Loading ...
Sorry, an error occurred while loading the content.

Gripping Fragile parts with Robotic Arm

Expand Messages
  • David Jason
    Hi,    I am a newbie. It might be very basic and typical Question :)   I have a 6 DOF Serial robotic arm manipulator (ED7220C). At the moment it is only
    Message 1 of 17 , Nov 1, 2012
    View Source
    • 0 Attachment
      Hi,
       
       I am a newbie. It might be very basic and typical Question :)
       
      I have a 6 DOF Serial robotic arm manipulator (ED7220C). At the moment it is only equipped with motor encoders. What I want to do is to let the gripper hold various kinds of objects (fragile objects like egg without breaking it and rigid objects like metallic). One way is to add Force sensors to the Gripper and then implement closed-loop feedback control system. What kind of sensors I need? Any clue (part no)? Would be happy to read your experiences? Secondly, is there more better and easy way of achieving my goal?
       
      Thanks for your help.
       
      Regards,
      DJ
      ____________________________________
       
       
    • Max Cato
      Hi DJ, I m by no means an expert on this, but I thought I d take a stab. For fragile stuff, I d recommend a piezoelectric or capactive sensor. A quick google
      Message 2 of 17 , Nov 1, 2012
      View Source
      • 0 Attachment
        Hi DJ,

        I'm by no means an expert on this, but I thought I'd take a stab. For fragile stuff, I'd recommend a piezoelectric or capactive sensor. A quick google search turned up this: http://www.trossenrobotics.com/c/robot-force-sensor-fsr.aspx which may or may not be what you're looking for, but could at least be a good starting point. 

        Hope this helps!

        -Sean


        From: David Jason <davidjason441@...>
        To: "SeattleRobotics@yahoogroups.com" <SeattleRobotics@yahoogroups.com>
        Sent: Thursday, November 1, 2012 11:32 AM
        Subject: [SeattleRobotics] Gripping Fragile parts with Robotic Arm

         
        Hi,
         
         I am a newbie. It might be very basic and typical Question :)
         
        I have a 6 DOF Serial robotic arm manipulator (ED7220C). At the moment it is only equipped with motor encoders. What I want to do is to let the gripper hold various kinds of objects (fragile objects like egg without breaking it and rigid objects like metallic). One way is to add Force sensors to the Gripper and then implement closed-loop feedback control system. What kind of sensors I need? Any clue (part no)? Would be happy to read your experiences? Secondly, is there more better and easy way of achieving my goal?
         
        Thanks for your help.
         
        Regards,
        DJ
        ____________________________________
         
         


      • Peter Balch
        Hi DJ As no-one else has answered yet, I ll have a go. ... I d say there isn t really a standard part that people use. You ll be pretty much building your own.
        Message 3 of 17 , Nov 6, 2012
        View Source
        • 0 Attachment
          Hi DJ

          As no-one else has answered yet, I'll have a go.

          > What kind of sensors I need? Any clue (part no)?

          I'd say there isn't really a standard part that people use. You'll be pretty
          much building your own.

          In general, people line the inside of the gripper fingers with something
          that measures pressure. If it were me, I'd look at the following:

          QTC. QTC is good stuff but you buy it as little rubber "pills" whose
          resistance goes down with pressure. But they are _really_ hard to work with.
          In particular, you don't just get a pair of wires to solder to - you have to
          somehow put a metal conductor on the surface of the pill. How? Glue it on
          with conductive glue? I've never found a good answer.
          http://www.peratech.com/qtcmaterial.php

          Resistive sensors. These are paper-thin pads that measure pressure. They're
          reasonably responsive (conductance proportional to pressure) but have
          problems with noise, non-linearity and drift (and, it seems to me, have
          strange reactance properties so your amplifier oscillated easily). They're a
          bit expensive and hard to buy in small quantities.
          http://www.tekscan.com/
          Here's an article comparing different makes
          http://www.google.co.uk/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&cad=rja&ved=0CEgQFjAB&url=http%3A%2F%2Fciteseerx.ist.psu.edu%2Fviewdoc%2Fdownload%3Fdoi%3D10.1.1.186.4191%26rep%3Drep1%26type%3Dpdf&ei=3chqUOGeH8mk0QW7oIDAAw&usg=AFQjCNGSTAi7Ud7WzivqRksz9JmMYQtwqw&sig2=dT37Z5O04J-EKeDX_8DSjA

          Strain gauges: A possibility, if you're building a big robot arm, is to use
          strain gauges. They're generally expensive devices and not often used by
          hobbyists but, if you take apart an electronic bathroom scale (from a
          charity shop?) you'll usually find four small good quality sensors in the
          form of "pads". Kitchen scales can also be good - they usually contain one
          beam sensor. You'll have to make your own precision amplifier (which isn't
          as hard as it sounds).

          Carbon impregnated foam. This is the really cheap and cheerful alternative
          often used by hobbyists. Just put a sheet of tinfoil on either side of some
          conductive foam (the sort electronic components often come packed in. There
          are various grades of foam - some is spongy some is more brittle. The
          disadvantage is that it only lasts for a few thousand operations before it
          starts to deteriorate. And it's not particularly linear. And it has a lot of
          noise. But for hobby robotics, it will do the job and it's really cheap.
          https://www.google.co.uk/search?q=conductive+foam+sensor+images

          There are other sensors you might consider that don't use resistance -
          capacitative sensors for instance. I suspect they'd give better results than
          many cheap resistive sensors but that's more of a research project.

          Peter
        • jamericanfreddy
          i have done some testing on sensors and capactive works better then piezo and sparkfun has a lots of them, also flex force sensors work great too
          Message 4 of 17 , Nov 29, 2012
          View Source
          • 0 Attachment
            i have done some testing on sensors and capactive works better then piezo and sparkfun has a lots of them,
            also flex force sensors work great too

            --- In SeattleRobotics@yahoogroups.com, Max Cato <maxsthekat@...> wrote:
            >
            > Hi DJ,
            >
            > I'm by no means an expert on this, but I thought I'd take a stab. For fragile stuff, I'd recommend a piezoelectric or capactive sensor. A quick google search turned up this: http://www.trossenrobotics.com/c/robot-force-sensor-fsr.aspx%c3%82%c2%a0which may or may not be what you're looking for, but could at least be a good starting point. 
            >
            > Hope this helps!
            >
            > -Sean
            >
            >
            >
            > ________________________________
            > From: David Jason <davidjason441@...>
            > To: "SeattleRobotics@yahoogroups.com" <SeattleRobotics@yahoogroups.com>
            > Sent: Thursday, November 1, 2012 11:32 AM
            > Subject: [SeattleRobotics] Gripping Fragile parts with Robotic Arm
            >
            >
            >  
            > Hi,
            >  
            >  I am a newbie. It might be very basic and typical Question :)
            >  
            > I have a 6 DOF Serial robotic arm manipulator (ED7220C). At the moment it is only equipped with motor encoders. What I want to do is to let the gripper hold various kinds of objects (fragile objects like egg without breaking it and rigid objects like metallic). One way is to add Force sensors to the Gripper and then implement closed-loop feedback control system. What kind of sensors I need? Any clue (part no)? Would be happy to read your experiences? Secondly, is there more better and easy way of achieving my goal?
            >  
            > Thanks for your help.
            >  
            > Regards,
            > DJ
            > ____________________________________
            >  
            >  
            >
          • Peter Balch
            ... I can t see any capacitive force sensors there. Which are they? And, yes, piezo sensors tend to respond to rate of change of force whereas for a gripper
            Message 5 of 17 , Nov 30, 2012
            View Source
            • 0 Attachment
              > i have done some testing on sensors and capactive works better then piezo
              > and sparkfun has a lots of them,

              I can't see any capacitive force sensors there. Which are they?

              And, yes, piezo sensors tend to respond to rate of change of force whereas
              for a gripper one generally wants the static force.

              Peter
            • jamericanfreddy
              it was at work while back when i was doing some testing for a digital scale ,i was in charge of design and test lab,mostly i do testing on sensors they give me
              Message 6 of 17 , Dec 3, 2012
              View Source
              • 0 Attachment
                it was at work while back when i was doing some testing for a digital scale ,i was in charge of design and test lab,mostly i do testing on sensors they give me and i build high precision in-house testers
                they gave me piezo sensor and capacitive type of force sensors
                if you go to sensors then capacitance you will see some chips
                little of hard part is to make sensor ,from what i remember you need 2 copper plates and material in between them
                sparkfun does have a lot of flex force sensors that might be much easy,i have some havent added to my arm design,i just ordered my metal lathe machine and milling machine to make my arm parts and more

                --- In SeattleRobotics@yahoogroups.com, "Peter Balch" <peterbalch@...> wrote:
                >
                > > i have done some testing on sensors and capactive works better then piezo
                > > and sparkfun has a lots of them,
                >
                > I can't see any capacitive force sensors there. Which are they?
                >
                > And, yes, piezo sensors tend to respond to rate of change of force whereas
                > for a gripper one generally wants the static force.
                >
                > Peter
                >
              • Peter Balch
                ... Thanks. I can see some touch sensor chips but not any capacitive force sensors or chips. I know that capacitive force sensors exist but I haven t found an
                Message 7 of 17 , Dec 3, 2012
                View Source
                • 0 Attachment
                  > they gave me piezo sensor and capacitive type of force sensors
                  > if you go to [sparkfun ] sensors then capacitance you will see some chips

                  Thanks.

                  I can see some touch sensor chips but not any capacitive force sensors or
                  chips.

                  I know that capacitive force sensors exist but I haven't found an easy
                  supply yet.

                  Peter
                • jamericanfreddy
                  Here is the page for one type ,they do have others https://www.sparkfun.com/products/7902 Now to make the sensor you need a 2 tiny copper plates and some soft
                  Message 8 of 17 , Dec 4, 2012
                  View Source
                  • 0 Attachment
                    Here is the page for one type ,they do have others
                    https://www.sparkfun.com/products/7902

                    Now to make the sensor you need a 2 tiny copper plates and some soft insulator material to go between them
                    Its been a long time since i work on that project,but since i got a full machine shop with every machine need,like a lathe and milling machine
                    Will be working on my hand design some time soon
                    I am thinking thin layer of open cell foam might work,can easy get it at home depot and one side is sticky,i think the copper needs to be facing each other,
                    ALSO there is circuits for capacitance to analog
                    FRED
                    --- In SeattleRobotics@yahoogroups.com, "Peter Balch" <peterbalch@...> wrote:
                    >
                    > > they gave me piezo sensor and capacitive type of force sensors
                    > > if you go to [sparkfun ] sensors then capacitance you will see some chips
                    >
                    > Thanks.
                    >
                    > I can see some touch sensor chips but not any capacitive force sensors or
                    > chips.
                    >
                    > I know that capacitive force sensors exist but I haven't found an easy
                    > supply yet.
                    >
                    > Peter
                    >
                  • jamericanfreddy
                    Do see they are touch sensor chips many circuits on the internet for capacitance to voltage and if you have a really good meter it has the circuit,its great to
                    Message 9 of 17 , Dec 4, 2012
                    View Source
                    • 0 Attachment
                      Do see they are touch sensor chips
                      many circuits on the internet for capacitance to voltage
                      and if you have a really good meter it has the circuit,its great to test the capacitance sensor first and then build the circuit around it
                      most like that chip i gave a page too the front end uses that type of circuit and then add a touch on/off circuit

                      --- In SeattleRobotics@yahoogroups.com, "Peter Balch" <peterbalch@...> wrote:
                      >
                      > > they gave me piezo sensor and capacitive type of force sensors
                      > > if you go to [sparkfun ] sensors then capacitance you will see some chips
                      >
                      > Thanks.
                      >
                      > I can see some touch sensor chips but not any capacitive force sensors or
                      > chips.
                      >
                      > I know that capacitive force sensors exist but I haven't found an easy
                      > supply yet.
                      >
                      > Peter
                      >
                    • jamericanfreddy
                      small search found this AD7150 capacitance to analog voltage chip with I2C buss,i think a 555 timer can work too
                      Message 10 of 17 , Dec 4, 2012
                      View Source
                      • 0 Attachment
                        small search found this AD7150 capacitance to analog voltage chip with I2C buss,i think a 555 timer can work too

                        --- In SeattleRobotics@yahoogroups.com, "jamericanfreddy" <jamericanfreddy@...> wrote:
                        >
                        > Do see they are touch sensor chips
                        > many circuits on the internet for capacitance to voltage
                        > and if you have a really good meter it has the circuit,its great to test the capacitance sensor first and then build the circuit around it
                        > most like that chip i gave a page too the front end uses that type of circuit and then add a touch on/off circuit
                        >
                        > --- In SeattleRobotics@yahoogroups.com, "Peter Balch" <peterbalch@> wrote:
                        > >
                        > > > they gave me piezo sensor and capacitive type of force sensors
                        > > > if you go to [sparkfun ] sensors then capacitance you will see some chips
                        > >
                        > > Thanks.
                        > >
                        > > I can see some touch sensor chips but not any capacitive force sensors or
                        > > chips.
                        > >
                        > > I know that capacitive force sensors exist but I haven't found an easy
                        > > supply yet.
                        > >
                        > > Peter
                        > >
                        >
                      • Peter Balch
                        ... Excellent. I hadn t spotted that chip. When I saw touch sensor I assumed it was on/off. I haven t got my head around the data sheet yet but it seems to
                        Message 11 of 17 , Dec 6, 2012
                        View Source
                        • 0 Attachment
                          > Here is the page for one type ,they do have others
                          > https://www.sparkfun.com/products/7902

                          Excellent. I hadn't spotted that chip. When I saw "touch sensor" I assumed
                          it was on/off.

                          I haven't got my head around the data sheet yet but it seems to imply that
                          maybe one should make a sensor with a capacitance change of less than 1pF
                          over the force range you're interested in.

                          For two plates,
                          Capacitance = permittivity * area / distance.

                          If we assume permittivity = 10pF/m then, for example, plates with an area of
                          1 sq cm separated by 1mm give a capacitance of 1pF. (I think I've got all
                          the zeros right!)

                          Of course, for a force sensor, one would be looking for a _change_ of 1pF
                          and I think the chip allows a "parasitic" capacitance of 40pF on top of that
                          change.

                          > Now to make the sensor you need a 2 tiny copper plates and
                          > some soft insulator material to go between them

                          It's a nice idea to make a home-made sensor. I'd be worried that the foam
                          would degrade pretty quickly. Open-cell foams often can't take many crush
                          cycles (although closed-cell neoprene isn't too bad). I wonder whether a
                          silicone baking sheet between two copper-clad boards would do the job. One
                          of the boards could have the chip on it to keep the sensor leads short and
                          stable. I'm generally measuring a max of a hundred kilos but you may be
                          trying to measure less force.

                          The chips are not cheap but they could well be worth it if the completed
                          sensor is accurate. I think I'll buy some.

                          Peter
                        • jamericanfreddy
                          I was over 5 years ago when a professor of physics that was the part owner of the company i worked for that ,design that sensor for a scale it measure up to
                          Message 12 of 17 , Dec 7, 2012
                          View Source
                          • 0 Attachment
                            I was over 5 years ago when a professor of physics that was the part owner of the company i worked for that ,design that sensor for a scale
                            it measure up to 100 lbs

                            SO be nice to make that sensor,i know we didnt use that chip
                            it might used a 555 timer to get capacitance to voltage

                            I am thinking small a circuit that used in most meters is what we need.

                            --- In SeattleRobotics@yahoogroups.com, "Peter Balch" <peterbalch@...> wrote:
                            >
                            > > Here is the page for one type ,they do have others
                            > > https://www.sparkfun.com/products/7902
                            >
                            > Excellent. I hadn't spotted that chip. When I saw "touch sensor" I assumed
                            > it was on/off.
                            >
                            > I haven't got my head around the data sheet yet but it seems to imply that
                            > maybe one should make a sensor with a capacitance change of less than 1pF
                            > over the force range you're interested in.
                            >
                            > For two plates,
                            > Capacitance = permittivity * area / distance.
                            >
                            > If we assume permittivity = 10pF/m then, for example, plates with an area of
                            > 1 sq cm separated by 1mm give a capacitance of 1pF. (I think I've got all
                            > the zeros right!)
                            >
                            > Of course, for a force sensor, one would be looking for a _change_ of 1pF
                            > and I think the chip allows a "parasitic" capacitance of 40pF on top of that
                            > change.
                            >
                            > > Now to make the sensor you need a 2 tiny copper plates and
                            > > some soft insulator material to go between them
                            >
                            > It's a nice idea to make a home-made sensor. I'd be worried that the foam
                            > would degrade pretty quickly. Open-cell foams often can't take many crush
                            > cycles (although closed-cell neoprene isn't too bad). I wonder whether a
                            > silicone baking sheet between two copper-clad boards would do the job. One
                            > of the boards could have the chip on it to keep the sensor leads short and
                            > stable. I'm generally measuring a max of a hundred kilos but you may be
                            > trying to measure less force.
                            >
                            > The chips are not cheap but they could well be worth it if the completed
                            > sensor is accurate. I think I'll buy some.
                            >
                            > Peter
                            >
                          Your message has been successfully submitted and would be delivered to recipients shortly.