T7 Pro but not a software person - Geo sensors solution? | LabJack
 

T7 Pro but not a software person - Geo sensors solution?

2 posts / 0 new
Last post
Mark
MDVolle's picture
T7 Pro but not a software person - Geo sensors solution?

I have a rudimentary monitoring system set up with a previous unit but now have a T7 Pro - 

I would like to interface a sensor coil (analog magnitometer) in the quietest and widest dynamic range mode possible - coil is typically connected as a balanced source and is about 4k Ohm dc resistance - (45,000 turns on a 1" core 30" long). Frequency range of interest is 0.01 Hz to about 40 Hz. If the upper limit were reduced to as little as 20 Hz it would still be an improvement over my current arrangement. There is significant signal at 60 Hz from AC utilities - 

Typical signals of interest range from a microvolt (or less) to over a volt during seismic activity.

Suggestions? Best input configuration for this?

I am also looking for the ability to display a 24 hour spectrum display - which seems like an equally difficult challenge for a rookie.

This is a personal hobby lab - so expensive software suites are out - 

The T7 will also have an Epi Sensor triaxial sensor attached - but those outputs are straight forward balanced signals from the sensor's internal processing board. It is inherently a dc to ~200Hz bandwidth but the DC to 40 Hz is of most interest.

The Epi Sensor runs in its highest gain mode and is able to display the lunar influence on the z-axis acceleration.

Any guidance wil be greatly appreciated

Mark

 

 

 

 

LabJack Support
labjack support's picture
Unfortunately, we do not have

Unfortunately, we do not have any particular recommendations for the spectral analysis. The solutions we are most familiar with are paid options such as MATLAB, LabVIEW, and 3rd party post-processing software. A likely solution would be to do the analysis after logging in a programming language with good math and graphing libraries. Python and R are two options that come to mind. You could handle the logging side with our Windows based LJLogM software:

https://labjack.com/support/software/applications/t-series/ljlogm

My initial recommendation for the AIN setup would be to start by sampling both signals with the 16-bit ADC at resolution index 8 and ±10V range. Resolution index 8 will provide oversampling and averaging to help filter out noise and you will be able to use a high enough data rate to avoid aliasing. The following data rates page is useful for understanding typical sampling rates under various AIN settings:

 
If the measurements under these settings do not meet your requirements in terms of resolution or noise, there will be some trade offs to consider for your setup. Please see the considerations below.
 
The 24-bit ADC could provide excellent filtering for lower frequencies such as 60Hz, but could in turn filter some of your signal, and limit the acquisition rate of the accelerometer due to the sequential sampling of the AIN. For example, if we neglect any communication overhead to request samples, you use resolution index of 10 (which uses the 24-bit ADC) on the magnetometer, and use the fastest setting of the 16-bit ADC for the accelerometer, you would be limited to less than 75Hz sampling rate due to the 13.4ms sample time of the 24-bit ADC at resolution index 10. Additionally, this would result in aliasing for frequencies above ~75/2≈37.5Hz. We expect the -3dB frequency to be around 48Hz using the 24-bit ADC at resolution index 10, so aliasing should not be an issue for the 24-bit ADC and you could expect 60Hz to be fairly well filtered on that channel. The 16-bit ADC sampling would be subject to errors due to aliasing.
 
A possible solution to avoid the tradeoffs in AIN sample time and resolution index is to create low pass filters in front of the AIN.  This could also give you better control over the filtering. We sell an accessory called the LJTick-Proto that could be useful:
 
Expected filtering and data rates for other AIN settings can be extrapolated from the channel sampling times described in our data rates page, the AD7190 (24-bit ADC) datasheet, and the information described in note 8 on our AIN general specifications page: