Many thanks to everyone who responded to my question regarding this
subject. As usual, the responses were rapid and extremely useful. This
phase shift problem has been quite a head scratcher and the Bruker
service representatives have tried numerous things to remedy it. These
attempts include replacing the ecoupler, MCI board, receiver board,
preamps, and several other components. None of these helped the
problem. The conclusion is, as many of you also noted, that this a
design flaw in this particular model spectrometer. It does not however
cause a problem in most experiments and can be worked around using PHCOR
in experiments such as 1D selcosy, which it would affect. I have posted
below the original question and a summary of the answers.
In an issue also related to this spectrometer, we had a problem of
frying MCI boards and preamp housings. We went through about 5 boards
in a period of a couple of months and 4 preamp housings. Some of the
boards would work for a week or so and then die. Some of the boards
would smoke as soon as they were powered up. After much detective work,
our Bruker service representative tracked the problem down to pin LL on
the N1 connector (on the MCI board). This was +5 V to the preamp
housing which was grounded. This was fixed by removing pin LL from the
N1 connector. Just something to check if you also experience this
problem.
Sincerely,
Brian
K. Brian Killday
Research Specialist
Division of Biomedical Marine Research
Harbor Branch Oceanographic Institution
5600 U. S. 1, North
Ft. Pierce, FL 34946
Phone: 561-465-2400 ext. 456
Fax: 561-461-2221
Email: killday@hboi.edu
_________________________________________________
On January 22, 1998, Brian Killday wrote:
I have a possible problem with an AMX-500 that someone may be able to
help me with. When changing power levels, keeping the p1 constant, the
phase changes. For example, when going from hl1=1 to 2, the phase
changes about 15 degrees. This phase changes decreases as the power
level decreases. From 2 to 3 it changes about 10 degrees, from 3 to 4
about 7 degrees, etc. I noticed this while doing a paropt changing hl1
to determine the low power level for MLEV spinlock. Has anyone else
experienced this? Are there experiments where this may pose a problem
(such as the MLEV) and if so, how might this affect the experiment?
Thanks in advance for your input.
Sincerely,
Brian
________________________________________________
RESPONSES:
________________________________________________
We have seen this phase shift, at least on a 1989 vintage AMX.
I would certainly expect that o be a problem if you try to spin
lock at a given angle. You can probably use lower power for the observe
pulse
to reduce the phase difference between 90 deg pulses and spin lock.
Generally,
using HL = 0 will result in worse t1-noise in 2D than if you use HL = 6,
or whatever
you can get away with.
________________________________________________
Brian:
I think this phenomenon was common on older consoles (possibly like the
amx), on newer ones it is not observed. It is a lot nicer if it does not
occur, but its presence will not preclude performing most experiments
(including MLEV-tocsy). Perhaps someone else will respond as to whether
it
should occur with an amx console or not.
________________________________________________
Is the peak you are looking at on-resonance when you do this? if it is
far
off then you may just be observing the effect of a frequency offset.
When
calibrating pulses I find it best to monitor a pulse that is almost on
resonance.
________________________________________________
Dear Brian,
We also observe this effect with our AMX system. It causes problems
when
using shaped pulses.
The problem appears to be a design fault with the AMX system.
________________________________________________
Dear Brian,
I you have a linear amplifier, you may use Cortab to correct this
problem. But it is very unlikely that you have it.Then use the :r option
on the phase in the pulse program. For spin lock experiment, you do not
need to correct the phase. I use to the phase correction on selective
excitation experiments such as selcosy ( selco...), so have a look on
these pulse sequence so you can see how it is implemented.
________________________________________________
Brian,
This is normal for the type of high speed amplitude control used in the
AMX series of consoles. Programmable attenuators have less of a phase
change with power, but are MUCH slower. In fact, I don't know of
any power control system that won't have some phase change with changes
in power.
If two pulse of different power levels require a defined phase
relationship
(eg. 1D TOCSY) then it is necessary to calibrate the phase difference
and
allow for this in a pulse program. eg
(p11 ph1:r):tp1
.
.
.
ph1=(360) 0 180
where the acquisition parameter PHCOR[1] compensates the phase
difference between
the shaped pulse and the spin lock pulse.
For 2D TOCSY the 90 degree preparation pulse and the spin lock are
usually
at the same power. For ROESY they are at different power levels but the
absolute phase difference between them does not matter because of
the evolution time between the prep pulse and the spin lock (this is the
same with TOCSY as well).
________________________________________________
the amx does not have linear amplifiers, thus a phase shift with
changing power levels is to be expected, although your numbers are
much too high for normal operation.
in any experiment where you use different power levels in one
channel, such as a tocsy, you have to make sure that the pulses come
from the same direction. Otherwise you get bad or no results. for the
amx this can be easily done by adding the statement :r after the
pulse in the pulse program:
p1 ph1:r
Then adjust the phase with the statement phcor1 in the parameter set.
A further description how to do this you will find in Exp. 7.1 of
our book 100 and more basic NMR experiments.