10k Sampling Rate from Load Cell

Hi all, I want to collect data of force exerted during the impact of a free-fall object using "100kN" Load Cell. The object load is about 100kg. I need to get 10000 sample/second because the experiment only run not more than 1 second.

From my research, load cell output needs to amplify example using HX711 due to the low resistance of the strain gauge. But HX711 sample rate is very low, up to 80 sample/second.

Is there any way I can get data up to 10k sample/second? I'm planning to use a higher bit of ADC with a high sampling rate such as AD7175-2CH. Is this way can work?

The HX711 appears to be a combination of the analog bridge amplifier and A/D converter. Using such a combination device limits you to the options given. Advantage is one chip to do two jobs. With that said, get a separate bridge amp and connect it to a faster A/D (such as the one you have selected). The next question to be asked: can the SPI interface on the Arduino initiate and rate the 10k 24-bit samples in a second?

Some thoughts: How fast is the load cell (slew rate)? What's the slew rate of the ADC? What's the total data size in 1s at 10kSPS and 16 or 24 bit resolution?

You'll need one of the bigger Arduinos with at least 20KB free RAM.

Have you worked out whether the load cell will withstand the energy of impact?

Is 10,000 samples a second fast enough ?

adwsystems: The HX711 appears to be a combination of the analog bridge amplifier and A/D converter. Using such a combination device limits you to the options given. Advantage is one chip to do two jobs. With that said, get a separate bridge amp and connect it to a faster A/D (such as the one you have selected). The next question to be asked: can the SPI interface on the Arduino initiate and rate the 10k 24-bit samples in a second?

aswsystems. my first thought to use the AD7175 is to get rid of the amplifier so I get more range of reading. then I realize, the problem that will arise is the range between zero point voltage and max load voltage from the load cell. if the range is small, getting rid of the amplifier will not work. I would go as your opinion, combining amplifier with faster 16bit adc. I need to find the load cell voltage range first to find a suitable amplifier.

DrDiettrich: Some thoughts: How fast is the load cell (slew rate)? What's the slew rate of the ADC? What's the total data size in 1s at 10kSPS and 16 or 24 bit resolution?

You'll need one of the bigger Arduinos with at least 20KB free RAM.

DrDiettrich, I'm not sure about the slew rate. it is a used load cell I got from the part store. besides the load cell, I need to get a reading from the rotary encoder to read the speed of load falling. I'm planning to get a faster controller but right now I only have Arduino for testing.

johnerrington: Have you worked out whether the load cell will withstand the energy of impact?

johnerrington, sorry I write 10kn. my mistake. actually, it is a 100kn load cell and it can withstand the force. the force will be around 10kn. correct me if I'm wrong.

Two light barriers could be used to determine the falling speed and also give a starting point for data storage.

Bare load cells do not have a fixed supply voltage, instead they are fed by the bridge amplifier of the ADC. Adding another amplifier IMO is not a good idea.

it is a 100kn load cell and it can withstand the force. the force will be around 10kn. correct me if I'm wrong.

Depends on the speed of the falling mass and the way energy is absorbed on impact, whether its an elastic or inelastic collision.

What exactly are you trying to measure? - the behaviour will vary depending on the nature of the two bodies (load cell surface and falling mass) the shape of the mass, its composition etc etc.

Unless the falling mass is spherical and of very uniform composition, structure, and Young's modulus (eg a ball bearing ) the results will be mostly meaningless