Troubleshooting RS4D Geophone Readings: What Am I Missing?

Applications/Uses
1

Hello Raspberry Shake Community,

I have a Raspberry Shake 4D (RE038), and recently, I attempted to compare its performance with the IMV VP 7000L by manually shaking both devices. I conducted the test at a frequency of 1 Hz with a peak-to-peak displacement of 10 inches. This corresponds to an amplitude of 5 inches (0.127 meters). For the Raspberry Shake, I performed vertical (up-down) shaking and recorded both velocity from the geophone and acceleration from the accelerometer. Below is a comparison between the theoretical calculations and the recorded data:

Theoretical Values:

We know that:

  • Displacement, x=Asin⁥(ωt)=Asin⁥(2πft)
  • Maximum Displacement: A = 0.127 meters
  • Maximum Velocity: Aω=A×2πf=0.127×2×3.1416×1= 0.79797 meters/second = 798 mm/second
  • Maximum Acceleration: Aω²=A×(2πf)²=0.127×(2×3.1416×1)²= 5.013 meters/second²

Recorded Data:

I collected 10 seconds of data and plotted the results:

Screenshot 2024-11-30 100817
Screenshot 2024-11-30 100817820×482 48.4 KB

  • Plot 1: Velocity data from the geophone (EHZ channel)

Screenshot 2024-11-30 100928
Screenshot 2024-11-30 100928795×480 56 KB

  • Plot 2: Acceleration data from the accelerometer (ENZ channel)

Data Conversion Code:

I used the following Python code (ObsPy library) to convert counts to physical units:

Converting Counts to Acceleration:

import obspy  

# Load the inventory file  
inv = obspy.read_inventory('RE038.xml')  # Inventory file of AM.RE038  

# Load and read the MiniSEED file  
file = "./AM.RE038.00.ENZ.D.2024.323"  
st = obspy.read(file)  

# Attach response and remove it to convert to acceleration  
st.attach_response(inv)  
st.remove_response(output="ACC")  # Convert to acceleration (m/s²)

Converting Counts to Velocity:

import obspy  

# Load the inventory file  
inv = obspy.read_inventory('RE038.xml')  # Inventory file of AM.RE038  

# Load and read the MiniSEED file  
file = "./AM.RE038.00.EHZ.D.2024.323"  
st = obspy.read(file)  

# Attach response and remove it to convert to velocity  
st.attach_response(inv)  
st.remove_response(output="VEL")  # Convert to velocity (m/s)

Observations and Questions:

I understand that manually shaking the devices is not a precise method. However, the accelerometer’s recordings align closely with the theoretical value of approximately 5 m/s². In contrast, the geophone’s velocity readings show a maximum amplitude of around 40 mm/s (considering around 80mm/s peak to peak), which is significantly lower than the expected theoretical value.

Given these results, I would like to seek your guidance on the following:

  • Is there an issue with my data conversion process from counts to physical units?
  • Are there any other factors or aspects that I might be overlooking in this experiment?

Thank you for your time and assistance. I look forward to your valuable insights.

Best regards,
Rahat Ibn Nabi

RSH.RE038.2024-11-30T04_51_10.logs.tar (2.9 MB)

AM.RE038.00.EHZ.D.2024.323 (4.6 MB)

AM.RE038.00.ENZ.D.2024.323 (4.8 MB)

RE038.xml (18.3 KB)

2

Hello Rahat Ibn Nabi, and welcome to our community!

Thank you for all the detailed observations you’ve made; it was quite an interesting read. I’d like to address your experiment and provide some insights, as some key points have to be considered:

  • As you know, the geophone in the Raspberry Shake is highly sensitive but has a limited dynamic range, optimized for detecting small ground movements. Manually shaking the device with a displacement of 10 inches at 1 Hz can introduce forces and displacements far beyond its design specifications. In such conditions, the geophone is almost certainly saturating, which could explain the discrepancy you have noticed.

  • Unlike the geophone, the accelerometers in our RS4D model have a broader dynamic range and are better equipped to handle higher amplitudes. This is why your acceleration data aligns more closely with theoretical values, while the geophone’s data appears attenuated or inconsistent.

As an important note: if the geophone is exposed to mechanical displacements or forces beyond its design parameters, there is a risk of physical or electrical damage to the geophone or its circuitry. We strongly advise against manually shaking the device to avoid unintended damage.

Also, while your calculations are theoretically sound, manually shaking the device introduces uncontrollable variables such as uneven motion, inconsistent force application, and non-ideal coupling with the geophone sensor. Such variables make precise comparisons difficult and can lead to unreliable data.

We recommend conducting tests in a controlled environment with known input signals from specialized vibration equipment, such as a shake table (or similar). This ensures accurate and reproducible results while staying within the sensors’ operational limits.

If you suspect any performance degradation or inconsistencies, we suggest running diagnostics on your Raspberry Shake and reviewing its recorded data under standard operating conditions. Please let us know if you encounter persistent issues, and we’ll assist further.

I hope the above can be of help.

1 Like
3

Thank you, Stormchaser, for your detailed and insightful response.

Could you kindly clarify what you mean by running diagnostics on the Raspberry Shake device? Are you referring to recording signals in a controlled vibrating environment under the standard operating conditions of the RS4D and then comparing the recorded signal? Or are there specific software tools or official diagnostic procedures for assessing the performance of the Raspberry Shake?

Thank you again for your support!

4

You’re more than welcome!

The ideal would be the first point you highlighted, that is, analyzing recorded signals in a controlled environment under the standard RS4D operating conditions.

Speaking about precise tools, as also mentioned in my previous reply, if you have access to something like this (Multi-Sensor Calibration Table - Earthquake and Infrasound Sensors), you can make precise measurements and validate the data with your calculations to see how they measure against the theoretical values.

You can also read more on how our Shakes compare to broadband seismometers here, if you want to get into more technical details: How does the Raspberry Shake compare to a broadband seismometer?