Email: Password: Remember Me | Create Account (Free)

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
09/11/12 06:03
Read: times


 
#188304 - Calibration + redundant force-limitation
Responding to: ???'s previous message
With a reasonably sensitive contact pressure sensor, you can pick up a sensor value just after having made initial contact (so a small measurement error before the sensor is sure it has contact) and then continue to move the probe down until the sensor reaches the force limit.

An issue here with an automatic device is that if there is a problem with the contact pressure sensor, the system will continue to push down so there must be a second set of mechanics that limits the amount of force available. That is very very hard to do when you have multiple sensor probes concurrently pressed to the skin unless each probe both have a contact pressure sensor for detection and some engage mechanics that have hard limits for how hard the pin may move forward (like a spring or similar). Your individual pins are very small, so even very low forces can do damage.

Another issue is that multiple contact pressure sensors will not be identically sensitive. So in soft tissue, there can probably be significant differences in how far you can press the individual probe pins before the system decides the required force has been reached. So there must be a way to calibrate each sensor tip.

One option for the mechanics would be pneumatics. With a fixed air pressure used to drive all measurement pins forward, you know exactly what force each sensor pin may push forward with before reaching equilibrium. So you wouldn't even need a contact pressure sensor to decide when to stop - just some sensor to decide when the measurement pin makes initial contact with the skin. And of course a sensor that measures how far it has been pushed out by the pneumatics. The skin contact could potentially be capacitive when it's enough to just know "touches skin" and the need to know how hard has been removed.

But still consider the manual route - one trick used there often is to project an optical grid on a surface. An operator can very quickly press a probe to each grid point. It isn't unreasonable that an operator with a single probe may be just as fast as an automatic contraption with 20 concurrent probes just because of the dexterity of our hands and our ability to quickly switch between high movement speed and great precision.

List of 49 messages in thread
TopicAuthorDate
How would you do this?            01/01/70 00:00      
   I have no idea if this would work, but ....            01/01/70 00:00      
      How would one actuate them?            01/01/70 00:00      
   strain guage            01/01/70 00:00      
      Yes, you might be onto something.            01/01/70 00:00      
         Computer-controlled?            01/01/70 00:00      
            computer controlled            01/01/70 00:00      
               Calibration + redundant force-limitation            01/01/70 00:00      
               Interesting problem            01/01/70 00:00      
                  Yes, it seemed coarse, at first.            01/01/70 00:00      
                     Precision is cheap ?            01/01/70 00:00      
                        I don't follow ...             01/01/70 00:00      
                           Precision            01/01/70 00:00      
                              other considerations            01/01/70 00:00      
         Tomographic imaging?            01/01/70 00:00      
            There's one major problem ... maybe two ...            01/01/70 00:00      
               $100 MRI unit            01/01/70 00:00      
                  I'm not sure ...             01/01/70 00:00      
         Voice coil and LVDT?            01/01/70 00:00      
            Not having a lot of experience with this sort of stuff ...            01/01/70 00:00      
               Danger/problems with hydraulic since no spring effect            01/01/70 00:00      
                  A few things puzzle me with this approach            01/01/70 00:00      
   Ultra Sound Probing            01/01/70 00:00      
      Ultra-sound is better for measuring surface or sub-surface            01/01/70 00:00      
   I would not use volumetric imaging            01/01/70 00:00      
      I'm not sure what you mean            01/01/70 00:00      
         another thought            01/01/70 00:00      
            I consider that one of the more "pregnant" approaches            01/01/70 00:00      
               Lots of options available            01/01/70 00:00      
                  You've got to consider cost            01/01/70 00:00      
                     Cost/safety            01/01/70 00:00      
                     Cost target?            01/01/70 00:00      
                        The price target comes from a few prosthetists            01/01/70 00:00      
   comercial CMM            01/01/70 00:00      
      This only received cursory examination             01/01/70 00:00      
         pneumatic actuator            01/01/70 00:00      
            Yes ... but what would this cost?            01/01/70 00:00      
               100 rods?            01/01/70 00:00      
                  Patient size ...            01/01/70 00:00      
                     That's what I'm considering            01/01/70 00:00      
                     AAACK! ... another double post ...             01/01/70 00:00      
                  it's not as big as you think!            01/01/70 00:00      
                     90% coverage            01/01/70 00:00      
                        So maybe vacuum pump for the too fat people            01/01/70 00:00      
   I do not know if this is relevant, but            01/01/70 00:00      
      I'll have to look into this.            01/01/70 00:00      
         Single air pressure sensor sensor for tip force guestimate            01/01/70 00:00      
   What about infrared?            01/01/70 00:00      
      How would you do this so it would retail for under $5k            01/01/70 00:00      

Back to Subject List