Journal Notes, October 2011 G2 magnetic storm
Build a Geomagnetic Observatory ! 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 magnet
ometer) 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
NOAA Space Weather Prediction Center, Top News of the Day! Space Weather Canada current space weather and regional forcasts .
Project Documentation, Links and References
Journal Notes:
Saturday, October 1, 2011
Sunday, October 2, 2011
Monday, October 3, 2011
Tuesday, October 4, 2011
Wednesday, October 5, 2011 G1
There is a warning for a possible minor G1 storm today: Space Weather Message Code: WARK05 Serial Number: 731 Issue Time: 2011 Oct 05 1315 UTC WARNING: Geomagnetic K-index of 5 expected Valid From: 2011 Oct 05 1315 UTC Valid To: 2011 Oct 05 2359 UTC Warning Condition: Onset NOAA Scale: G1 - Minor. There are many types of NOAA Space Weather Prediction Center email notices. You can sign up for NOAA space weather alerts at the NOAA SWPC Product Subscription Service. After you sign up for the email notifications (bottom of the SWN page) be sure to go to the "subscription" page to check off which categories (of the many types of notifications that are available) you want to receive (or, you get nothing). Since the beginning of our project, I had been glancing at the various space weather pages, with many of the indices dependent on several hour averages, so those indices are useful only well into an event. Some of the email notices (a free service) provide comprhensive warning and analysis.
The total field (F scalar) was very active here in upstate, NY today (USA) ~10 pm (0200 UTC), however not very storm like without large field excusions of 50 to 100 nT PDF. The 24 hour plot from NRCan Ottawa shows a similar plot including the three vector components that make up the total F field PDF.
Thursday, October 6, 2011
Overnight: PDF.
Friday, October 7, 2011
Saturday, October 8, 2011
Instrument health - polarization controller chart PDF. For new readers: the light blue fluid temperature curve is calculated directly from the NMR Tau2 fall time (measured each cycle). There is no fluid temperature sensor. The polarization time is automatically varied each cycle to achieve a constant amplitude precession signal. Our setpoint is presently for 0.85V out of the narrow band low noise amplifier (NBLNA). The orange curve is the ambient temperature at the sensor stand measured with a LM34 temperature sensor. Note that as the evenings get colder polarization times have fallen at times to near 0.6 seconds. Also, note that the fluid temperature is always about 20 degrees F higher than the sensor stand temperature and that there is a several hour lag in heating and cooling over the ambient temperature (with somewhat faster heating in direct sunlight). The NMR working fluid is presently Prestone De-Icer windshield washer fluid, good for upstate NY winters to -30C is in a 4 oz. (125 mL) plastic Nalgene bottle. The bottle is inside of a thick walled PVC pipe (open at the top), inside of a plastic container and the sensor stand is covered by a thick inverted plastic bag (the rain cover). The spikes in the green Tau2 curve are some sort of interference not yet uderstood. The Tau2 calculation has no affect on the field measurement related to the frequency estimator calculation (independent of amplitude affects).
Sunday, October 9, 2011
Monday, October 10, 2011
I updated our Article Part I introduction to include a reference to the Space Weather Prediction Center Email notification service - alerts, warnings, etc., there are many notifications to choose from. After you sign up for the email notifications (from the bottom of the SWN page), be sure to go to the "subscription" page to check off which categories of the many types of notifications that are available that you want to receive (or, you get nothing).
Also, I updated the Part IV coils and stand article with the following note about coil orientation: My understanding is that maximum precession signal from a solenoid can be had when the coil side (i.e. the polarization field) is normal to the local total field vector (the "F vector"). You can picture a plane perpendicular or normal to the vector. Any rotation of the coil within that plane should give a maximum precession signal (all other factors working properly). I prefer the coil's longitudinal (long) axis pointed N-S and tilted up by the inverse of the inclination (dip) angle (i.e. 90 degrees minus the local inclination angle) because it helps experimenters to visualize, as well as to explain to visitors, the direction of the Earth's total magnetic F vector, the direction of the total magnetic field. However, as you picture the plane, you can see that as the coil rotates in the plane to E-W for the long axis, it is now flat (picture the plane going through a table top). Also, since the coil is cylindrically symmetric, probably once it is E-W (longitudinal axis), you can now lay both coils down on a flat surface (e.g. a flat table top), for in that special case, any parallel plane probably also gives a maximum precession signal response.
Since a solenoid coil pair (coils always parallel to each other) can work optimally in any orientation within a plane normal the direction of the total field (the Earth's local F vector direction), this does raise the possibility of rotating the coils system in the plane to try to minimize any particular source of local interference (e.g. from nearby power lines). Ideally, if the coils are perfectly balanced, there would be no change in RFI/EMI pickup. However, if the coils are not perfectly balanced and there is a very strong source of local interference, there could be a significant improvement had by changing the orientation within the normal plane.
Tuesday, October 11, 2011
Overnight: PDF.
Wednesday, October 12, 2011
Thursday, October 13, 2011
Friday, October 14, 2011
Overnight: PDF, TXT, the geomagnetic field was very quiet overnight, nearly "flat".
Instrument health: polarization controller PDF (the fluid temperature calibration, based on each measured tau2, needs to be verified. The wide separation of fluid temperature with ambient sensor stand temperature overnight is possible, but suspect), FOM, FDM amplitude PDF (statistical mode 2e-7). The overall instrument health is excellent.
FID frequency as a function of time during the precession waveform: I have been thinking about a post processing routine to study very small changes in frequency across an individual precession (the free induction decay or FID) waveform. One approach is a moving window multiple FDM calculation (in a loop) to make a modulation domain analyzer type of frequency vs. time display. A graph of FID frequency of time should show, for example, the effects of a large or nearby vehicle in motion during the time interval of the precession waveform. In other words, it should be possible to observe the time variation of the magnetic field B(t) during a precession waveform. Since FDM calculations take about a second for on the order of 10,000 input data points, a loop of FDM calls might be less practical for the operating FDM PPM magnetometer, at least where the application is running in the background during other uses of the PC. On the other hand, if it works, a display of precession waveform frequency with time might be quite interesting. (Note that we already mostly exclude large delta f events (e.g. passing vehicles in motion) using the FOM filter and the auto-retry process).
I wrote a post processing LabView vi to analyze a particular precession waveform in detail. (This is not part of the basic FDM magnetometer system, and is only intended for academic interest and possible use someday in instrument accuracy verification and noise floor studies). The program takes a relatively long signal record of 2.5 seconds (25,000 samples at 10 kHz; the number of FDM points on the FDM magnetometer front panel that are saved to a "signal" file for the fdm.exe executable to act on). Now, where the FDM magnetometer runs the FDM executable once each measurement cycle (less than a second to execute), this new post-processing routine single steps (e.g. 500 times) across a single precession waveform record every millisecond with a moving window. FDM was run over again for each one millisecond shifted 12,500 point window. I used the same auto-retry thresholds that we use in the FDM magnetometer and discarded (did not plot) any results that were beyond the thresholds. The results are very interesting. Here is a sample precession waveform that had little variation during the first 1/2 second (the moving window (typically 12,500 points wide) with 1 ms steps for a total time shift of 500 ms from the first point that is normally used in the magnetometer). The first half second of exponential decay can be clearly seen in the amplitude graph (keep in mind, that while the delay range here is 1/2 second (500 ms, the longest delay), the window still goes out another 1.25 seconds, which is why the amplitude gets so small) PDF. Here are the results for a second free induction decay (FID) waveform with much more frequency (field) variation during the precession signal (note also the much higher FOM values) PDF. Using this method, it should be possible to somewhat separate measurement artifact variations from an actual changing frequency during some interval (e.g. the first 1/2 second) of a single FID waveform. It is encouraging to see that the relatively quiet waveform stays mostly within +/- 0.1 nT during the 500 successive FDM calculations. Note also that some excursions correlate well with higher figure of merit (FOM) values which might help to distinguish calculation variations with actual field changes in the 0.01 nT to 1 nT range. Keep in mind, however, that our instrument noise floor is still unknown.
Sunday, October 16, 2011
Instrument health: Several days ago I noticed the the fluid temperature (calculated directly from tau2) was no longer falling to match the ambient temperature before morning sun. I do not know yet why the fluid temperature equation constants c1 and c2 needed to be changed. I changed out the NMR fluid with another bottle of Prestone De-Icer to rule out any chemical changes (e.g. heating in the summer sun). There was no significant change in the fluid temperature value. The fluid temperature calculation has no direct relation to the PEV servo loop or more importantly on the FDM field value (the F scalar). I can only guess that somehow I inadvertantly changed a constant somewhere in the tau2 computation. I roughed in new constants c1=0.3944; c2=0.018, (they still need to be refined) All is well now: sample polarization controller chart, PDF. The lag in temperature change between the ambient temperature at the sensor and the fluid temperature can be hours and is normal.
Monday, October 17, 2011
Tuesday, October 18, 2011
Wednesday, October 19, 2011
Thursday, October 20, 2011
Friday, October 21, 2011
Overnight: PDF.
Saturday, October 22, 2011
Overnight: PDF.
Sunday, October 23, 2011
Monday, October 24, 2011 - G2 Geomagnetic Storm !
Overnight: PDF, TXT. Evening, there was a very fast positive excursion of the total field (>100 nT) about 7:20 pm local time. 200 nT vertical scale (normally 100 nT) PDF. 8:53 PM PDF. 10:10 PDF There has been an extremely fast negative going edge and negative "spiked" wavform PDF with very high dB/dt. We are now on a 300 nT vertical scale, the last excursion was from about 53,680 nT to 53,535 nT (about -145 nT). NRCan OTT (Ottawa) before 0000 UTC PDF, since 0000 UTC, 10/25/2011 PDF.
Interesting, a view of the USGS plots show a relatively low impact across most of the US (USGS PDF), however we apparently experienced at least at times, a "major storm" PDF here in the Eastern North America Sub-auroral zone PDF. The Eastern North America Sub-auroral zone has been down graded to "active" as I write this entry.
Email from the Space Weather Prediction Center has been quite good, including an alert at 7:17 and this warning at 7:26 (there were earlier notifications about an upcomming geomagnetic event):
Space Weather Message Code: WARK06
Serial Number: 182
Issue Time: 2011 Oct 24 2321 UTC
WARNING: Geomagnetic K-Index of 6 expected
Valid From: 2011 Oct 24 2325 UTC
Valid To: 2011 Oct 25 0300 UTC
Warning Condition: Onset
NOAA Scale: G2 - Moderate
You can sign up for NOAA space weather alerts at the NOAA SWPC Product Subscription Service. After you sign up for the email notifications (bottom of the SWN page) be sure to go to the "subscription" page to check off which categories (of the many types of notifications that are available) you want to receive (or, you get nothing).
Instrument Health: Sample spectra PDF, log spectra PDF. Instrument health is excellent, polarization controller graph PDF, FDM figure of merit (FOM) and amplitude chart PDF.
Tuesday, October 25, 2011
Overnight: PDF, TXT, return of magnetogram vertical scale of 300 nT to normal 100 nT PDF.
Wednesday, October 26, 2011
Thursday, October 27, 2011
Friday, October 28, 2011
Saturday, October 29, 2011
Overnight: PDF, TXT, interesting example of a very quiet overnight period turning into an unsettled period during the morning part of a quiet day diurnal cycle.
Monday, October 31, 2011
Overnight: PDF, TXT. There was an interesting triangular shaped waveform this evening PDF.
Want to build your own FDM Proton Precession Magnetometer?
Project Documentation, Links and References
QUESTIONS/COMMENTS/notice of typos, etc. send email to joegeller @ gellerlabs dot com
COPYRIGHT © 2009, 2010, 2011 JOSEPH M. GELLER, All rights reserved.