🚀 go-pugleaf

In article <BzDI7y.53I@cantua.canterbury.ac.nz> geoffrey@cosc.canterbury.ac.nz (Geoff Thomas) writes:
>Subject: Nuclear Magnetic Resonance (NMR/MRI)
>From: geoffrey@cosc.canterbury.ac.nz (Geoff Thomas)
>Date: Wed, 16 Dec 1992 22:23:57 GMT
>
>I am wondering whether an NMR machine could be run 'inside-out'. As
>far as I understand them, NMR imagers consist of a toroidal magnet
>system, through which the subject is placed. Would it be possible to
>use a bar magnet type of configuration so that the subject didn't
>need to be placed within the physical magnet structure?
>
>i.e.
>this:				instead of this:
>...........			    MMMMMMM
>...........			  MMMMMMMMMMM
>...........			MMMMMMMMMMMMMMM
>MMMMMM.....			MMMM...SS..MMMM
>MMMMMM.....			MMMM.SSSS..MMMM
>...........			MMMM..SS...MMMM
>.SS....S...			MMMM....S..MMMM
>SSSS...SS..			MMMMMMMMMMMMMMM
>..SSSSSS...			  MMMMMMMMMMM
>...........			    MMMMMMM
>
>where:
>M = magnet structure
>. = magnetic field
>S = subject
>
>
>Can it be done? What range would the system typically be capable
>of measuring? How big would it be? How much power would be required?
>
>Any help much appreciated. Thanks in advance,
>Geoff Thomas

To do NMR, you need uniform field. Single bar magnet can not produce that,
then you need to couple it with another bar magnet in the opposite side, and 
to contain the field, the magnets have to be coupled together, which you 
end up to have a toroid, and subject between the poles. To do imagine, 
you need to add gradiant field to it.
By the way, the toroid you are thinking, has zero field outside the core.   

--
Peace and Prosperity!
  ---==< 110 >==---