Just thought I would add my $0.02 worth.
Yes, the gradient amplifier is the most likely culprit. However, I believe that the gradient
amplifier probably receives its blanking pulses from the GCU, so that could also be at fault. On
Bruker systems, the gradient amplifier is blanked unless specifically unblanked during a pulse
sequence (the UNBLKGRD command). I have seen the odd case where a pulse sequence fails for some
reason with the gradient amplifier unblanked but with the gradients off, and a varying lock signal
is indeed the result. A pulse sequence that stops with the gradients on will trigger an error
shut-down of the gradient amplifier.
I have never seen this diode box that Rich is talking about - our INOVA600 system doesn't seem to
have one. Personally, I think that placing such a non-linear device in a part of the system that
should be extremely linear IS A REALLY DUMB ENGINEERING IDEA! This would distort the shape of any
shaped gradients and make specific gradient ratios (as required for coherence transfer in some
sequences) harder to achieve. You would always have to compensate for the fixed drop of the diodes,
which would be proportionally greater at lower gradient levels. Many years ago, these crossed-diode
blocks were also used in our RF amps as well. A blanking system (as now used in ALL our RF
amplifiers) would be the proper way to solve this engineering problem.
Cheers
-Kirk
Kirk Marat, Ph. D., NMR Facility Manager
Dept. of Chemistry
University of Manitoba
Winnipeg, MB, R3T 2N2, CANADA
C#, What C++ should have been
ph. (204) 474-6259 FAX: (204) 474-7608
kirk_marat_at_umanitoba.ca
ALL SPAM forwarded to Spam Cop
----- Original Message -----
From: "Richard Shoemaker" <Richard.Shoemaker_at_Colorado.edu>
To: <ammrl_at_ammrl.org>
Sent: Tuesday, August 12, 2008 2:00 PM
Subject: RE: AMMRL: Gradient current leakage and shimming...
> Dear AMMRL folks,
>
> Regarding the issue of gradient "off" current affecting shimming, just to
> add a few more details on the subject based on Josh's comments:
>
> This is often (if not always) a problem with gradient amplifiers. Varian
> has provided a solution to that for many years, which is a little box with
> diodes in it to cut-off current leakage after the output voltage drops below
> the turn-on voltage of the diodes. Most Varian system w/ gradients shipped
> with a little silver-box with XLR (micrphone/gradient) connectors on it....
> that's what this box is used for.
>
> Since the introduction of the cold-probe, the "new" version of this box has
> two switches called "Deadband" and "Damping". The Deadband serves the
> purpose mentioned here (to eliminate current leakage w/ the gradients off),
> and the Damping is to slow/dampen the slew-rate of current rise to prevent
> gradient ringing.
>
> With such a passive diode-switch on the output of the gradient amplifier,
> there should be no difference in shimming Z1 with the amplifier on or off,
> provided the gradient amplifier is not too far out of balance.
>
> Cheers,
>
> -Rich Shoemaker
> ---
> Richard K. Shoemaker, Ph.D., Director, NMR Spectroscopy Facility
> University of Colorado at Boulder
> Phone: (303) 492-7062 Fax: (303) 492-5894
> E-Mail: Richard.Shoemaker_at_Colorado.edu
>
> -----Original Message-----
> From: Josh Kurutz [mailto:jkurutz_at_uchicago.edu]
> Sent: Tuesday, August 05, 2008 6:34 PM
> To: Xi An Mao
> Cc: <ammrl_at_ammrl.org>
> Subject: Re: AMMRL: Cryoprobe tuning problem and gradient dirturbing problem
>
> Xi-an Mao,
>
> Regarding your lock instability, I suspect the gradient amplifier/
> power supply is the problem. We had the same phenomenon on our Varian
> Inova a little while back.
>
> When on, the gradient amp supplies a small current to the probe's
> gradients, which then act like shims. For fun, try shimming up a
> sample on a working system, then turn off the (z) gradient power. You
> should find that the lock levels falls and the lineshape gets
> distorted, but both can be restored by adjusting the z1 shim.
>
> If your problem is like ours, your gradient's "off" current is
> unstable, effectively mimicking an unstable shim. Thus, your gradient
> amp needs service. If you're sending it in for repair, you'll need to
> be in close communication with your repair engineers. I got the
> impression this is a weird failure, and the amp may still pass all
> their normal tests.
>
> Good luck!
>
>
> Josh Kurutz, Ph.D.
> Tech. Dir., Biomolecular NMR Facility
> University of Chicago
>
> On Jul 29, 2008, at 10:09 AM, "Xi An Mao" <xmao_at_wfubmc.edu> wrote:
>
>>
>> Dear All:
>>
>> We have problems with our two Bruker cryoprobes. One probe does not
>> response to proton tuning/matching, but the other channels (13C, 15N
>> and 2H) work fine. The other probe has unstable lock level. One when
>> the gradient cable is disconnected can the lock level remain calm.
>>
>> I am wondering if anyone of you had similar problem and how did you
>> fix it. I appreciate all feedbacks.
>>
>> Best regards.
>>
>> Xi-an Mao
>> Cleveland Center for Structural Biology
>>
>
>
Received on Thu Aug 14 2008 - 06:55:09 MST