Hello all,
Looking for some insight and advice. Fairly new to the game but not too worried about setting aside time to study what needs to be done. Basically I am working on a project that needs some high speed datalogging over the span of <5s. And then a second project that is more basic that I think would be covered by most of what I would learn from the first project. (listing the abbreviated project brief at the end of my post).
For the most part I was planning on using a Raspberry Pi to record everything. I did consider using Sparkfun's datalogger/IMU combo but that would limit my ability to expand to the ADXL sensor and setup for the second experiment (I dont think it supports the load bar and slide pot). But realizing that the data rate for the ISM and ADXL would be quite high at their max sample rate, I thought that it might be better to use an Arduino board for each sensor, and then write them into a single file on the PI afterwards. The Pi would also be used to trigger the hammer and datalogging start through the use of a analog switch.
My questions are:
- Am I on the right track with my approach?
- Should the Pi and/or Arduino boards for this? Alternatively, the Sparkfun OpenLog Artemis sounds like a decent option but I cant tell if its confined to the Qwicc compatible modules.
- In either case (pure Pi logging or Pi + Arduino), is there a concern for bottlenecking? From what I've read SD cards are generally the biggest bottle neck but in this case I feel like the processing power and timing of data logging would be a bigger issue.
- Any particular datalogging libraries that you would recommend? For the most part I would be fine with a CSV output and processing later on PC, but if I am able to monitor data live though a USB-C monitor, that would be a big+.
Computing
- Raspberry Pi 3 (though if I can get away with a Pi 2 then I can reuse one I already have)
- Arduino Nano x2 +RTC module.
Alternatively, the Sparkfun OpenLog Artemis sounds like a decent option but I cant tell if its confined to the Qwicc compatible modules.
Setup 1 - Recording Accelerometer (and potentially gyroscopic) data from an impact.
Sensors
○ SparkFun 6DoF IMU Breakout - ISM330DHCX (Qwiic) [SEN-19764] (16g accelerometer)
○ ADXL375 - High G Accelerometer (+-200g) with I2C and SPI - STEMMA QT /
Sensors will be mounted to a test subject on a sled, then impacted with a "hammer". Hammer force is generally expected to be below 16g so the ISM330 will be the primary sensor. The ADXL375 is a backup sensor in case readings bottom out above 16g and expected to provide at least some redundancy in data collection should this occur. The ISM has a sample rate of ~6khz, while the ADXL is only 1khz. Logging time is expected to be around 5 seconds.
Setup 2 - force and distance of a slow moving load
Sensors
○ Straight Bar Load Cell - 5kg (e.g., SEN-13329)
○ HX711 Load Cell Amplifier [SEN-13879]
○ Slide Pot - X-Large (10k Linear Taper) [COM-09119]
○ ADS1115 ADC Module
In this experiment there is a lever that connects to what is essentially a compliant mechanism or trigger. I want to pull the primary lever and measure the load exerted at any given time or distance from its resting state. As the lever travels, resistance will increase until a breaking point where the resistance will break and the compliant mechanism will break off. The lever will be pulled using a linear actuator and expected to only reach around 10lbs at most.
Looking for some insight and advice. Fairly new to the game but not too worried about setting aside time to study what needs to be done. Basically I am working on a project that needs some high speed datalogging over the span of <5s. And then a second project that is more basic that I think would be covered by most of what I would learn from the first project. (listing the abbreviated project brief at the end of my post).
For the most part I was planning on using a Raspberry Pi to record everything. I did consider using Sparkfun's datalogger/IMU combo but that would limit my ability to expand to the ADXL sensor and setup for the second experiment (I dont think it supports the load bar and slide pot). But realizing that the data rate for the ISM and ADXL would be quite high at their max sample rate, I thought that it might be better to use an Arduino board for each sensor, and then write them into a single file on the PI afterwards. The Pi would also be used to trigger the hammer and datalogging start through the use of a analog switch.
My questions are:
- Am I on the right track with my approach?
- Should the Pi and/or Arduino boards for this? Alternatively, the Sparkfun OpenLog Artemis sounds like a decent option but I cant tell if its confined to the Qwicc compatible modules.
- In either case (pure Pi logging or Pi + Arduino), is there a concern for bottlenecking? From what I've read SD cards are generally the biggest bottle neck but in this case I feel like the processing power and timing of data logging would be a bigger issue.
- Any particular datalogging libraries that you would recommend? For the most part I would be fine with a CSV output and processing later on PC, but if I am able to monitor data live though a USB-C monitor, that would be a big+.
Computing
- Raspberry Pi 3 (though if I can get away with a Pi 2 then I can reuse one I already have)
- Arduino Nano x2 +RTC module.
Alternatively, the Sparkfun OpenLog Artemis sounds like a decent option but I cant tell if its confined to the Qwicc compatible modules.
Setup 1 - Recording Accelerometer (and potentially gyroscopic) data from an impact.
Sensors
○ SparkFun 6DoF IMU Breakout - ISM330DHCX (Qwiic) [SEN-19764] (16g accelerometer)
○ ADXL375 - High G Accelerometer (+-200g) with I2C and SPI - STEMMA QT /
Sensors will be mounted to a test subject on a sled, then impacted with a "hammer". Hammer force is generally expected to be below 16g so the ISM330 will be the primary sensor. The ADXL375 is a backup sensor in case readings bottom out above 16g and expected to provide at least some redundancy in data collection should this occur. The ISM has a sample rate of ~6khz, while the ADXL is only 1khz. Logging time is expected to be around 5 seconds.
Setup 2 - force and distance of a slow moving load
Sensors
○ Straight Bar Load Cell - 5kg (e.g., SEN-13329)
○ HX711 Load Cell Amplifier [SEN-13879]
○ Slide Pot - X-Large (10k Linear Taper) [COM-09119]
○ ADS1115 ADC Module
In this experiment there is a lever that connects to what is essentially a compliant mechanism or trigger. I want to pull the primary lever and measure the load exerted at any given time or distance from its resting state. As the lever travels, resistance will increase until a breaking point where the resistance will break and the compliant mechanism will break off. The lever will be pulled using a linear actuator and expected to only reach around 10lbs at most.
Statistics: Posted by tolebelon — Sun Dec 08, 2024 12:21 pm