Dear all,
The response to this discussion was considerable. A digest of the
responses may be downloaded from
http://andysoper.synthasite.com/ammrl-digests.php
I summarise it as follows.
1. There seemm to be two basic models:
The Vendor Support Model.
One person, generally a scientist, runs the instruments and breakdowns
and major servicing are/is undertaken by vendor service staff.
2. The Parolis Model.
This model I have named after Professor Haralambos (Lambe) Parolis who
bought our first instrument. He asks me to acknowledge that he got the
idea from the University of British Columbia, back in the late 1980's
or early 1990's. He also implemented an 'open shop' approach where the
engineer taught postgraduate students and staff to use the instruments
independently. That is another discussion point,perhaps :-))
A scientist and an engineer share the management of the instrument.
The Scientist is primarily responsible for applications support but
also works closely with the engineer regarding administration of the
instrument(s). The Engineer is responsible for the administration of
the instrument(s). It has to be this way since the engineer requires
intimacy with the instrument to keep her/his skills sharp. The
engineer has to have the freedom to administer the instrument(s) on a
day by day basis and to configure and reconfigure them and change
things in order to reinforce his/her training.
The Engineer needs to have a deep interest in NMR and needs to have
regular training by, and contact with, the vendor at the factory level.
3. In some cases a hybrid model is used where a scientist runs the
instruments, some service is undertaken by the institution's
electroics service staff, and the vendor staff are called for more
serious problems. I have the impression that this model can be quite
stressful. Having a non-nmr engineer work on a system, especially an
unshielded magnet system, may be risky.
4. The Vendor Model flourishes when the vendor is near-by ( a few
hours by road or air) and where a large support budget is available.
Some respondents mention maintenance contracts but I believe that all
abandoned them in recent times.
5. The Parolis model comes into its own when the instrument site is
relatively far from vendor support and / or support funds are limited.
Clearly paying for the services of an in-house nmr engineer also
requires funding. Our experience is that the Parolis Model more than
covered its costs in the avoidance of vendor service charges, reduced
down time and the improved applications support due to having two
people with rather diffent backgrounds but a common goal involved.
The Parolis Model promotes engieering skills in the scientist and
science/applications skills in the engineer, providing a measure of
protection agains the temporary unavailability of either one of these
staff.
The Parolis Model requires the active cooperation of the vendor.
Generally vendors are not too happy about committing skilled and
valuable staff for days at a time to distant instruments, and are
likely to be supportive. I can only speak for Bruker who have given us
wonderful assistance over the years. For them a plus factor is having
an engineering person on-site who can speak to and work with their
engineers. They speak the common language of nmr-engineering.
Again, you may find the digest of responses on my web site at
http://andysoper.synthasite.com/ammrl-digests.php
A warm thank you to all who participated. There is some talk of
continuing the discussion at ENC. If you would like to be directly
involved in this please contact Josh Kurutz at
jkurutz_at_uchicago.edu
Sincerely,
Andy
Received on Wed Jan 21 2009 - 15:58:00 MST