Tau 2 Feed Forward Polarization Controller Mode
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 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, Links and References (very early stages)
Journal Notes:
Monday, March 28, 2011
Overnight: PDF, TXT, Fast Forward Tau 2 (FFT2) polarization time controller PDF. There was a school bus parked and slightly moving within 100 feet of the sensor this morning for about 10 minutes near 8:00 am, turns out they were training a new driver. This has become a regular event, I showed them the sensor and asked them to use another location.
The FFT2 polarization time controller worked well overnight. The fluid temperature lags the sensor air temperature as previously determined. The lag is very significant, some tens of points at 2 minutes per point (about an hour). It seems the thick walled 2" pipe, even with the open end, is good thermal insulator. There is a strange disturbance in the Tau 2 measurement near minimum sensor air temperature (blue curve is fluid temperature derived from the Tau 2 mesurement). It does not correlate to the bus near the sensor. Our gas furnace vents from the side of the building and I might have guessed the heat comming on in the morning, however the LM34 sensor air temperature (orange curve) shows no sign of heating at the counter-wound coil sensor stand. Fortunately, the Tau 2 exponential curve is near its lowest sensitivity at such cool temperatures, so the peak envelope amplitude did not change much. Another mystery. I suppose it could have been some sort of electromagentic interference which distorted the precession waveform filtered envelope causing an error in the Tau 2 measurement.
The present working equation for the Prestone DeIcer is: Tau 2=0.42*exp^(temperature*0.0126). The slight deviation from 2 V peak amplitude early in the record indicates a slight error in our measured NMR Tau 2 c1 and c2 constants. The open loop "regulation" with the FFT2 polarization time controller is almost as good as the closed loop PEV servo.
Late morning, first solar heating, Fast Forward Tau 2 (FFT2) polarization time controller PDF.
Once an NMR fluid is chosen - and Tau 2 is measured each cycle, the fluid temperature (calculated only from Tau 2) is given by:
Temperature = (ln (Tau 2/c1))/c2, or Tau 2=c1*exp^(temperature*c2) .
The polarization time (to achieve a desired precession waveform peak envelope voltage) is given by:
Correction - [Polarization Time = (c3*Tau 2)+c4 ] Afternote: This was probably incorrect as an approximation over a relatively narrow range of Tau 2.. Later PEV servo data is indicating an exponential function: Polarization Time = (c3*exp^(Tau 2*c4)) 3/30/11 fit: PDF
It should be possible to determine c3 and c4 for the polarization time taking peak envelope amplitude data at two Tau 2 values. Calibrating the fluid temperature (determining c1 and c2) might need two actual immersion temperature measurements in the fluid.
Before shutting down for reprogramming, I took a view of the Tau 2 and assciated data for the day PDF Magnetogram PDF, TXT.
Here is the new FDM Magnetometer front panel PDF. The selector in the lower right hand corner allows for either a fixed polarization time, the peak envelope servo, or the feed forward Tau 2 controller (PEV and Tau 2 to hold a desirable 2 V peak envelope precession waveform voltage). New calibration constants were entered for Tau 2 to fluid temperature (c1, c2) and Tau 2 to polarization time (c3, c4).
Tuesday, March 29, 2011
Overnight: PDF, TXT, FDM FOM PDF, FDM Amplitude, PDF The local geo magnetic field was extremely quiet overnight (very flat, outdoor sensor calibration quality!). The figure of merit (FOM) most common value or "mode" was a very respectable 2e-7 overnight.
Feed Forward Tau 2 Controller: PDF, We added front panel push buttons last night to be able to switch between a fixed settable polarization time, the PEV servo, or the Feed Forward Tau 2 controller. We ran overnight on the Tau 2 controller. The constants still need some fine adjustment (as does the unrelated FDM Amplitude calibration). One reason these controller options are important is to study the NMR properties of various fluids. The FDM magnetometer has been performing perfectly in all modes, however we believe there might be some advantages to running in one of the two automatic modes that hold the precession waveform amplitude relatively constant over temperature.
I will begin a separate calibration page on how to make measurements to approximate the calibration contants. For users mostly interested in only magnetic measurements and not the NMR fluid characteristics of different fluids, as long as they use a fluid already measured, it will simply be a matter of plugging in the correct constants for that fluid. Calibration of the Tau 2 derived fluid temperature can be done with any suitable immersion probe at two temperatures and then by doing an exponential curve fit. The heat capacity of the 125 mL of NMR sample fluid is sufficient to take the temperature before there is significant change from being temporarily removed from the powered coil. Of course it takes some care not tip over the opened Nalgene bottle!
The LM34/LM35 style temperature sensor resting in the open end of the PVC sensor powered coil form has proven invaluable in understanding both the NMR fluid equations and the relatively large lag (on the order of an hour) between sensor air temperature and the fluid temperature. However, many users, especially those mostly interested in geomagnetic measurements, will probably find the Tau 2 derived fluid temperature plenty good enough. In fact Tau 2 is probably the most reliable and accurate measurement of the sensor NMR fluid temperature. It is remarkable that a time constant can so easily be converted to a fluid temperature!
Afternoon: I changed the polarization controller over to the peak evnvelope voltage servo mode (on the fly). With PEV servo data from the high temperature today and a low temperature tonight, I hope to reset the Tau 2 feed forward constants. The polarization controller graph shows how beautifully the PEV servo works PDF. So, why have a Tau 2 feed forward controller mode. There are some minor trade-offs. For example, the mean amplitude value with the PEV servo enabled, is very close to the desired set point, however the variation can be a bit higher then some periods of relatively quiet operation in the Tau 2 feed forward controller mode. I think for now the primary function of the Tau 2 feed forward controller is to verify the exponential equations for a given fluid. 10 pm local PDF, TXT.
Evening: There was a minor geomagnetic impulse event after 8 pm local (UTC -4 hours) PDF, also seen throughout North America USGS PDF.
Wednesday, March 30, 2011
Overnight: PDF, TXT, Polarization Time Controller PDF. Following the impulse early last eventing, the field was quiet overnight.
The polarization controller curves show the manual transition (just after peak temperature) from the fast forward Tau 2 control to the peak envelope (PEV) servo. Since the Tau 2 FF constants are not yet correct, the PEV servo operation is more successful at holding the desired set point, a peak envelope amplitude of 2 V. The trade-offs are also apparent. For example, note how while nearly perfectly holding the desired mean value of 2 V, the control parameter (polarization time) has more short term variation to accomplish the goal. The 50 or so point variation in Tau 2 before point 900 is similar to one a few days ago and remains a bit of a mystery (although, this time there is a change in the internal sensor air temperture before the Tau 2 disturbance.
Evening data: PDF, TXT, Polarization Time Controller PDF. Will shut down for a short period to change the Tau 2 to polarization time equation. Back up and running with the new equation relating Tau 2 to polarization time for the feed forward mode, so far so good PDF, polarization controller PDF.
Thursday, March 31, 2011
Overnight: PDF, TXT, Polarization Time Controller PDF. The geomagnetic field was very quiet overnight.
The polarization controller ran well overnight in the Tau 2 feed forward mode. It will take a warm day to see performance over a wider temperature range, however initial performance looks promising. Afternoon: The field remains very quiet, here is a good representation of our diurnal cycle PDF. The Tau 2 FF mode is still running well PDF. 10:35 pm PDF.
Friday, April 1, 2011
Overnight: PDF, TXT, Polarization Time Controller PDF. The geomagnetic field was very quiet overnight. The morning disturbance in the Tau 2 measurement remains a mystery, perhaps a period of direct sunlight, something to look at.
Afternoon: There is some increased geomagnetic activity this afternoon as seen by a down-turn in the field (long after the normal chnges or our local diurnal cycle) PDF. Evening: There was an interesting relatively fast and short positive going pulse, followed by a rapid drop of more than 20 nT from the pulse peak value, just after 8 pm local (NY, US, -4 UTC) PDF. The two day view shows the increased activity today PDF, TXT. The polarization controller continues to run well in the Tau 2 feed forward mode PDF.
I noticed that the polarization controller pdfs have been large files. I will try to remember to reduce the resolution for these plots, they seem to look fine at around 250 kbytes PDF.
Project Documentation, Links and References (very early stages)
QUESTIONS/COMMENTS/notice of typos, etc. send email to joegeller @ gellerlabs dot com
COPYRIGHT © 2009, 2010, 2011 JOSEPH M. GELLER, All rights reserved.