Hi,
I'm working with a 300 Avance spectrometer. I'm trying to use a RARE sequence
to take flow encoded 2D images about every second or so. I've run into a
rather annoying limitation. I can take sixteen 128 x 64 images every 1.5s. If
I try and take a larger number, the program starts taking images and then fails
with the error message "tcuerror: <location in program where it failed> -
timing of pulse program too short".
The location in the program where it fails varies depending on the resolution,
number of images, and delays; there's not one particular place that's the
problem. It will stop running in the acquisition, during a gradient step,
during d1 and so forth.
The issue seems to be a combination of the number of phase steps, number of
images I want to acquire and my TR. The more phase steps and number of images
I want, the longer I need to make my TR. It doesn't seem to matter where I
increase the delays within the program so long as the total TR is increased.
Unfortunately, it's a necessity for the experiment that the images be taken on
an order of a second or quicker.
Something else I've noted that the higher number of images I try to take, the
sooner it will fail. For example, in a series of 32 128 x 64 images, it will
take 17 images before it dies. For a series of 64 128 x 64 images, it will die
after say 12 images.
I've tried using a single and multiple memory buffers. The write time doesn't
seem to be the issue; the program will still fail even if the data are not
written to disk. I thought perhaps it had something to do with using
acq_start, so I tried single point acquisition. Same problem. I thought
getting rid of the multiple 180 pulses might help, but with an EPI sequence the
same thing still happened after multiple scans. Multizg will take too much
time. I'm running out ideas. Any suggestions?
Best,
Kate
-------------
Kate Washburn
PhD Student
School of Chemical and Physical Sciences
Victoria University of Wellington
Wellington
New Zealand
email: kew@mit.edu
Tel: +64 04 463 5911
This archive was generated by hypermail 2b29 : Thu Dec 29 2005 - 01:02:06 PST