We had the same problem when we installed our GE Omega 500 ten years ago.
The application chemists at GE NMR told us we had a problem with large
periodic temperature swings that were affecting our 2D data sets. We
immediately went over
to our wall mounted temperature chart recorder and said "Impossible,
the temperature has been rock stable all week." After a little thought
about how thermostats work I came to the conclusion that our A/C could not
be that stable. As an experiment I touched the gas-filled temperature sensor
with my thumb and forefinger. The temperature almost immediately went up.
At this point I took my hand away and watched the temperature take nearly
20 minutes to come back down to baseline. Temperature recorders are
frequently set up to record average temperatures and this was our case.
As our recorder was too old to upgrade we got a new one from Omega
Engineering.
I ordered one that could use thin-film RTD's. This setup gives near
instantanious readings and showed that the room temperature varied about
5 degrees over a 20 minute time frame.
We then looked at our thermostat and found that it was rated for
+/- 2 degrees and was doing exactly what it was designed for.
We replaced it with one rated at +/- 1 degree. The temperature
swing was at least halved and the period went down as well.
However, the shorter on/off time of the A/C is now right at the rated
on time for the compressor motor. The motor needs a minimum run time
for the inrush starting current heat load to dissipate. It is difficult
to closely regulate room temperature with standard A/C's. I have been
told that the ideal setup is an A/C compressor that runs all the time
but uses "proportional electronic reheat" to control the final air
temperature delivered to the room. This works a lot like the VT unit
on the NMR spectrometer, however I haven't found anyone on campus who
has heard of "proportional electronic reheat". The office areas in our
building have hot water reheat, which rarily works well. I have also
been told that steam reheat is worse that no reheat at all.
The major concern though is the temperature of the compressed air going
to the probe. After the above problem was addressed we tested the VT air
temperature and found that it varied along with the room's temperature.
To minimize this we insulated all exposed air lines in the room including
regulators, values, filters, and pressure gages. We also insulated the tygon
tubing for the VT air, body air, and spinner air from the spectrometer to
the magnet/probe. This reduced the compressed air variation to
approximately 0.5 degrees. To improve on this we installed a huge,
insulated "passive" heat exchanger inline between the spectrometer and
the magnet/probe as close to the probe as possible. This improved the
variation to about 0.1 degree. We tried using just the passive heat
exchanger but it was not good enough without the insulated air lines.
If you are looking at small organic molecules in organic solvents you
probably will not have problems. If you are looking at proteins in D2O,
exchangable protons, or heteronuclei (Se, Te, Hg,...) that are temperature
sensitive you probably will have problems. And the problems will be greater
at higher field. We came up with better methods of temperature control
but I am not sure if they are really safe for the magnet/probe so I won't
go into them here. You can email directly if interested.
To answer the question asked, we use a circular chart recorder ( CT1000 series
~$1200 ) from
OMEGA Engineering, Inc.
P.O. Box 4047
Stamford, CT 06907-0047
1-800-826-6342
We also have used a Digital Transmitter (D1411, $325) with Pt thin-film RTD
to record data to disk via RS232. They now have free PC software but you have
to ask for it. I had to roll my own with UNIX shell scripts and PC basic.
I just noticed that their latest temperature catalog lists a lot of handheld
temperature meters, many for under $100.
At 05:19 PM 4/9/98 -0500, you wrote:
>Dear All,
> I want to record the temperature variations in our NMR room. I
>used a
>DICISON SL445C7 temperature record and the recorded results shows that
>the
>variation of temperature is less than 0.2 degree in one week.
> After site inspection by NMR vendors, I was told that the
>temperature
>variation is +/-2 degree C. There are two maximum and two minimum in
>each
>hour.
> I would like to get an inexpensive and fast responding
>temperature meter
>to record the temperature variations, but I do not know which company
>sell this
>kind of equipment. Any suggestion is appreciated.
>
>Weixing Zhang, Ph.D.
>NMR Manager
>St. Jude Children's Research Hospital
>Department of Structural Biology
>332 North Lauderdale Street
>Memphis, TN 38105
>Phone: (901)495-3169
>Fax: (901)495-3032
>http://www.stjude.org
>
>
-------------------------------------------------------------------
John A. (Art) Anderson, Ph.D. http://www.rrc.uic.edu
Spectroscopist Spectroscopy Facility
voice: 312-996-6640 Research Resources Center (M/C 937)
fax: 312-996-0539 The University of Illinois at Chicago
email: art@uic.edu 901 S. Wolcott Ave, Rm E102 MSB
Chicago, IL 60612-7341