Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Help With Dynamic Force Calibration

  1. Mar 30, 2009 #1
    I am currently working on a project which involves me using a force sensor. So far I have tested the sensor statically. However, I need to calibrate it dynamically so that the output voltage can be used to determine a force in Newtons.

    I have done a bit of research so far and it seems that it is quite difficult to dynamically calibrate. However, I had thought about building some sort of spring-loaded device like the one below:


    I figured if I know the exact force being exerted on the force sensor by the strike pad, then I can determine a relationship between output voltage and force.

    I am a little unsure though. It is not really an area I am familiar with. If there is anyone on here who could help then it would be much appreciated.

    This is the force sensor I am using:

  2. jcsd
  3. Mar 31, 2009 #2
    Your method is pretty much what is carried out on dynamic load cells but with a little bit more sophistication. Generally a highly accurate strain gauge load cell is used to calibrate piezoelectric load cells, and would also be used to calibrate the Flexiforce. Your method relies on a spring constant being accurately known, as well as being able to accurately measure the displacement of the spring.

    Strain gauges are used to measure the deformation of the object they are bonded to, and by use of a gauge factor their output is converted to load. Whilst expensive strain gauge load cells might not be readily available to you, depending on your skills or those of a technician you have use of, you can bond strain gauges to a component yourself and use that as your calibrating load cell.

    The biggest drawback to using and calibrating the Flexiforce (conversely also one of its selling points) is the small area of application. You need to ensure that all of your applied load is transmitted through the sample area, and as such you generally use a 'puck' when calibrating. This would allow you to use or manufacture a load cell (whether it's a strain gauge component or your rudimentary spring measurement system) that's larger in diameter than the sensing surface.

    Out of interest, what forces are you measuring with the Flexiforce sensor?
  4. Mar 31, 2009 #3

    What is the easiest way to do it? I am not particularly clued up on what is required for calibration. I have actually designed a puck already to ensure that the load is transmitted through the sensing area. I will be measuring boxing punch forces. So between 1000-4000N most likely.
  5. Apr 1, 2009 #4
    Interesting - I've done a bit of work in that before. We tended to use accelerometers instead of load cells and the equivalent forces we were working with were towards the limit for the Flexiforce (4.4 kN if you change the resistor value in the drive circuit).

    You'll have to calibrate it against another load cell. Unless you know pretty accurately your spring constant and can measure the deflection accurately, you won't get much of a calibration. The ideal method is to apply the same load (much as you do through static testing) to the sensor and another load cell.

    Make sure your other load cell has a quicker rise time than the sensor, then apply a dynamic load through both (for instance tape the sensor flat on to an anvil attached to the load cell, then drop a falling weight onto your puck). You can compare a variety of different impacts between the two devices. You especially need to compare the duration of impact to how closely the sensor's output follows the load cell's output, as this correlation will be lower the shorter your impact.

    Edit - Flexiforce have a decent page on the process here.
    Last edited: Apr 1, 2009
  6. Apr 1, 2009 #5
    Thanks for that. I had a read at the Flexiforce page you linoked me to. I wasn't quite understanding what they were meaning. I am also not particularly clear on what you were suggesting above. Could you perhaps describe in a little more detail exactly how I would set this up? I am just not quite clear on what you are suggesting I do with the anvil.

    I don't actually have a load cell at the moment either. If I was to buy one of these would I need to build a circuit for that as well? On the Flexiforce site it suggested that if a load cell is not available then the next best option is to use the sensor for comparitive studies. Do you know what they are meaning?
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook