2.12 gripper rapid prototyping [2.12 Design Notebook]
Hereee we go, going to make a granular jamming gripper just for testing, nothing fancy, honestly the simplest gripper we can do.
Grabbed some balloons, grabbed some coffee grounds, tape, munchies, etc. We don't have a funnel like in the make video so we're going to improvise.
making the gripper:
So step one is to fill the balloon with granules of coffee. And I started by trying many strategies for this. All of the below strategies failed because it didn't open the balloon enough and I was spilling too much coffee everywhere.
The last strategy with the makeshift funnel ended up like this:
Yes, oops, okay so I went out tried to find an actual funnel and apparently the room across the hall had one and that worked so so so much better to fill the coffee. The strategy is fill the funnel with coffee and allow the grains to enter the balloon by stretching it. Then push the grains further into the balloon opening the mouth of the balloon up allowing more grains to fall in and then repeat.
Finally we end up with this onion looking nonsense. He's pretty sizable (the hand is for size reference, it's about the size of a large tomato or a mango. Now we need to add some things, a tube to keep the mouth open that we can attach the vacuum pump to, and a filter of some sort so I don't keep getting mouth-fulls of raw coffee.
That's basically it, we just need to hook up a vacuum to it and grab some things to test with!
testing the gripper: quick and dirty
So now it's time to test the gripper, and here's the strategy.
(1) To get the grains loose, we will blow air into the balloon while descending on the object and allow the gripper to settle around the object.
(2) We will then remove the air blowing and switch to a shop-vac which can produce the suction force required to suck the air out and see if the thing picks it up.
(3) If it does we will remove the suction and blow on it to see if it ejects the part.
We added extra tape to the top of this gripper so the vacuum has a surface to rest against, I'm afraid if we leave the balloon exposed the vacuum will suck it in and rip it.
We'd ideally be trying to pick up water bottles that are crushed or in different configurations/etc. We only have one of these and its in the normal configuration but we will try this.
TEST ONE: picking up Vitamin Water (not)
TEST TWO: picking up a marker (not)
TEST THREE: picking up an open-mouth bottle (kinda)
TRYING A NEW STRATEGY:
The above tests didn't work so well, that could be because the granules aren't loose enough or the gripper doesn't have enough suction. My guess is on the first one, which means that the granules need to be larger, and that we will need to pump air INTO the system to pick up objects before we draw the vacuum. I'm also going to try to see if its the vacuum seal itself hence:
TEST FOUR: random object on the table
This doesn't seem to be going so well. Maybe we really need to make a better/proper test gripper like the hackaday one and this maker one just isn't good enough.
So it's also possible that it isn't being constrained enough from the sides. Let's try sticking some PVC around it. So we're going to try 3D printing this: https://hackaday.io/project/158551/instructions
And let's see if this thing works with a much better design. We're going to change the following things:
(1) I THINK we overfilled the balloon - so we're going to try again with much less filling next time.
(2) I don't think it's supported enough, we need a better seal, we need a better vacuum etc.
Okay so we're going to come back to this prototype on Sunday or tomorrow night.