Help me in making a self-made robot arm

[poorak]

i need a simple pickup arm that can pick up a cube of 10*10*10 mm and place it on a table of some height ( less than 100cm).
The arm should be self-made & no kits are allowed.

 

[Hero999]

Whilst we enjoy helping people; in order to help you we need to know how far you've progressed. Please post what you've already done, including any ideas you've had. If you need help that your lecturer should have provided, please tell us (and your lecturer) what you found was lacking. It is not right if you manage to attain a high grade from the help we have provided if your lecturer is negligent; it will make them appear to be better than they really are and put you at an unfair advantage to your classmates.

 

[Hank Fletcher]

The arm should be self-made...

You mean like this? Seriously, though, put some more effort into your post, if at least by providing more detail about what you need, and you'll be more inclined to get the answers you're looking for. We're not a custom robot-hand building business, but if you offer how much you're willing to pay, you might find takers. If you're looking for free advice, you get what you give and then some, so it's still a good deal.

Will it be easy to crush the cube (e.g. is it made of sugar?)?
Besides grabbing, raising, rotating, releasing, and lowering, how else will the hand have to move?
Will the height be a fixed distance, or variable? If the latter, how will you detect the height?
Define "self-made," i.e. to what extent can you use pre-made parts, such as geared motors, or even scrap toys?

 

[dknguyen]

Well...if the table is 1m tall, then the arm probably has to be longer and at that length your cost just sky rocketed (like into the several hundreds of dollars at least) because the lever arms involved are likely too long for any RC servo to lift the weight of the material of the arm, let alone a load. Seems like a crane would be the best way to go since you can get around the size of the crane by using small pulleys and ropes as there are no rotating "elbow" or "shoulder" type joints.

Even if you used the most powerful RC servos available (~$120USD each), at 1m they would still probably struggle to lift the weight of an arm of that length, especially with a load. Not to mention it would be already expensive anyways. Any arm even vaguely resembling a human arm needs at least 3 joints so you're already at $260 just for motors- that's assuming each joint only has one motor which is probably not enough due to the length of the arm.

So my advice is this: I hope you don't have something like a human arm in mind. I'd go tower-crane-type method- a tower crane only needs a couple of motors and the lever arms aren't a problem since you can just use small pulley wheels for the cables. There's a reason we use construction cranes and not giant polar-revolute coordinate arms (human arms). Cost, simplicity, and load capacity among other things.

A simple 90 degree bar with one motor to rotate the base radially and another motor to drive the cable and pulley wheels (to lower and raise the load) should be sufficient. If you need to be able to precisely place it, then you need a 3rd motor to move the location of the load along the horizontal boom. You might need one more small motor at the end of the cable on the crane to grasp the load, but you might not depending on how you decide to grasp the load (vacuum perhaps?)

Study this image:
https://en.wikipedia.org/wiki/Image:Tower.crane.bristol.arp.jpg

You could use a more "traditional" crane like this:
https://en.wikipedia.org/wiki/Image:Trier_Germany_Alter_Krahnen.jpg
but then you get the problem of the lever arm at the base (similar to the shoulder joints on a human arm)

 

[Hero999]

I hate threads like this, when people seem to make more effort responding than the original poster.

 

[Ambient]

Sounds like a school project or some contest that is supposed to be the OP's work.

 

[poorak]

Sorry for replying a little late. i chose wood for this purpose. I have now designed the arm, it's all wood now. The length of wooden towers where the arm is screwed is about 65 cm. The length of the arm where grabber is placed is about 60cm. Lets visualise it, Their are two wooden pieces of length 65cm which are glued to base at 90 degree. at the top of these pieces their is a wooden hub which is holding the tower at sides and at the middle of this hub i have 3 arm size square holes. Via this hub the arm motion is free.
i have made a moving base of 30*25 (cm). I have placed a motor @3v at each corner of base with 4 tyre each for giving full support to the bulky machine.
Now the problem is at large related to pulley these hot woods. For this as of now i have buyed a printer motor @24v. I have built a remote. Another thing that is watchworthy, is the 24V maximum power used in the machine.
For a constant power supply i have used a step down transformer and made a full wave rectifier circuit to get a constant supply of voltage.
my machine survives on two motions namely lifting of arm up vis-a-vis down and motion of the wooden grabber i.e. Open and close.
The problem area that i wish to consult , how can i make a pulley that lifts the arm, using printer motor ?
Another question of same nature is how can i use a simple DC motor @ 3v to implement a open and close.
Note: The opening and closing of the jaw in this machine is entirely based on pulling of middle arm piece( i have three side to side placed wooden pieces that make the arm)

So i need some rough idea how i am going to implement the pulley action and the jaw mechanism via motors.

Please reply soon.... thanks in advance......

 

[Hero999]

Could you please post a rough sketch of what you're intending to do?

24V maximum power used in the machine

That doesn't make any sense, power is measured in Watts which is volts times amps so saying the maximum power used is 24V is totally meaningless.

 

[poorak]

It's DC

it's DC motor and it works in volts. not meaning less but has a fact to tell.

i can give you a sketch.
Next post i will proide it as a link. sorry for delay.

 

[xheavenlyx]

Poorak, by your first post I am assuming you aren't experienced, I too am not much but I can tell you for sure if a person with more than 5000 posts is telling you something, then listen!

I myself was on this project a while back and the most important thing you have to keep in mind is Power and Weight constraints.

Your motor works in volts, yes, is a fact. But the power it consumes is not the same as voltage. And multiple motors will have same voltage, ie 24, but total power consumption will be much higher.

Actually Hero999, if you would be kind enough to please explain how is power mesured in motors, mainly the amps (used later to calc V * A)? do we connect an ammeter in ser while motor is running at 24 volts OR is it also ok to measure a rough resistance of armature winding by a ohmeter and devide it by 24V in this case? Till today I have never really figured out how to use all the theory about motors we learnt in class, and apply it to small DC motors!

 

[Hero999]

Power is measured in motors the same as it is in light bulbs and resistors.

Measuring the armature is simply not good enough as it'll only give you the stall current.

Measure the current the same as you would in any device by connecting an ameter in series with it. If it's a large motor and draws >10A then a multimeter won't do so use a clamp meter.

 

[poorak]

Sorry boss...!!
I have used a 3 Amp step down transformer that provides 12 and 24 volts parallely. So this might solve the watt problem.
I have been a little out of touch, so sorry for that.
The distance for which the motor must pull is about 60cm.
Actually the machine is still in workshop for integration of parts. Once completed i will surely send a photo of the same.

Regards

 

[xheavenlyx]

Oh yeah, the armature reading (resistance & / by V ) will just give the current at NOLOAD. Which is the stall current rite?

So, for proper tests is it advisable to measure the current by ser ammeter (low power) and varying the load and noting down the current and hence power consumed at each load level. By this we can check for different voltages (12,24 etc) and get the best efficiency?

 

[Hero999]

Oh yeah, the armature reading (resistance & / by V ) will just give the current at NOLOAD. Which is the stall current rite?

Yes, it's the stall current but it's opposite to the no load current, it's the maximum current when it's overloaded.

So, for proper tests is it advisable to measure the current by ser ammeter (low power) and varying the load and noting down the current and hence power consumed at each load level. By this we can check for different voltages (12,24 etc) and get the best efficiency?

Just measure the current when the load that you intend to use is connected.