I have been struggling recently in deciding whether to buy a Bruker DMX or a
Varian UnityPlus to replace one of our old Bruker AM consoles. Based on the
discussions with both Bruker and Varian sales persons, I am listing the
following statements, which can be shared with you. If you know of other
facts and info either from your own experiences or from other sources, please
send them to me. I'll summarize all the feedbacks later on.
1. Both UnityPlus and DMX have open structure for RF configuration. UnityPlus
can control up to 6 channels, while DMX can have 8 channels.
2. DMX has digital lock system which guarantees that the centerband is no
more than 50% of the height of the satellites in the spin-echo-difference (SED)
experiment. UnityPlus does not have digital lock or does not need one to
maintain the field stability to an adequate but somewhat lower grade (200%
vs 50%).
3. DMX uses digital receiver with oversampling technique, which is controlled
and processed by a separate cpu (Intel i960). With a 16 bit ADC running at
300kHz, they get effective 18 bit digitizer resolution. No more baseline
rolling problem that is common in older Bruker spectrometers.
Varian claims that they will have a software-version of oversampling
available. They do not have baseline problem anyway.
4. Both DMX and UnityPlus are using linear amplifiers. Varian claims that
Bruker uses cascaded amplifiers which might still have phase coherence problem,
while Varian is using one single amplifier for each channel.
5. DMX uses Hamiltonian (Bruker built, based on R3000 cpu) computer, which
is claimed to be totally compatible to SGIs and SUNs. They also offer SGI
Indigo, or Indy to replace Hamiltonian (This is still at the testing stage).
Varian is using SUN SPARC 10 Station (or older SUN SPARC 2). Bruker's
software is a little behind Varian's.
6. Varian claims that their pulse field gradient accessary is much more
reliable. Bruker, however is now using AcuStar pulse field gradient unit
developed based on GE's device, which does not seem to have the problems of
affecting the lock and the rf electronics. Varian also claims that Bruker's
shielded gradient probes suffer about 50% of S/N loss.
7. Bruker's probe has one advantage as far as service is concerned. The coils
are not glued on the quartz inserts. If a sample breaks in the probe or low
grade sample tubes have scrolled the insert, only $50 is needed to replace
the quartz insert. If Varian's probe is damaged this way, the repair cost
will be at least 10 times.
8. Maintenance costs for both DMX and UnityPlus can not be compared because
of lack of data. But some of you who have both Varian's and Bruker's NMRs
can perhaps provide a comparison. (The parts from Varian used to cost more
than from Bruker, but might not be true now, but might not be true now.)
Cheers,
Shaw
tel & fax: 617-495-3939 e-mail: huang@nmriris.harvard.edu