Journal notes, Software Article, FID amplitude, Magritek Videos
GELLER Labs "Backyard Science"
Thoughts on a proton precession magnetometer design - a Proton Magnetometer Project. Build an Earth's field magnetometer.
The FDM MAGNETOMETER1 project is a low cost high performance proton magnetometer (a digital magnetometer) kit under development for universities and amateur scientists to be able to accurately measure and monitor changes in the Earth's total magnetic F field and to observe geomagnetic storms. Magnetic storms can cause large excursions in the field and are of concern to interests ranging from electrical power grids, radio communications, and satellite operations, to aurora watchers and amateur radio operators.
1 Filter Diagonalization Method "FDM" (harmonic inversion), see Jan 21 and Jan 23 entries, based on: Vladimir A. Mandelshtam, Howard S. Taylor, Harmonic inversion of time signals and its applications, Journal of Chemical Physics (1997), Volume 107, Issue 17, 1997, Pages 6756-6769
(be sure to hit refresh to pick up our latest changes and entries)
Project Documentation (very early stages)
Journal Notes:
Monday, December 13, 2010
Overnight: PDF, TXT. Evening: the NOAA / Space Weather Prediction Center Costello Geomagnetic Activity Index web page is predicting increased short term geomagnetic activity PDF. The Costello index occasionally over-predicts, however in our limited experience, we have found it to be one of the first and most reliable indicators of increased geomagnetic activity. There has been increased activity in our local magnetograms last evening and this evening PDF, but so far at sub-storm levels here in upstate, NY (6:37 pm est). It was another over-prediction: PDF, our magnetogram PDF, however I still think the Costello index is one of the best early warning / prediction indices. It did properly predict short term increased geomagnetic activity, just a bit too much.
Tuesday, December 14, 2010
Overnight: PDF, TXT several day view PDF, continued elevated geomagnetic activity (sub-storm levels), Costello 1 day index PDF, 7 day index PDF . Aftenoon: Curious fast geomagnetic event PDF at just about exactly 12 noon on our local magnetogram (NY, USA, 1700 UTC) seen throughout North America USGS PDF. The geomagnetic field remains slightly disturbed, but at sub-storm levels Costello PDF. Same event at NRCan (Canadian) stations OTT PDF, VIC PDF, FCC PDF, and BRD PDF. Another smaller fast event at 2:30 pm local (1930 UTC) PDF, and yet another at 18 UTC PDF. Well, for now the curious fast waveforms are done. Here is a good overall 12 hour view of the three of them from today's geomagnetic record PDF. I am a novice at these observations; it is entirely likely that such events are common and well understood. Evening: a little more activity this evening, now a NOAA k4 alert, PDF, magnetogram PDF.
Wednesday, December 15, 2010
Overnight: PDF, TXT, one relatively large slow disturbance, followed by a few smaller slow pulses overnight, otherwise the geomagnetic field here in upstate, NY was relatively quiet. Several day view PDF (probably going to reboot for Windows updates this afternoon which loses the cached values (the data remains in our TXT files).
Long term readers will recall our delight at finding online and reading the ME Halse PhD thesis (our July 26, 2010 journal note). From her writings, we learned of the NMR / MRI demonstration apparatus offered by the Magritek Corporation (no affiliation). A reader recently brought to our attention a very informative series of You-TubeTM videos by Magritek, which most readers of our project will probably want to view. These videos provide an excellent non-mathematical and very intuitive introduction to basic NMR and MRI principles.
Note, however that by comparison, our apparatus is a very simple (yet powerful) implementation using Earth's field NMR (EFNMR) for geophysics applications. For example, with carefully designed electronics and FDM signal processing, we use a single 2 second DC polarization field (on the order of tau 2), similar to Magritek's five second DC pre-pulse (which rides higher up towards saturation on a polarization vs. time curve). We do not need the millisecond (90 degree) transverse resonant oscillatory (AC) magnetic field. On the other hand, in most geophysics applications (e.g. an Earth's field geomagnetic observatory), we are looking only for a single high resolution measurement of the fundamental Larmor frequency from each FID signal, whereas the Magritek apparatus can be used to perform a wide range of teaching experiments ranging from spin-echo to NMR spectroscopy. Also, we do not need gradient coils for shimming, relying on the exceptional homogeneity of the relatively undisturbed outdoor field sensor. Of course this simplicity precludes the Magritek's MRI gradient coil experiments.
Thursday, December 16, 2010
Friday, December 17, 2010
Saturday, December 18, 2010 - First Draft of our article on software for the FDM Magnetometer
Overnight: PDF, TXT , on rare occassion, such as overnight, a vehicle parks relatively close (40 to 50 feet) to the sensor resulting in an offset such as the ~11 nT depression seen until about 2 am. In low traffic areas, such offsets are easily removed in post-processing. Automatic processing could identify this type offset and remove them, however there is also some risk that an automatic correction scheme could also inadvertantly remove part of a large excursion related to an actual geomagnetic event or geomagnetic solar storm. In higher traffic areas where vehicles arrive and depart often, such as high volume street side parking, where possible, it will be highly desirable to place the Earth's geomagnetic field observatory counter-wound sensor pair 70 to 150 feet from the road (more distance is better, however, less improvement is gained at the larger distances by the square law). For low occurance offsets, it is relatively easy to identify a single parked vehicle interaction to remove the offset from the magnetogram in post processing. Unfortunately, at about three or more nearby vehicles, it becomes very difficult to identify and correct individual vehicle offsets unless they each happen to have very different offsets. We have discussed previously the possibility of using array of three or more fluxgate sensors for automatic rejection of parked vehicles (by separating directionally sensitive signals from situations where all three or more fluxgate sensors move the same amount, thus confirming a geomagnetic event), while preserving the absolute accuracy of the FDM magnetometer reading, however good site placment of the FDM magnetometer sensor is a much better approach.
We posted a first rough draft of our article on the FDM magnetometer software! See the articles page. While writing the software article, I noticed and corrected a coding error. The ratio of acceptable measurements to total measurements (an indication of how many auto-retries have occured) is up to about a 70% success rate (over about 4 hours) from the previous ~60% success rate. The average FOM is also lower. It will be interesting to see if this higher success rate continues (as related to the coding error), or if it is simply a very good evening for EMI/RFI (low electromagnetic interference) with less passing vehicles.
Sunday, December 19, 2010
Overnight: PDF, TXT. It was just a quiet night last evening, by this morning the ratio of total measurements (including auto-retries) was back to about a 60% success rate. We attribute most of this average auto-retry rate to local EMI/RFI conditions and the rejection of moving vehicles.
I added a concept Drawing on the Docs page under the SWCTRL section to more intuitively show how the counter-wound coil pair is re-configured (remember to do a "refresh" to see the newest page) by the SWCTRL board. The reconfiguration is from a powered coil for proton spin polarization to a center-tapped counter-wound coil pair to pick up the precession signal with ambient RFI/EMI noise cancellation. The drawing is just to help visualize the two coil configureations and does not describe zero-current switching or the coil energy dump process (to achieve a fast fall time so that an EFNMR instrument (having a coil system oriented normal to the total field vector) does not need an additional transverse resonant oscillatory pulsed signal).
We have commented months past on the temperature dependence of the peak amplitude of the precession signal. The graph linked below shows peak FDM amplitude for all measurements, including the auto-retries, from about 1:40 pm yesterday afternoon to 4 pm this afternoon. While during the summer, if memory serves me correctly, the amplitude increased for cooler overnight temperatures. Now entering winter here in upstate, NY, it appears that the amplitude falls again as outdoor temperatures drop to below 10 degrees F. PDF The working fluid is about 4 Oz (125 mL) of Prestone De-Icer windshield wiper fluid. Note that we are presently operating with an automatic crossover supply (constant voltage to constant current), so it is unlikely that amplitude change is caused by the lower resistance of the coil and cable at colder temperatures.
As discussed previously, an inexpensive constant voltage power supply can be used for the polarizaton power supply. It might be necessary to make minor voltage adjustments to keep the polarization current between about 1.5 A and 1.8 A. Our present polarization supply, the hp or Agilent E3610A, is generally available for a used cost in a range of about $90 to $180. With an automatic crossover power supply, one sets the voltage slightly higher than needed for the warmest prevailing temperatures. Then the power supply automatically throttles back the voltage to what is needed to provide a predetermined desired polarization current. We have been operating for many months at a 1.55 A polarization current.
Note that our measurements are not averaged. Each measurement, even during the auto-retry cycles, results in a high resolution result that stands on its own.
Evening: Looks like there could be some geomagnetic activity early early morning: PDF NOAA Costello K index PDF, but we are closing up for the night, so we'll see what the magnetogram looks like later this morning.
Project Documentation (very early stages)
QUESTIONS/COMMENTS/notice of typos, etc. send email to joegeller at gellerlabs dot com
COPYRIGHT © 2009, 2010 JOSEPH M. GELLER, All rights reserved.