next topic »

[on:]

Hi gang,

What methods exist to make sure that a gripper gets a good enough hold on the item its gripping without crushing the sh!t out of it?

TIA,

Jim

[Reply #1 on:]

feedback is the word:
use a pressure sensor inside the gripper, some like - https://www.sparkfun.com/products/8685 -

You could also use a spring that counteracts a servos

[Reply #2 on:]

Strain gauges are also used as force feedback sensors.

Lefty

[Reply #3 on:]

Thanks Gents

[Reply #4 on:]

These are a lot cheaper,

http://www.pololu.com/catalog/product/1696/specs

[Reply #5 on:]

A lot depends on what type of gripper you plan to use. For good grip a thin layer or grippy type foam can be put on the inside of the grip surfaces. A limit switch/contact type of grip force detector might be used, or maybe a rubberband that actuates the gripper that would stretch preventing servo damage. Lots of possibilitys.

[Reply #6 on:]

I think I have run across some code that involved calculating when you get close to the stall rate of a motor. Don't ask me how to do it, but I have seen it somewhere in these forums, and it seems that it wouldn't involve any further purchases.

[Reply #7 on:]

I think I have run across some code that involved calculating when you get close to the stall rate of a motor. Don't ask me how to do it, but I have seen it somewhere in these forums, and it seems that it wouldn't involve any further purchases.

But it will most likely make my brain hurt....

For now I'm going with a small electromagnet as end effector.

[Reply #8 on:]

For now I'm going with a small electromagnet as end effector.

I should think that measuring motor stall current would be past the point of putting
too much force on whatever is being picked up.

Could you explain your end-effector idea a little more?

Also, with those resistive force sensors, I should think you could mount them between
a thin piece of relatively hard rubber and the metal frame, to pickup transmitted forces.
Eg, glue the rubber down at the 2 endpoints, and slide the force sensor into the slot
between the rubber and metal.

[Reply #9 on:]

Could you explain your end-effector idea a little more?

Like those huge crane thingies in scrapyards, which can lift whole cars... mine is really just a proof-of-concept that a robot arm can actually do something, in this case say pick up a bunch of paperclips when the joystick button is pressed.

[Reply #10 on:]

Paper clips, wow, I would think Lego blocks might be a lot easier. Or even eggs.

[Reply #11 on:]

Paper clips, wow, I would think Lego blocks might be a lot easier. Or even eggs.

Well no.... last time I checked, a magnet wouldn't lift Lego or eggs 

[Reply #12 on:]

Maybe your chickens have an iron deficiency.

Any current sensing circuit to the gripper's motor will give you an indication of how much force is being applied. It will still work with an RC servo as well. Murphyslaww, perhaps you were thinking of the code that would tell you the servo's position based on how much current was being used by the servo (a moving or pushing servo uses more power than a stationary one).

[Reply #13 on:]

Not sure what kind of servos you are using, but the simplest method is to use a spring to limit the force of the servo. RC "Servo Savers" can be had for a few dollars.

http://www.google.ca/imgres?um=1&hl=en&sa=N&tbo=d&biw=1920&bih=1075&tbm=isch&tbnid=WPi9RIeIDeBu1M:&imgrefurl=http://www.horizonhobby.com/products/servo-saver-hitec-hs-55-mini-t-LOSB1226&docid=5u7P6cSCD1Hj1M&imgurl=http://s7d5.scene7.com/is/image/horizonhobby/LOSB1226_a0%253F%2524pdpLand%2524&w=900&h=500&ei=fczrUMbTOMaKjAKEpIGYBQ&zoom=1&iact=hc&vpx=1420&vpy=443&dur=821&hovh=167&hovw=301&tx=177&ty=78&sig=112198703044298569596&page=1&tbnh=138&tbnw=236&start=0&ndsp=58&ved=1t:429,r:27,s:0,i:175

[Reply #14 on:]

Servo savers typically have a limited amount of travel, so if the servo keeps trying to
turn, the saver will bottom out and you'll still get too much force.