Shop OBEX P1 Docs P2 Docs Learn Events
Load or Force Sensors — Parallax Forums

Load or Force Sensors

ArchiverArchiver Posts: 46,084
edited 2000-06-17 21:22 in General Discussion
I am trying to measure individual finger and/or full hand grip
strength (Pressure excerted will be 150lbs or less) and then sending
the signal through an A/D converter to get a digital readout. I have
just purchased the BS2 and the 12bit A/D converter.
I would like to know if anyone has had any experience with Load or
Force Sensors. I have been looking at the Flexiforce Sensors from
Tekscan. They seem to be compact enough for my application. I am open
to any and all suggestions.
Thanks
Kay

Comments

  • ArchiverArchiver Posts: 46,084
    edited 2000-06-14 02:15
    Hello Kay,
    Load Cells transducers are utilized in weighing system.They needs a special
    kind of IC amplifier.AD7730 bridge transducer from Analog Devices
    is a complete front end for weigh-scale.I have ordered a one to build load
    cell based electronic scale.

    Regards
    Mohamed Refky

    >From: Kayclifton@a...
    >Reply-To: basicstamps@egroups.com
    >To: basicstamps@egroups.com
    >Subject: [noparse][[/noparse]basicstamps] Load or Force Sensors
    >Date: Mon, 12 Jun 2000 19:35:12 -0000
    >
    >I am trying to measure individual finger and/or full hand grip
    >strength (Pressure excerted will be 150lbs or less) and then sending
    >the signal through an A/D converter to get a digital readout. I have
    >just purchased the BS2 and the 12bit A/D converter.
    >I would like to know if anyone has had any experience with Load or
    >Force Sensors. I have been looking at the Flexiforce Sensors from
    >Tekscan. They seem to be compact enough for my application. I am open
    >to any and all suggestions.
    >Thanks
    >Kay
    >
    >
    >
    >
    >

    ________________________________________________________________________
    Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com
  • ArchiverArchiver Posts: 46,084
    edited 2000-06-14 16:39
    Here is a response to an earlier question on the topic of the Tekscan
    sensors. This had to do with their sensors configured in a fine grid to
    measure pressure distribuition inside a shoe:

    Beau Schwabe wrote:
    These sensors are "Smile" ...One reason we developed our own sensor
    for the purpose of monitoring pressures inside a prosthetic foot or
    a diabetic foot with the onset of "distal neuropathy" (<-loss of
    feeling in the extremities).

    The FSR's from TEKSCAN will not last long under the SHEER stresses
    inside a shoe, and quickly become "de-laminated". Any deep creases
    and the sensors become saturated. The software allows you to
    "normalize" or "zero-out" the returned values to accommodate for
    any sensor bending, but again, when you do this it severely dulls
    the response and range of the sensor.

    The TEKSCAN sensor is great if you are standing still, but for any
    "real" gait analysis where a patient must walk, the sensors poorly
    survived our study. After $10K+ we quickly learned our lesson the
    hard way.

    My own application of this type of sensor a few years ago was under pans of
    food that were being visited by animals, as a kind of indication of the
    animal's presence. So that was more like "standing still"-no shear stress.
    The sensor resistance is temperature dependent and quite nonlinear, and
    repeatable only to within a few percent. But you cannot beat it for low
    mass.

    -- Tracy Allen
    Electronically Monitored Ecosystems
    http://www.emesystems.com
  • ArchiverArchiver Posts: 46,084
    edited 2000-06-14 18:48
    Thank you for your response on the Tekscan force sensor. Since I don't want
    to totally rule it out as a possibility, since it seems reasonably priced and
    pretty straight forward compared to any thing else I've come across. But
    since I have no experience with sensors, it's hard to make a wise decision.
    You mentioned that:
    "The sensor resistance is temperature dependent and quite nonlinear, and
    repeatable only to within a few percentage. But you cannot beat it for low
    mass."
    What is meant by temperature dependent?
    When you said nonlinear, would that mean I would get gaps in my readings?
    What is considered low mass? Would that describe the 0 to 150 LB range that I
    want to measure?
    Since my intention is to compress the cell "head on" would this eliminate
    shearing problems.
    How long or how many repetitions were you able to get from the sensor?
    Did you use the independent sensors (the ones that did not have to be
    connected to the software, that is the ELF) and if so how did you calibrate
    them?
    I know I've thrown a lot of questions your way, and any response will be
    appreciated.
    Thanks
    Kay Clifton
    Kayclifton@a...
  • ArchiverArchiver Posts: 46,084
    edited 2000-06-14 23:51
    >>>
    Date: Mon, 12 Jun 2000 19:35:12 -0000
    From: Kayclifton@a...
    I am trying to measure individual finger and/or full hand grip
    strength (Pressure excerted will be 150lbs or less) and then sending
    the signal through an A/D converter to get a digital readout. I have
    just purchased the BS2 and the 12bit A/D converter.
    I would like to know if anyone has had any experience with Load or
    Force Sensors. I have been looking at the Flexiforce Sensors from
    Tekscan. They seem to be compact enough for my application. I am open
    to any and all suggestions. Thanks Kay
    <<<

    Kay, there are quite a number of ways to do this - depending on degree
    of motion you can tolerate from 0-150# and accuracy, repeatability.

    One of the simplest things you might consider is to use a potentiometer
    mechanically coupled to a squeeze spring [noparse][[/noparse]like a grip exerciser] so
    that squeezing the spring makes the potentiometer turn. Read the pot.
    with RCTIME

    Another possibility with close to zero travel is a piston coupling a
    non-compressible fluid to a pressure transducer which has appropriate
    electrical output.

    Another possibility is an oscillator with a mechanically actuated
    capacitor or inductor which varies its frequency: then divide this
    down to a lower frequency with flip-flops and measure with PULSIN.

    Resistor pads that vary resistance with pressure have been made
    - one company is Interlink.

    You can also use an encoder wheel coupled to a spring - I've seen this
    in bathroom scales. Digital bathroom scales are a cheap source of the
    whole system - spring electronics and digital display... [noparse][[/noparse]most seem to
    use LVDTs or linear voltage differential transformers as the transducer].

    You can modify a digital bathroom scale to be hand squeezable with some
    levers and pivots... but that would leave out the STAMP!

    Then there are strain gauges on springy material.... and carbon
    piles and granules which change resistance when you squeeze...

    You don't say anything about speed of response or stability etc. so
    I in turn don't know what to say about that... The encoder wheel and
    spring give a direct digital output which you'd read with as many
    ports as you need for resolution - or turn the signal into a serial
    output to one port. With muscle contraction that you imply you are
    liable to get quite a bit of "noise" or varying grip tension so you'll
    need to deal with that...

    Good luck - if any of this is unclear to you ask [noparse][[/noparse]direct if you wish].

    ________________________________________________________________
    YOU'RE PAYING TOO MUCH FOR THE INTERNET!
    Juno now offers FREE Internet Access!
    Try it today - there's no risk! For your FREE software, visit:
    http://dl.www.juno.com/get/tagj.
  • ArchiverArchiver Posts: 46,084
    edited 2000-06-17 21:22
    Kayclifton@a... wrote
    > Thank you for your response on the Tekscan force sensor. Since I don't
    want
    > to totally rule it out as a possibility, since it seems reasonably priced
    and
    > pretty straight forward compared to any thing else I've come across. But
    > since I have no experience with sensors, it's hard to make a wise
    decision.
    > You mentioned that:
    > "The sensor resistance is temperature dependent and quite nonlinear, and
    > repeatable only to within a few percentage. But you cannot beat it for
    low
    > mass."
    > What is meant by temperature dependent?

    It means that the resistance of the sensor depends not only on the force
    applied to the sensor, it also depends on the temperature by about 2% per
    degree C. Say that the resistance at a certain applied force is 10000
    ohms. If you see a change in resistance down to 9000 ohms, it might be due
    to a 5 degree C change in temperature, or it might be due to a change in
    applied force. Does that matter? Not if you are dealing with relatively
    large changes in force and relatively small and slow changes in
    temperature.

    > When you said nonlinear, would that mean I would get gaps in my readings?

    No, it just means that the relation between resistance of the sensor and
    the applied force is not an exact simple math function. It is
    approximately resistance = 1/force. But there is more to it. The sensor
    has "wear in", that is, the properties change the first few times you apply
    pressure. Then there is a slow drift of the properties, if a load is
    applied continuously, that can amount to 5% change. There is "hysteresis",
    which means that the resistance is history dependent--from a constant
    force, going up and back down will be different from going down and back
    up, by a few percent. Again, these considerations are unimportant if the
    range of forces and time scale you are looking at are well over the few
    percent range and faster than the slow drift. If you need to "push the
    envelope" your software may have to compensate in some way, say, to
    establish a slowly changing baseline, or to compensate for change
    sensitivity with temperature.

    > What is considered low mass? Would that describe the 0 to 150 LB range
    that I
    > want to measure?

    I was referring to the sensor element itself. It is very thin, light in
    weight and simple and the electronics to read it out can be very simple.
    You can purchase the sensors with a wide range of sensitivity to applied
    force.


    > Since my intention is to compress the cell "head on" would this eliminate

    > shearing problems.
    > How long or how many repetitions were you able to get from the sensor?

    They still work after almost 10 years, but most of that time has been in
    storage! Total iterations maybe 1000. Hardly can make any extrapolation
    to what you want to do with finger pressure.

    > Did you use the independent sensors (the ones that did not have to be
    > connected to the software, that is the ELF) and if so how did you
    calibrate
    > them?

    Yes, we purchased the individual sensor "buttons" in bulk. I sandwiched
    three of them under three legs of low-profile pans, so that we could
    "weight" the objects on the pans. The sensors were in parallel, so that to
    first approximation,
    total weight = Force1 + Force2 + Force3
    = 1/R1 + 1/R2 + 1/R3
    And then we calibrated the pan by putting known weights on top. The
    software also had to establish a base line and some temperature
    compensation. We had to reduce our expectations several times, but the
    study did provide useful qualitative and quanititative information at low
    cost.

    > I am trying to measure individual finger and/or full hand grip
    > strength (Pressure excerted will be 150lbs or less) and then sending
    > the signal through an A/D converter to get a digital readout. I have
    > just purchased the BS2 and the 12bit A/D converter.

    If you really want to be quantitative about static forces, forget about the
    FSR technology. It is not accurate. Premena and others provided some
    alternatives. If you are looking for qualitative information in the 10%
    accurate range or with the time course of the grip strength, well it might
    work. But keep in mind the issue of shear forces.

    > I know I've thrown a lot of questions your way, and any response will be
    > appreciated.
    > Thanks
    > Kay Clifton
    > Kayclifton@a...

    You're welcome, and good luck!

    -- Tracy
Sign In or Register to comment.