No rifle Walter but I can hide in the shed
OK........ on to Phase 4
After the 'packaged' sensor with its new stable power supply had been running overnight in the shed I once again compared my plot with the Intermagnet web data from Dourbes.
Unfortunately the nice agreement had been lost and the trace showed a big dip - I assume as the temperature dropped overnight and started rise again?
There is a very interesting 'blip' at 09.03 - when I opened up the shed/observatory to set up for a quick imaging session - boy, is this thing sensitive to temperature
The new 5v supply has given some data that can be mapped to professional data
This performance is clearly compromised by changes in temperature
OK - now for a solution..........
After wasting an hour trying to image between the ever thickening clouds......... I packed up my gear, decided to bite the bullet and started to dig a hole
My garden has about 12" (30cm) of soil over heavy clay and building rubble so this was never going to be easy. After about an hour of digging and cursing I had a hole about 18" (45cm) deep with a solid clay base, away from metal objects and in the shade of a tree - probably as good as its going to get without major work.
I placed a concrete slab in the bottom - checked it was level in all directions using a spirit level - and added the waterproof sensor tube, bolted to a wooden plank. Checked again that it was level and used the Gyrocompass on my iPad to orientate the sensor E-W.
A bit of careful back-filling and routing of the cable along the fence line and viola - we have a thermally insulated (buried) sensor
Back in the house, I re-tuned the Bat 5 and then placed it in a plastic sandwich box surrounded by foam packaging.
This was then placed in a cool box also packed with insulating foam.
I think I have now done as much as I can - short of placing everything in thermostatically controlled packaging as Stuart has done
OK - this was at about 12.30 and I gave it an hour or so before before starting to record data.
The trace settled down quite quickly and was soon seen to be moving up and down rather than just drifting as it has in the past when voltage and temperature changes were clearly more significant than changes in the magnetic field.
At 9pm I fed the data collected into Stuart's spreadsheet and tweaked the graph to show it to best effect.
I then downloaded data from the Dourbes Magnetometer; scaled, inverted and tweaked it for comparison with my data and got this:
There seems to be some general agreement between the two data sets. It is quite difficult to get the time-lines to match and there are minor differences in detail but.......................... not bad for a first run.
There is nothing very dramatic happening at the moment with the trace staying between -20 and +10 nT. This is much smaller than the 140nT swings shown by some of Stuart's captures so........ maybe there is better to come.
I'm a bit baffled as to why my readings are all around the +19 or 20 nT value whereas the Dourbes data ranges from -20 to +10 nT??????????
Also not sure why my Rate of Change graph only shows a change of +/- 0.02 ????????????/
I will have another look at the spreadsheet - and hope Stuart comes riding to the rescue
I'm going to try re-tuning the Bat 5 to invert the trace and leave it running for a day before re-assessing progress.
Watch this space for the next thrilling instalment of........................ The Magnetometer Project