So, had to go for an MRI earlier today, and was chilling (literally, they cool them with liquid nitrogen I am told) away there in the pipe for two hours, and got to thinking.. can we make a rudimentary MRI?
Stepper motor positioning system for at least x and y axes, with an electromagnet above and a hall effect sensor below. Different frequencies of modulation can even be easily done, and power modulated via PWM. Pulse the magnet and map the field as it decays. The induced field will be different based upon the material's ability to be magnetized. Vary the pulse to find the resonance and you have rudimentary MRI, correct? For that matter, it wouldn't be strictly necessary to have the sensor below if we were measuring the field decay and not the transmission of the flux itself, where the detector would opposite the emitter (the magnet). Math's going to remove the field of the magnet itself, so a side-by-side arrangement could work if just measuring field decay. Going to have to do some googles on rates of field decay and whether or not a Hall sensor and Arduino could keep up...
The hall sensor won't detect actual spin and we wouldn't be able to work in RF ranges like a "real" MRI, but we certainly can detect variations in the induced field. Though the sensor itself has plenty of error, that noise is easily removed by overlapped measurements and application of a little math. Move the magnet, pulse, measure, repeat... the map will build an image of the ability of the "sample" to hold an induced magnetic field.. which means it ought to map (at least in 2d) variances in that ability based upon it's makeup at a particular location.. voila, MRI. For example, if you embedded a wrench in concrete, something like this ought to be able to get a pretty good image of the wrench in contrast to the concrete. The question is how far you can push that in terms of resolution and material discrimination.
Going to have to think this over. Yes, it would be very low power and crude.. but it certainly could map the changes in the field enough to produce images at least as clear as a rudimentary xray in metals, certain minerals, or even maybe calcium-dense bone. I guess it's more like a glorified metal detector than a true MRI, but it does present an interesting idea in terms of an imaging project that a hobbyist could easily build..
Having done an electron spin resonant experiment as an under graduate (early 70s) I would say no. It was extremely tricky to see the resonant point on top of the large sin wave shown on the oscilloscope. And this was just in a simple crystal placed in a coil.
Also as you say the fields involved are so big that you have to have a super conductor magnet to generate them.
Yes, but aren't the big fields required because of the frequencies used (because of target being electron spin)?
Let's say I am placing a typical home-made electromagnet a student might make within three inches of that wrench in the concrete. I would think that based upon the proximity to the metal, there will be a variation in the field detected at the same time (or as it decays) by a hall sensor, and the field ought to be pretty easy to detect. Not super resolution or material discrimination, but at least workable. I would think this level would require VERY low power levels and would be rather easy to make... in fact, it's pretty much a stud sensor, isn't it?
The question I would think is how to go beyond being able to map a wrench an inch below concrete, and differentiate more than just ferrous/non-ferrous as well as have reasonable resolution..
Well that is not looking for ESR is it, it's just using a magnetic field like a metal detector.
You might get something from a hall sensor but I doubt it.
One reason for big fields is that the magnetic field drops off with the inverse cube of the distance so it drops off very rapidly with increasing distance.
There is an instrument that is a magnetic field force microscope. This scans a surface and reacts to any magnetic field it finds, but the field is way too small for a hall sensor, you have to use a flux gate magnetometer. With it you can do things like look at the magnetic markings on hard disc drive surfaces. I did help a colleague make one about 20 years ago.
Hmm. To be honest I know very little of Hall sensors other than they function as magnetic field detectors.. I guess I imagined that Hall sensors are substantially more sensitive than they actually are. Interesting.
looks like another subject I'm going to have read up on... lol
hehe.. xrays are a heck of a lot more dangerous than a magnetic field, though. Another fairly dangerous one might be using a Magnetron and a microwave antenna.. that's a lot closer to an actual MRI than what I'm proposing, anyway-- but insanely dangerous also. The x-rays would be worse in the long run, however.
I guess you'd call what I'm proposing an "induction magnetometer" or something of that sort.. glorified metal detector, but still, I like the idea from a "high school science fair" level..