AMMRL: Faraday specifications summafication
Dave Scott (scott@iastate.edu)
Tue, 18 Mar 2008 16:07:13 -0500 (CDT)
The short answer is that Faraday cages commonly contain
MRIs but don't often accommodate a(n) NMR. Spectroscopists
tend to change the field to avoid the birdie blues.
The l o n g answer is not approved for patients with ADHD.
Political, economic, and health concerns bring contrasting views
to the discussion. You can support nearly any position.
Most relevent to much of the group are last years site planning guides:
Varian: 2.10 Radio-Frequency Environment
The site should be checked for radiofrequency interference (rfi)
at or near the operating frequencies of the spectrometer, listed
in Table 11, and most common nuclei, listed in Table 12).
The level of any interference should be attenuated to an
electrical field strength of less than 150 µV/ m at the
site of the magnet.
Radio-Frequency Interference
Most radio-frequency interference (rfi) comes from transmissions
for radio, television, paging systems, and cellular telephones.
Some emission comes from electrical and electronic equipment in
the immediate area. The horizontal NMR imaging system has a,
shielded magnet bore and is not sensitive to average rfi levels.
If the room level exceeds 10,000 µV/m, additional shielding may be required.
Bruker: 11.4 RF Interference
as a rule of thumb only broadcasting transmitters located within
a radius of apporximately 5km would normally be a possible source
of interference. As a general guideline a site should be checked
at the respective frequencies shown in table 11.2 ( 12MHz to ~=8%
above the 1H frequency). The level of any RFI should be
attenuated to an electrical field strength of -64 dBm at the
side of the magnet.
Even the specifications are difficult to compare. µV/m implies an
antenna which introduces a frequency dependence. At 42 MHz, a 150µV/m
signal will input -64dBm to the receiver. At higher frequencies the
induced signal is smaller.
Where do these spurious signals arise?
Radio frequencies vary from 1Hz to 100 gigahertz.
In the US, these frequencies are licensed as follows:
The AM band runs from 0.535MHz to 1.605MHz
3 to 30MHz is used by short-wave radio. (the power ranges from 1 to 500 kw)
25.6 to 26.1 is DRM
cordless devices (43.69 to 50MHz)
VHF television (54-72 and 76-88MHz)
FM radio (88-108MHz) => usually around 10 mW/cm2
To find how much rf AM and FM is feeding your spectrometer, go to
http://www.antennaweb.org/aw/welcome.aspx => Choose an antenna =>
enter your zip code (address not necessary)
aviation (118-137MHz)
VHF ham radio (144-148MHz)
VHF television (174-216MHz)
To find the strength of TV signals at your spectrometer, go to
http://tvfool.com/index.php?option=com_wrapper&Itemid=29 =>
enter your zip code (address probably not necessary)
UHF ham radio (420-450MHz)
UHF television (470-608 and 614-806MHz) => 10 mW/cm2
cell phones (824-849 and 869-894 MHz) => 0.45mW/cm^2 = 4.5W/m^2
industrial-scientific-medical 902-928MHz
global positioning 1227 and 1575 MHz -160dBw
PCS-personal communication services (1850-1990MHz)
wireless router 802.11a/g (2.4GHz, 5GHz, 3.7GHz) => 13-20dBm (20-100mW)
Its not the wild west, but these are valuable frequencies,
so useage can change with time. The FCC is now auctioning frequencies at
1.4GHz, 800MHz, 700MHz, 454MHz, 220MHz, 152-159MHz, 35MHz
The NMR signal we're trying to protect is on the order of millivolts.
The wavelength varies from 0.3M for 1000 MHz to 30M for 10 MHz.
If you use the 1/10 wavelength rule of thumb to block RF,
that would mean no penetrations larger than 3cm. Most cages
limit penetrations to 0.5cm.
Shielded rooms are difficult to build. Not only do you need
to block through space RF, but any wires that penetrate
the shield can act as antennas. All wires into the room
need to be grounded. The cost is substantial, between $50-110K.
There is no certification process although both Lindgren
and Imedco have expertise. At the end of the process,
your facility could contract "ugly room" syndrome.
If you build it, your increasingly wireless
users could go into withdrawl in the NMR lab :).
Because both the computer and entertainment industries
believe more rf is better, the problem isn't likely to
go away. The courts won't help because the
Telecommunications Act of 1996, Section 704 states that,
"No State or local government or instrumentality thereof may
regulate the placement, construction, and modification of personal
wireless service facilities on the basis of the environmental
effects of radio frequency emissions,,,"
Since the console is already a faraday cage and all
wires to the magnet are shielded and grounded, a
spectrometer is almost its own faraday cage.
I imagine that if the problem grows new spectrometers
will either be increasingly "RF hardened" or there
will be a Faraday option. Some imaging magnets
are RF shielded, but patients are more claustrophobic
with extended tubes and end caps.
If you want to "future proof" your lab renovation,
but can't justify large expenditures, brass screen
costs around $3/sq ft. If you overlapped and sandwiched
it between two layers of drywall, the seams wouldn't be
too difficult to seal.
It would be good to begin monitoring rf power vs frequency
in our facilities. Is there a standard way to measure this?
-
Thanks to Andy, Clemens, Jerry, Kelly, Marty, Michael, Paul, Razvan,
Sara, and Simon for the following insights:
-
Your best bet is to discuss this with a Faraday cage vendor, such a Lindgren
or Imedco. Since the design of a cage must take into consideration the shape
of the room, the openings (doors, ventilation, lighting, sprinklers) and the
frequencies of the instruments, it really requires expertise for the results
to be acceptable. We generally assume you'll need 40dB attenuation for MRI,
so I'd aim for the upper end of the range you specified.
-
I will look forward to what you find out. We are shortly to move our
500 out of reach of the local HDTV station - direct line of sight. If
we are to operate in the planned to building we will need a shielded
room. My understanding is that they are not cheap and not easy to build.
-
They built one to house the world's second unshielded 600WB NMR magnet
at the University of Warwick Physics Dept.
-
30-40 dB sounds too small.
MRI rooms are RF shielded for 100 dB attenuation
in the pertinent (resonant) frequency ranges.
-
There should be no need for a Faraday cage. The magnet can, probe body, and
console case are Faraday cages if properly grounded. The common practice
I've seen using faraday cages is to keep interference in, rather than out.
I have never heard of cellular phones interfering with NMR.
I have seen many birdies go away after troubleshooting led to the discovery
of a bad filter or cable. If the birdie is truly external, 30dB is a
significant reduction. Whether it is enough is impossible to say unless you
already know the magnitude of interference.
-
this is my personal opinion. A faraday cage might keep the birdies away
but there is no guarantee. I think it will only work if this is built as
a Rf shielded room. This means that all conductors going in and out of
this room or cage are separated or filtered.
Our colleagues from --- build such rooms. Usually they are ugly and very
expensive.
-
Do you have two instruments
with the same nominal field strength?? If so
their fields need to differ by about 100 kHz
otherwise you will have problems.
400 Mhz instruments are prone to interference
from FM radio on 13C spectra - common
frequency ca. 100 MHz. Watch out for campus
FM station(s) which can put in a signal strong
enough to cause problems.
-
Try Varian site and look up the Installation Planning Guide.
There is a section on faraday cages.
-
I know that from time to time imaging suites have
magnet rooms that have rf shielding.
I do not know of any chem NMR lab with rf shielding.
Nor am I aware of any initiatives within -----
to build spectrometers (i.e. magnets & probes) with improved rf shielding.
In lieu of shielding, the most common technique
to minimize external rf interference is
energizing the field such that the important
resonance frequencies have minimal interference
-
I only know of one site at -------.
We have never had to implement RF Shielding for our NMR systems
to my knowledge. We have good contacts with two very good RF Shielding
companies since we are installing RF shielding enclosures for our
MRI horizontal bore systems with much larger bore sizes (20cm, 30cm, 40cm).
-
hth,
dave scott
iowa state university
For a list of all the ways technology has failed to improve
the quality of life, please press eight."
-- Alice Kahn