Gears for a arm joint confusion 2

[GlennW125]

Hi There, I am a electrical engineer trying to design a robot arm for automation that does not contain a motor to which I am looking into gears. I have been scratching my head on terminology and what gears to use for this scenario. A arm that has the first and second part 15in X 3in X .75in , the arms are going to be made out of aluminum so not so heavy. the end of the arm is going to have a max weight of 2 lbs. What I have been trying to figure out is accuracy and how to achieve this accuracy. I want there to be a motor to turn at the base of the arm and its going to drive a pulley to the arm joint, probably a GT2 belt as there accuracy is pretty good. I want the joint itself to receive a 2 degree turn and output a to the attaching arm 4 degrees. It needs to be along 1 axis, the X, and not move a hair up or down. Which is why accuracy is key. I have attached a picture to better describe what problem I am having. Anyone in mechanical engineering know what I could use for that joint? Maybe a gearbox or whatever linear box I could purchase to put on that joint?

 

[MikeHalloran]

I hope you will not be insulted by this, but you need to get help from a real mechanical engineer; you are in way over your head.



Mike Halloran
Pembroke Pines, FL, USA

 

[GlennW125]

Kinda why I posted on the eng-tips forums...to get help from idk a real mechanical engineer?

 

[3DDave]

Interesting. What you seem to be looking for is a straight line generating linkage, of which there are dozens of versions.

The question to ask is how long would it take an engineer to design this? Based on the type of questions and constraints, 40-80 hours to document a supportable concept. There just isn't enough typing time to handle it with the forum, so you need to find a mechanical engineer who can work directly with you.

I've diagrammed your request, which is 3 questions and 10 constraints, though some of the constraints are not good ones.
For example, 'c', confuses density with mass. It is often the case the strength to weight ratio of steel is so much better than aluminum, so that a steel structure is lighter. 'i' is really terrible as it suggests the only solution is an infinitely rigid item with no backlash in the joints. 'j' is good, but the assumed mechanism seems impractical.

**************
Hi There, I am a electrical engineer trying to design:
a) a robot arm for automation that does not contain a motor to which I am looking into gears.
I have been scratching my head on terminology and what gears to use for this scenario.
b) A arm that has the first and second part 15in X 3in X .75in ,
c) the arms are going to be made out of aluminum so not so heavy.
d) the end of the arm is going to have a max weight of 2 lbs.
Question 1: What I have been trying to figure out is accuracy and how to achieve this accuracy.
e) I want there to be a motor to turn at the base of the arm and its going to drive a pulley to the arm joint,
f) probably a GT2 belt as there accuracy is pretty good.
g) I want the joint itself to receive a 2 degree turn and output a to the attaching arm 4 degrees.
h) It needs to be along 1 axis, the X,
i) and not move a hair up or down.
Which is why accuracy is key.
j) I have attached a picture to better describe what problem I am having.


Question 2: Anyone in mechanical engineering know what I could use for that joint?
Question 3: Maybe a gearbox or whatever linear box I could purchase to put on that joint?

 

[GlennW125]

Thank you so much 3DDave. I love the detail you put into your answer. I really do appreciate it. Ill get into contracting a mechanical engineer to see what options for price I have. I really wish constraint (i) was not terrible, as I was hoping there was a way for it to be a certain height above the assembly. Ill look into those Straight line generating linkage and see what comes up, but again thanks for your amazing input.

 

[tbuelna]

GlennW125-

Gear meshes always have some amount of backlash (clearance between the teeth), and belt drives have similar issues when the direction of load on them is reversed, since the belt is only capable of transferring force in tension.

If you are going to use a set of bevel gears for positioning your robot's arm, you should remember that when the arm segments rotate about the gear axes at the elbow, you may need to account for some amount of differential rotation between the input/output gears.