*********************original question*************************************
hello everyone,
we, bowling green state university, are fairly new into this techniques of
pulsed-field-gradient nmr at our unity+400 instrument. we would like to ask
for everyone's opinion/suggestion/experience on the following issue.
our pfg system was installed and tested by varian's engineer and we did run
couple pfg experiments with good results. the question we have is -
are there ways of routinely checking at the performance of our pfg system?
just like the line width/line shape specification, are there standard pfg
experiments we can run (say, every 6 months) and check at its specification?
(for example, the s/n ratio, or lines' separation of standard pfg experiment
with standard sample.) we'll appreciate all of your responses and be glad to
summarize the information we receive. chen
**********From jwg4@slc10.INS.CWRU.Edu Wed Oct 4 10:41:55 1995**********
I am moving and have all my journal papers in boxes...,
but there was an interesting paper in the journal
"Magnetic Resonance Imaging"
where in Europe, there they tried to compared different machines.
Sorry I do not have the reference, but it was in 85-88 (approx).
They detailed the experiments which were used on a whole body scanner.
jim
**********From cgregory@uiuc.edu Wed Oct 4 10:53:45 1995**********
Three standard tests are "gradient amplitude", "settling time", and
"stability". For 3-axis system, run gradient amplitude on each axis, and
settling time and stability on all 3 axes at once (to save time, if
possible).
Gradient amplitude:
Run a sequence (on our VXR-500S it is called "profile") that applies a
gradient during the acquisition. Measure the width of the line with a
standard sample - this should not change. (This is the same procedure used
to calibrate the gradient in gauss/cm).
Settling time (also called "eddy current" test):
Run an arrayed experiment with variable delays between a longish gradient
pulse (e.g. 50ms at 25% maximum amplitude) and an rf pulse/acquisition
(e.g. "g2pul" sequence). Look at the peak amplitude and linewidth vs delay
(typically a multiexponential decay) and the frequency vs. delay. There
should be a spec for your probe (e.g. linewidth with no gradient = x% of
linewidth with gradient at y msec after pulse).
Stability:
Run any favorite pfg sequence, with a dummy array (e.g. array d1 but make
all the steps the same). A 2dft should show all of the signal along the
center in f1. Any sidebands along f1 indicate instability. Measure their
amplitude and keep a record.
The first two of these at least should have been run by the factory rep
when the pfg accessory was installed.
Hope this helps.
Carl Gregory
Biomedical Magnetic Resonance Laboratory
College of Medicine
University of Illinois
1307 W Park St
Urbana, IL 61801
(217)-244-2350
**********From hilton@ncifcrf.gov Wed Oct 4 11:04:16 1995**********
Well, I would run the gradient calibration to check the gradient strength
(doped water) and the gradient recovery test. Both of these are quite
fast, and are a stand--ard part of the Varian psglib.
If you want more, get a concentrated sample of say 2,3 dibomoproprionic
acid and set up conditions so you can run an hmqc in 10 minutes (very narrow
narrow SW) or less.
Run a gradient hmqc under a uniform set of conditions and keep the spectra.
This will allow a fairly demanding test in under 30 minutes. But the tests
above are important because they concentrate on gradient preformance.
Dr. Bruce D. Hilton NCI-FCRDC hilton@ncifcrf.gov 301-846-1226
**********From kfw@lubrizol.com Wed Oct 4 12:56:49 1995**********
Chen:
I have had PFG for over 2 years on my Bruker AMX500 and routinely use
various PFG experiments.Typicall, I do very little as far as calibrations
or standard experiments besides tweeking the gradient amplitudes with 1D
exp before doing a 2D exp. All the experiments that I typically use run as
expected. I would be interested in the replys that you get back to see
what others have to say in about this subject matter. I guess you could
run the typical installation experiments to check gradient strengths and
linearity of the gradients but I would be interested to see if others are
doing more elaborate tests to check their PFG systems.
Thanks in advance,
Kurt
Kurt Wollenberg
e-mail:kfw@lubrizol.com
Research Chemist
ph#:(216)943-1200 ext2026 fax:(216)943-9022
NMR Spectroscopy
Lubrizol Corporation
**********From nieman@asu.edu Wed Oct 4 13:11:16 1995**********
Varian provides two tests, which start with the p2pul sequence: profile and
grecovery.
Profile will give you a very good indication of whether the efficiency of
the gradients has changed over time (look at the macro 'setgcal' and the
parameter 'gcal'. the grecovery test will tell you about any problems with
degredation of the coils.
Another set of tests involves using "real" pulse-sequences. We use the
ghmqc sequence, which is exquisitely sensitive to gradient strength (array
'gzlvl3' by 10 to 20 with ni=1, nt=32 on a 10-20mg/ml sample).
We have found the gradients on our 400 and 500 to be rock solid. We tested
them routinely each week. They just don't change. These tests every few
months will ensure that the gradients are working.
---------------------------------------------------------------------
Ron Nieman, Director
Nuclear Magnetic Resonance Facility
Department of Chemistry and Biochemistry
Arizona State University
Tempe, Arizona 85287-1604
(602) 965-3613
Fax (602) 965-2747
email: nieman@asu.edu
**********From scott@iastate.edu Wed Oct 4 14:34:13 1995**********
chen,
We had the same trouble when specifying gradient performance in
our rfp but settled on: After a 0.0002 second 10 gauss/cm gradient
pulse followed by a .001 second delay, the residual signal should be
95% of the signal obtained without a gradient pulse. The lineshape
specifications of the probe must be obtained with the gradient amplifier
blanked at zero current. Fast 2D gradient experiments (3 min hmqc and
20min hmbc) must produce spectra of similar quality.
You might want to talk to Paul Keifer (415-424-6609) at Varian because
he knows a lot about gradients and the tests you could perform. I
just tend to run a quick gradient experiment on a stable standard
sample.
good luck,
dave scott
iowa state university
515-294-4057
**********From drethwis@icaen.uiowa.edu Wed Oct 4 17:29:48 1995**********
We spend a fair amount of our research on PFG. We are using it to
measure the diffusion coefficient of gases dissolved in polymers. We
are using an MSL300 and our probe is a Doty probe (high field gradient
and up to 300 C). In terms of a standard, we have used 50/50 H2O/D2O
for our calibrations. We don't really spec the line width (our probe
doesn't give the sharpest lines since it was optimized for other things)
or the S/N but look more at the gradient strength.
If you want more details let me know.
Also, please let me know what you find out. Thanks.
Professor David Rethwisch
Dept of Chemical Engineering
University of Iowa
Iowa City IA
drethwis@icaen.uiowa.edu
**********From JSCARSDALE@Gems.VCU.EDU Wed Oct 4 20:39:14 1995**********
In response to your questions concerning pfg specifications, the
Varian engineer should have run two tests, a gradient recovery
test and a gradient profile test both of which should be in the
appropriate subdirectory. It is my understanding that the
gradient profile test is a good test of both the pfg current
amplifier and the gradient coils in the probe. The gradient
recovery test is a good test of the pfg amplifier and its
ability to deliver gradient pulses as well as insuring that
there are not other problems which interfere with the systems
ability to recover after a gradient pulse. Both of these tests
should be run on the doped D2O sample. If you run these tests
every two months, as you suggested, and compare the results to
those obtained after the initial installation, you should be able
to obtain at least a rough idea of gradient performance. I
believe these tests are documented in the accessory installation
guide, if I remember correctly, as I have only run them once
when we had a problem with our pfg amplifier after our upgrade
to vnmr 5.1. I find that I use the gradients often enough (whenever
possible) that I can generally note any changes in experimental
performance, and if I noted any such changes would run these
gradient tests to try to isolate any system fault.
J. Neel Scarsdale
Assistant Professor Biochemistry and Molecular
Biophysics,
Virginia Commonwealth University
*******From thio@johann.chemie.mu-luebeck.de Thu Oct 5 12:10:41 1995*******
Hi Deng-Ywan (hope that's your first name),
I would be very interested in getting a copy of
the resonces of your question about the calibration
of pgf's. We just got a new drx500 and will face the same
problem.
Many thanks in advance,
thomas
/\ /Thomas Weimar
/ \ _ / Institut of Chemistry
/ \ _| |_ / Medical University of Luebeck
/ \_| o |_/ Ratzeburger Allee 160
/ oOOOo 23538 Luebeck
| _______ OOOOO ______ |
| Tel. x451 500 4231
| _______ ______ Fax. x451 500 4241
| /C-O\ |
|_________ C C_____thio@johann.chemie.mu-luebeck.de
\C-C/
**********From gsr3g@avery.med.virginia.edu Sun Oct 8 21:08:24 1995**********
Your should check the gradient strength and gradient recovery
time on a semi-regular basis. Just follow the test procedures
that are given in the Varian install manual.
x-----------------------------------------------------------x
| |
| Gordon S. Rule gsr3g@Virginia.EDU |
| Department of Biochemistry (804) 924-2683 (office) |
| University of Virginia (804) 924-1337 (NMR lab) |
| Charlottesville, VA |
| |
x-----------------------------------------------------------x
**********From bantalek@Kodak.COM Mon Oct 9 10:22:38 1995**********
Dr. Chen,
We have had a pfg (and diffusion) package on our Unity 500 for over a year
now. We calibrate the gradients in our coherence transfer experiments
regularly and save the values. Usually they are very consistent. As far as
specifications are concerned, we usually are satisfied with two tests: 1)
eddy current ringdown time, 2) gradient strength. Both tests are described
sufficiently in the Varian manuals. We generally see a ringdown time between
300 and 400 microseconds. We can achieve 33 gauss/cm for a maximum gradient
strength. These have not changed during the time we have had the system. We
generally check them every 3-6 months.
Let me know if you have any additional questions.
/) .
/) /^ / (/ /)
Magnetic Resonance
Analytical Technology
**********From: "BrukerFan (tm)" <brukerfan@dmx.600>**********
Hi,
you should through away your (what's its name?) immediately and
buy a Bruker instead! The technology is way ahead and they have
it digital and on 800 MHz and all that...
And they have all these requested Experiments pre-installed.
If you can't do this, but I can't think of a reason, why not,
contact your (what's its name?)-office and ask for help.
Best wishes
BrukerFan (tm)
**********From west@qtp.ufl.edu Tue Oct 10 08:42:29 1995**********
Please summarize what you learn.
John West
west@qtp.ufl.edu
**********From: "BrukerFan(tm)" <brukerfan@dmx.600>**********
I wrote:
> you should through away...
############
Yes, I see, I'm sorry, you couldn't knew that!
It's Bruker-Slang.
It means: "antimatter-annihilate"
This can be simply achieved, buying a Bruker spectrometer,
which has an anti-(what's its name?) in his extent of
delivery. Throwing (now I meanthrowing) this device at the
(what's its name) you will get plenty of freespace for new
spectrometers.
These anti-(what's its name?)s are also available separately
from Bruker on request.
BrukerFan(tm)
*************************************************************