Maker Pro
Maker Pro

Searching for small load cell or method for measuring forces in the 1 g (10mN) range

rAlvarez

Feb 19, 2020
6
Joined
Feb 19, 2020
Messages
6
Hello,

I'm working o a project to measure the force response of very small polymer fibers and need a to find a miniature load cell (or another practical method) for measuring forces in the 10mN range.

Can anyone point me to some existing load cells or recommend ways for doing this?

I'm seeing some ~low-cost alternatives in the 10g range and some that claim 0.001g precision but I'm skeptical after previous experiences with 100g load cells.
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,878
Joined
Jun 21, 2012
Messages
4,878
Most load cells, at least the ones that I am familiar with, use strain gauges in an electrical Wheatstone bridge configuration to measure the minute compression or extension of a stiff "I-beam" shaped material when a longitudinal force is applied along the long axis of the beam. This type of load cell is well-defined and available with a huge range of full-scale load ratings. As you discovered, most of this range is on the high end rather than the low end of the range of forces. There are reasons for this, one of which is the limitation of strain gauges ability to operate over large ranges of strain.

Strain gauges are thin, serpentine, metal films whose resistance varies with the amount of strain applied. The percentage resistance variation from the original un-strained resistance divided by the strain applied is called the strain gage-factor. Typical gage-factors for foil strain gauges are about 2. Only slightly self-serving information can be found in this National Instruments Application Note 078. It's worth reading for starters.

Most load cells strive to be very stiff, only partly because the strain gauges mounted therein have a limited range of strain before they tear into pieces. Most people don't want any movement from the load cell when a load is applied, but this is impossible. Stress (force per unit area) is converted to strain (percentage compression or extension) because everything moves when stress is applied, some things more than others for a given amount of stress. It is the purpose of the strain gauge, tightly cemented to whatever the stress is applied, to measure this movement. Clearly, the smaller the applied stress, the smaller the resulting strain.

Eventually the change in strain gauge resistance becomes so small with decreasing load that it is immeasurable for all practical purposes. One way to get around this is to make the mechanics of the load cell more complicated, so that a given stress is "amplified" to produce more strain. For example, instead of mounting strain gauges on the long axis of an "I-beam" a thin flat metal leaf, cantilevered from a fixed support with strain gauges attached to opposite sides of the leaf, will bend with just a slight force applied to the free end of the leaf. The resulting strain is easily measured, but the downside is the leaf must "yield" (move) to effect a reading.

There are exotic techniques that have been used to "stiffen" load cells while allowing very small stresses to be measured. I recall one such technique employed optical interferometry, but that is not practical for most applications. Another technique, modulating the strain gauge excitation voltage and then using synchronous demodulation to detect the strain, is also a common means of extending the lower limits of the measurement range. Perhaps you could try this using one of the smaller, but still larger than ideal range, load cells.

If you cannot find a commercial load cell that meets your polymer fiber testing requirements, perhaps you can make your own, using a leaf spring of suitable stiffness. However, the art of attaching strain gauges is not easily learned.

You should also consider force-balancing techniques, where active feedback is used to balance fiber tension against the variable pull of, say, an electromagnet. A small loudspeaker with a cylindrical actuator device (for axial symmetry) cemented to its paper cone, said actuator attaching to one end of the fiber would be a simple way to implement such a device. A pair of photodiodes could perhaps "look" at a flag or a reflective tape attached to the actuator and adjust the current in the loudspeaker coil to keep the flag or tape centered between the two photodiodes as the opposite end of the fiber is pulled by some mechanism that varies the tension in the fiber. Other non-contact position-measuring schemes are also possible.

Years ago, a freshly minted mechanical engineer was given the task of isotonically supporting a newly dissected frog muscle that was electrically stimulated while the muscle contraction was measured. His solution, which I was then tasked to make work, was to build an elaborate wire-and-pulley mechanical balance with a movable weight to apply the tension. A motor moved the weight horizontally to select a desired tension while the position of the lever arm to which the weight was attached was sensed with a photocell arrangement. One end of the muscle tissue was attached to this contraption, while the other end was attached to a motorized fixture that pulled on the muscle to match whatever tension the movable weight applied through the wire-and-pulley arrangement. Because of the mechanical advantage between the balance arm with the movable weight and the attached muscle, the apparatus was quite "stiff" and therefore isotonic as requested.

This was a "fun" project with a practically zero budget, except our labor cost was covered by a small grant. Almost everything was constructed in-house using our rather extensive machine shop facilities. The guys there could build you almost anything if you had a ball-point pen and a paper napkin to draw it up on.

I am sure you want something commercial and off-the-shelf (COTS) so you can get on with testing your fibers, but sometimes COTS just isn't available and you have to "roll your own" instrumentation. AC-excited strain-gauge bridges are pretty much a "piece of cake" nowadays, so you might want to look into that as a means to extend downward the usable range of an existing load cell. Non-linearity and hysteresis problems might appear, but those can be "calibrated out" with appropriate data acquisition software and one-sided directional load application to remove the effects of hysteresis. In other words, you back off the tension, maybe even put the load cell into slight compression, before applying tension again to remove hysteresis.

Good luck with your testing. Please come back here if you have any questions.
 
Last edited:

rAlvarez

Feb 19, 2020
6
Joined
Feb 19, 2020
Messages
6
Dear @heavans1944,

Thank you so much for taking the time to share your experience and essentially confirming that I won't find and COTS solution.

Fortunately for my application, the displacement of the fibers is so large that reducing the stiffness of my loadcell is possible. I considered rolling my own load cell by mounting the strain gauges as you suggested but you have confirmed there is hidden art behind that.

Modulating the excitation sounds like an interesting approach that pushes the complexity to the electronics/software and allows me to use COTS load cells. I'll look into that to understand how it works and how to implement it. I took a peek at the subject and found some implementation notes, and confirmed that AC is the recommended way.

Very interesting muscle measuring contraption! My application is not that different.

You also made me think of perhaps the simplest solution is to use a small lever to amplify the force from the fiber by 10X so that the load cell can measure.

I saw in your profile that you occasionally consult. I'll reach out so we can explore the possibility of collaborating.

Thanks again!
 
Top