The basic point is, do you trust the pulse programming software to produce
what you code? For example I am trying to program the gradient enhanced
HNCACB experiment - many pulses and gradients on three channels including
1H and CO decoupling during the 'evolution'/'mixing' portions of the code.
I am not getting any signal (well, I get only noise). So how you I start to
debug all this? I agree the spins tell you a lot, but if there is no signal
then the communication lines have essentially been cut off. The DPS display
(Varian) for my code looks right, but is that what is really going to the
probe? The DPS display only displays the unpreprocessed code, not the compiled
sequence. At least if I could see what was happening at the probe I would have
more confidence in what was (not) going on. Colleagues who have set up the
in-line detectors tell me that they understand the characteristics of the
software/hardware interface a lot better, and are very enthusiastic about the
set-up. In principle the dectectors can be left permanently in-line because
the losses are low, and you do not need to manually adjust the pulse power
levels to do the testing, as I believe you would if you simply used a
directional coupler (divider) output to the scope. So this arrangement can act
as an essentially _passive_ monitor of the NMR experiment.
You might want the capability to look at the pulse characteristics too, which
reqiuires the high frequency (digital) oscilloscope. I am looking for a
cheaper, simpler solution that can gave you the multi-channel pulse envelope.
I wouldn't want/need it for 'simple' experiments.
I hope that helps,
Paul
---------------------------------------------------------------------
Paul Driscoll | Office (answer)phone: (44)-171 380 7035
Dept. Biochem. & Mol. Biol. | NMR lab phone: (44)-171 391 1354
University College London | driscoll@biochem.ucl.ac.uk