Building - Robot Arm, Help


I want to build a robot like a learning process but I would like the robot to be able to lift a can of coke (330Ml) for my project. Which weighs about 350g. Do you have any ideas or link to what parts I should pick and etc?

Thank you so much in advance for your help!

As cbenson mentioned in the shout you will need to be more specific. Do you want an arm that can reach out and lift it? How long do you want the arm? how far does it need to lift? how fast?
Or maybe you want a rolling robot that rolls over the top of the can and lifts it enough to take it someplace else.
Or maybe something different entirely?
Please give more details so we can help.

Thanks for the reply. My plan is to have a mini fridge where I store coke, and then some how make the door open by some programming or if the robot arm is strong enough? Then I hope I can make the robot arm lift the coke out of the mini fridge and put it on conveyor belt and it will come to me :smiley: . Do you have any ideas? Lenght, and speed doesnt really matter as long as it can do the movement smooth, but it would be nice if he could lift 350g with straight arm.

That is a pretty ambitious project, especially for a first robot. But it is doable. Building an arm from scratch is a large project, even for an experienced builder. An arm that can lift 350g fully extended is actually a pretty large arm for a hobby robot. This project can be relatively straightforward or quite complex, depending on how you approach it.
Here is what I recommend:

First, use a kit for the arm.
As you look through those kits you will see that the ones capable of lifting 350g are not simple or cheap. But if you try to build one yourself you will almost certainly end up spending at least as much.

Second, limit the requirements. What I mean by that is make it as simple as possible. Have the drinks in a dispenser like this:

so they are always in the exact same place. And make sure the arm and conveyor belt are always in the same place. Then it’s a simple matter of moving the arm through the same set of motions every time. If the robot has to “find” the drink, it is a MUCH more difficult task. Probably using a camera for vision, image recognition, and all sorts of difficulties.

Third, separte the parts. Have the mechanics that opens the door separate from the arm and the conveyor. You can build one part at a time, adding the others as each one is complete. Besides, an arm strong enough to open the door is going to be big, heavy, and expensive.

Take a look through the kits in the link above. I think you should find something suitable in the $150 to $250 range.

Big results require big ambitions :wink: Thank you again! What robot arm / kit would you personally recommend? I dont have must experience though but I am willing to learn :slight_smile: Would also be nice to make him do the task with a simple click of a button from my computer. If thats possible?

That’s a good attitude!
I think it’s best if I let someone from robotshop make a recommdation. They will be much more familiar with the products. All I can do is read teh specs. @cbenson is a good person to ask if no one jumps in here.

The task of opening the fridge should not be done by the arm itself - there’s quite a seal around the door which makes it quite hard to open. Use a linear actuator mounted to the top to open it:
You might think about how you’d mount it to determine the stroke length needed. These are fairly easy to control if you just want fully open and fully closed.

Regarding the arm, a small industrial would be best, but way beyond budget. The next best thing would be:
It can list ~500g at full extension, but does not include a gripper, power supply or controller.
Consider a computer power supply which can provide 5V at up to 10A, and the SSC-32U:

However, almost any servo controller will work well.

Gripper (add rubber to prevent the can from slipping):

You’ll need some servo extension cables.

Don’t try to do all at once - work on each section.

If I say that I am on a budget on the arm. Is it possible to decrease the price on the arm, ca 200 - 300 ? Does the parts ever come on sale?

The arm is the expensive part - inexpensive arms simply don’t have the reach or the torque. The parts do occasionally come on sale, but cann’t say when / why. A $300 arm would be something like:

This cannot provide the reach nor the payload you need.

You might want to consider a more “mechanical” approach with only one or two degrees of freedom in the arm. Think of how it would be done in a “manufacturing” plant - they would not invest in a complex robotic arm but take a look at the minimum motion required to move the can from an initial position to a final position. It might be as easy as a “release” mechanism like a vending machine.

I see might buy the 600$ one, if it someday comes for sale. Im not in a rush. Btw do you have any ideas how to make a convejor belt, or buy some cheap once. I need it to be thin and long

Cheapest way would not to have it motorized - use a “gravity chute”. Most of the production parts used for conveyors in hobby robotics are not meant to be very long and might break. The Lynxmotion one is robust, but would not be cheap if you need a long length:

VEX conveyor meant for relatively short distances:
If you’re ready to put in the effort, many people have adapted bike chain to tracks (which are essentially conveyors).

Is there something in between those arms price wise?

There are iterations of the Lynxmotion AL5, but in the <$650 range, there’s nothing aside from the M100RAK that would be suitable.
Take a look:

Im not sure what parts I need to make the robot arm work even if I decide to buy it.

Can you sketch the motion of the can and the fridge door up to where you want to grab it? be sure to include dimensions / distances.
For simplicity, have the can start on the top shelf of the fridge, and end up lower.
Based on that sketch, you’ll be able to see better just how many degrees of freedom your system needs.
Consider taking a few minutes to learn Sketchup (import a fridge, chair, can etc.).
You likely do not need a complete robotic arm.

To make it easy I was thinking of putting one of these in the fridge so the can will always be on the same spot.
Idk I can adjust to make it easy. I will be getting a 120cm long desk to have my tech work 'robot arm". The desk is connected to another desk in an L shape which is 200cm. Which is where I am going to have my desktop. I thougt it would be cool to have the robot arm serve me a can of coke when I feel like it. Get the can from the fridge and put it on the conveyor. Then later try to make it do more. Btw what benfits are there if I connect it to a rasberry pi?

It would be cool to use a robot arm, but as you’re seeing, there is a price to pay. A Raspberry Pi is a mini single board computer, which is really overkill for this application unless you want to add a camera, WiFi etc. You’ll need to learn to program in Python.

A post was merged into an existing topic: Robot Arm - 1 Pound Payload