything shows up well. This way it's easiest to
keep the facility clean, and minimize the probability of debris getting
into the bore or sticking to the sides of the magnet.
You may want to consider carpeting, because small metallic objects can't
"walk" or be scooted along as well. On the other hand, you do have to
vacuum it. We have a maroon/red carpeting, and although I don't like the
color (it is heterogeneous and hides metal well), it does stop staples
dead in their tracks. We have an utterly el-cheapo vacuum cleaner that
weighs only a few pounds and is almost all plastic.. we use this for
periodic vacuuming of the carpet. Perhaps another flooring material would
be better, but I just thought I'd bring up the point.
Another thing to consider is painting one wall a dark color, like a dark
blue or green. I spend a lot of time looking at nitrogen and helium
plumes, and it would be a lot easier to see them if the back ground was
dark, not "institution white" cinder block. The track lighting helps to
provide some contrast, but I ended up buying some poster board and taping
it to the wall so that I could see the plume from a distance.. it looks
silly. Painting the wall would look much better. Incidentally, track
lighting uses a lot of electricity and the lights are expensive. We left
them on all the time in the beginning, but then got sick of calling
maintenance all the time for replacement bulbs.. so we left them off after
that. The spectrometer is in its own room, with a door between it and the
Sun computer, so we don't have to listen to the hum of fans and A/C while
we're working. It also means that we can keep the lights off 99% of the
time.
Depending on your usage, you may want the lights immediately surrounding
the magnet to be on a separate switchable circuit so they could be turned
off when you're not over near the magnet. It's kind of a dramatic effect
as well. :-)
If any of the people who install the magnet have a way to map the gauss
lines, you might consider askign them to do this. We never had this done,
so we just "guessed" where the 5 and 10 gauss lines are, and posted our
signed on convenient doorways. I have always been curious about the field
strength at different distances, and if I could have, I would have asked
someone at the time (if they had appropriate equipment) to mark off a few
lines for field strength. As is, we just created an artificial line with
yellow and black tape, to discourage people from crossing that arbitrary
threshold. I'd like to see a little plastic chain fence or something
similar to keep the riff-raff out, personally. Just a thought.
One last thing - often there is a one-time budget for this sort of
installation. If you can, you may want to consider having a platform
built instead of using a ladder. Filling from a ladder is more dangerous
and physically demanding.. it's sometimes hard to juggle the extremely
cold transfer line, the cover for the transfer line access, and the helium
exhaust clamp without falling off of the ladder. I've seen other
facilities where there is a simple wooden platform along side of the
magnet with steps going up, and personally, I think this would be far more
practical than a ladder (although I'd have the ladder on hand for
non-routine access).
You'll probably get plenty of other stories and suggestions from others,
but these are some things that were overlooked during our installation, so
I thought I'd point them out. Hope this helps!
- Jessica Dion
Biochemistry Dept.
University of Vermont
**************************************************
Just a few quick thoughts on lessons learned the hard way.
1. Make sure all doorways and elevators have sufficient clearance to move
the 600 magnet from the loading dock to its final resting place.
2. Make sure you have sufficient ceiling ht for helium service, and
magnet
energization.
3. Don't even think about installing a 600 magnet on the third floor
without the best vibration isolators available from the manufacturer.
Those
who have not been through this particular nightmare cannot imagine what a
sensitive seismometer a high fileld NMR magnet can be. Normal building
vibrations often make it nearly impossible to measure lineshape. You will
probably need vibration isolators for the 300 as well, unless you don't
plan
any 2D work.
4. NMR magnets have a high center of gravity. In seismically active
areas
(I assume you are in the Bay area) they should be suspended from the
ceiling
so they will not tip over during an earthquake.
5. The fringe field of a 600 magnet that is not shielded will cause
problems with computer monitors on the floor below. The area directly
below
will probably be in the 5 guass exclusion zone for heart pacemakers, and
will need a warning sign. Although the real health hazards are miniscule,
these signs make some people very nervous.
6. Locate the magnets as far as possible from any air handling units on
the
roof above. You need to worry about both vibration and magnetic field
modulation from large motors.
__________________________________
David J. Wilbur, Ph.D.
Instrumentation Specialist
Department of Chemistry
Tufts University
voice: 617-627-2163
Fax: 617-627-3443
email: dwilbu01@emerald.tufts.edu
__________________________________
**************************************************
Have the company (Varian or Bruker) come in and evaluate the site: they
can
give good advice on what matters most. On the 3rd floor, you're probably
looking at antivibration legs. Beware motors on the roof that only turn
on
sometimes - have a physical plant guy force them on during the site
pre-inspection. Even though the site is "fixed", you (and management) may
get more creative if you get concrete examples of what high vibration or
fields fluctuating by more than 1 Hz will do to your multi-million-dollar
installation.
The active-shielded magnets are a godsend: they are also less sensitive
to
environmental perturbances, as well as throwing less fringe-field so can
be
installed in smaller space. I understand it's pretty hard to shield
magnetic fields, and you should be critical about how much field-bending
is
going to have to be shimmed out if it isn't cylindrically symmetrical.
One thing I'm happy I did was have a Liebert installed, which is a
computer-room type air conditioner that also can control humidity. This
may be less of an issue if Genentech temperature is stable through nights
and weekends (don't ask - measure), and if you don't deal with muggy New
England summers. Having room temperature stable to +- 0.5 C means we can
leave VT off, which (1) avoids dangerous user mistakes, and (2) eliminates
temperature equilibration when you put samples in.
One thing I regret is not getting a quieter scroll-type air compressor.
We
saved $2000 by getting a piston-type, and even through it's down the hall
out of earshot, I still wince thinking about the poor woman on the
adjacent
toilet when it goes off like WW III. Quiet is worth more than you realize
now. Put the compressor as far away as you can. Be sure that any
condensation in the pipes has a way to drain, ideally back into the
reservoir, and get an auto-bleeding reservoir.
Definitely get an air dryer. The kind that switches between silica
towers and runs for ten years. They also make noise - if it has to be in
the room, build a cabinet around it. We've seen imbalanced towers show
up
in the shims: it's good to have a ballast tank after the dryer to give
some
mixing if the pipes aren't very long.
I plumbed in a valve that lets me switch air supply to a liquid nitrogen
gas-pack by the door of the room when we use VT. In practice I'd have
been
happy trailing a hose across across the floor on these occasions, but
management was happy to pay a couple hundred dollars to avoid that
liability.
I added a couple of guages into the air lines before and after the dryer.
We could have skimped on the $5000 UPS. There's a theoretical advantage
that it isolates spikes better, but in practice I think we'd be OK with a
surge suppressor. (Blackouts are once-a-year.)
Counter space next to the computer for an open notebook, an open manual,
and a couple of spectra is a necessity, not a luxury. A PC networked into
your laboratory information system is a worthwhile luxury.
Don't ignore the amenities: If stuff gets piled around, the room feels
like a closet, users get irreverent and equipment doesn't last as long. A
rack on the wall for the transfer lines and wrench. A coatrack. Shelf
space for manuals. A second table for the wash bottle and Kimwipes,
orphan
samples, and extra d-solvents (printers can go on top of the console;
solvents make it look old, fast). Cabinet space for the magnet power
stick, spare parts kit, standard samples, spare cables, hoses, tools,
spare
paper and printer supplies. Sacred cabinet space for your other probes.
Gas cylinders and lN2 near the door but out of the way. A niche for the
non-magnetic plastic Sears shop-vac, and mop. The mandatory safety fence.
A whiteboard and bulletin board for schedules, how-to's, notices, and
explanations.
Management will want a flow probe within a couple of years (96 samples in
3
hours, unattended) and it will want a cart or table near the magnet.
That's the hindsight here.
-- Roger
"Roger A. Kautz" <rkautz@lynx.dac.neu.edu>
**************************************************
I wonder if I can help you with your quest for information about nmr
facilities.
I understand you anticipate putting the systems on a third floor. I
strongly recommend that you have a vibration analysis done to be sure
there
are no vibration problems. This would not effect the 300 very much but if
you anticipate a 500MHz or 600MHz system then it would be very important
to
know the natural frequencies of the building. We typically put vibration
isolation posts on the 500MHz and 600MHz systems as a matter of course;
since these spectrometers are usually used for long term multidimensional
acquisitions.
As far as active shielding is concerned we have been delivering this type
of magnet for about three or four years and have literally hundreds at
field right now. We offer the active shielding for 300MHz to 700MHz
systems.
The Avance(tm) 300 has a magnet which has the 5G line contained within the
dewar itself ( no, it's not a fat dewar). This of course is a space saver
as well as a safety advantage.
When considering the 500MHz or 600MHz systems the fringe field is of
course
much more important; and the three dimensional profile must be considered.
Use of active shielding reduces the ellipsoid volume contained by the 5G
contour by 85%.
For a standard bore (54mm) 11.7T magnet the vertical distance from magnet
centre of the 5G line is less than 6 feet and radially about 4 feet. For
the 14T magnet ( 600MHz) the axial 5Gline is about 8 feet above the centre
of the field and radially less than 6 feet.
We have found no compromise in the performance of the magnet after the
redesign to include active shielding. The dewar size is the same as the
nonshielded magnet and this does reduce the helium hold time by a small
amount.
As far as problems with fluorescent light is concerned this was certainly
a
problem at Scripps with the 600MHz, non shielded systems and the
Avance(tm)
750. The Avance(tm) 800 room does not use this type of lighting. With sh
ielded magnets I would not anticipate a problem since the fringe field is
much less at the ceiling. A good rule of thumb is to think of a shielded
magnet as having the same fringe field as a non shielded magnet 200MHz
lower in field.
You are perhaps aware that we have just installed an Avance(tm) 800 in
Genentech building 29 for Wayne Fairbrother, Nick and Melissa. We also
just
upgraded one of the two Avance(tm) 500 systems to an Avance(tm) 600. This
Avance(tm) 600 has a shielded magnet.
Since you are in South San Francisco ( I assume ) there would be no
problem
in having a Bruker person drop by to look at the proposed facility, talk
about possible configurations etc.
Please let me know if I can be of further assistance.
Dr. Malcolm Bramwell
West Coast Office
Bruker Instruments Inc
Phone 510 683 4301
Fax 510 490 6586
mrb@bruker.com
**************************************************