Universal gripper part 2
I have looked up more tutorials for the universal gripper and it seems that Sparkfun had demoed one successfully with a funnel in place of a showerhead.
I purchased a few funnels from the local hardware store ($3.75 for a set), and proceeded to search for the final two articles: coffee grinds and cotton ball.
Here’s a picture of the finished device.
I will soon upload a video of it actually working right here. For now enjoy watching my naive first attempts, and my first adjustment to the design: filling the volume of the funnel with cotton spacers:
Here’s a pic of the main materials:
- coffee grinds
- tubing (get diameter)
- cotton swab
- 12 volt Vacuum pump
- 12 Volt AC-DC adapter
The cost breakdown is as follows
- pack of 36 12in diameter balloons —
- coffee grinds
- cotton swab pack
- 4 feet of tubing — 5 dollars about (get specific quote from home depot)
- 12 Volt Vacuum pump (check sparkfun and amazon.com)
- 12 volt AC- DC adapter (check amazon.com)
- funnel (check amazon.com)
I am estimating that it would take around 40 dollars to create, but if you split the cost of the balloons and coffee grinds with friends, then it would basically just be the cost of the 12 V vacuum pump + ac-dc adapter (you can also just use an extra one from home).
Below are picture based instructions of how I built it:
I did a few trials of different types of heads: grocery bag with grinds, ziplok with grinds, and balloon
Balloon worked the best for two reasons:
- the material can conform over the other materials, and
- the rubber formed a high-friction grip onto the material it was conforming to
For next time:
I would try to use different coffee grinds, and see if finer grains would provide a more detailed grip
I will also attempt to use a 36 diameter balloon, to see if it would conform better, having more slack.
I like the general hook-up at the moment; however, I will try to find out if I can add:
- a vibration motor to help the grains settle back to normal
- a valve that can switch the direction of the air flow (to inflate the balloon)
- this could help the grains resettle and
- it could add the ability to throw/release objects.
- Affix the funnel to:
- a robotic arm with a rotating base
- a regular joint for up and down motion
- the ability to rotate the funnel to turn knobs.
Medium term, we could connect it to an arduino to automate it.
We could make the arduino record voltage data inside a time series for a specified amount of time, and have the robotic arm mimic the movement (once calibrated with the rotary potentiometers).
Long-term, we could fix the hand to a mobile base (with geared wheels)