How Effective Is Chebyshev's Linkage for Precision in Gripper Mechanisms?

[Femme_physics]

So we need to make a gripper in Solidworks only for this object

http://img819.imageshack.us/img819/2285/thepart.jpg

Our current goal (as per our professor) is to focus on our gripper mechanism.

My classmate suggested the following for the gripper final grip mechanism

Chebyshev's linkage - alternative form
While the picture here looks quite different from Chebyshev's linkage above, these two configurations actually generate the same curve. It is curious that a particular curve may be generated by more than one linkage. For a clue why this might work, have a play with the GeoGebra applet below, in which the two linkages are superimposed.

http://img13.imageshack.us/img13/5577/linkagep.jpg

The idea being that at the end of this straight movement that closes the gripper 2 jaw parallel effectors attached at the tip of this mechanical mechanism grab the object. We thought it's best for both sides of the gripper to close in a straight motion as this would increase accuracy, as opposed to a Two Jaw Cam Actuated Rotary Gripper...which mean they will be closing in an angel and decrease accuracy.

http://img834.imageshack.us/img834/3716/rotarygripper.jpg

What do you think of the idea so far?

 

[I like Serena]

Well, how does it look... sexy! :!)

As for the idea of a linear gripper... do you have a drawing?

 

[Femme_physics]

Better yet, I have a video! :)

 

[I like Serena]

Nice video. ;)

So are you finished already?
Or is there more to come?

 

[Femme_physics]

No, it's just a suggestion. I want to see if it can be improved, or if it's a good idea overall. We're still beginners on the issue, so every feedback counts. Do you have a feedback? :)

 

[I like Serena]

Should your gripper operate in 3 dimensions?
How would your gripper be fixed in a frame?

Did you already google for grippers? Trying to find a picture of a gripper that matches your object?
And perhaps one that will look sexy in SolidWorks?

Personally, I thinks it's more important that it looks sexy than that it is a functionally correct gripper.

 

[I like Serena]

Btw, is there anything I can get you for your birthday?

 

[Femme_physics]

Should your gripper operate in 3 dimensions?
How would your gripper be fixed in a frame?

Did you already google for grippers? Trying to find a picture of a gripper that matches your object?
And perhaps one that will look sexy in SolidWorks?

Personally, I thinks it's more important that it looks sexy than that it is a functionally correct gripper.

LOL, well the current design is pretty sexy

Yes, it's 3 dimension. But, I guess you can just imagine the 2D construct with volume, or width.

Yes, I googled for gripper, mostly we realize that the 2 jaw parallel gripper is good since it uses less friction than flat friction grip, therefor less energy. My issue is the gripping mechanism, not the actual gripper design. It looks ok, but I'm just a starter so can use the feedback :)

 

[Femme_physics]

Btw, is there anything I can get you for your birthday?

Aww yea what u brought last time just for uppers ;)

j/k! Your forum help is the best gift I can ask for!

 

[I like Serena]

Aww yea what u brought last time just for uppers ;)

j/k! Your forum help is the best gift I can ask for!

Well, the forum help is for free, so that does not count.

So you'd like something to match cybergrrl?
Well, that alone already gives you +25 intelligence points, let alone the charisma points.
I'm just wondering what more uppers I can get you...?

 

[OldEngr63]

Just a few comments about the whole thing.

1. The problem statement apparently does not specify the direction that gravity wills act on the combined gripper/object assembly, not does it specify the actions that the assembly must be able to make. This makes it difficult to determine just how the gripper must hold the object and how firmly the grasp must be. For example, can it simply scoop the object up in a shovel and dump it out, or must it pick it up in one orientation, carry it, and reorient it before placing it at the final position. How fast does it have to function; will dynamic loads be a consideration?

2. It appears that you have not made use of any of the features of the object to help secure the hold of the gripper on the object; shaped jaws could certainly help, but the closing/opening motion must not allow for a jam.

3. Your gripper design looks really bulky. Often the gripper needs to insert a part where there is not a lot of room for the gripper. You might want to look at ways to stream line the design. (Make your two operating wheels coaxial, so that they appear one over the other in the side view, for starters. I think you might think you might want to look at other mechanisms as well.)

4. In your second figure, the on labelled Fig 1, what appears drawn as a spring is labelled "cam." I cannot see it being anything other than a spring.

5. Remember that done properly, the grip/release motion is really very small. What you have drawn generates a very large motion which is easy to see, but far too big to be serviceable. Think about the sizes that are really needed, and the motions, and corresponding cycle times that could be achieved with them.

 

[Femme_physics]

1. The problem statement apparently does not specify the direction that gravity wills act on the combined gripper/object assembly, not does it specify the actions that the assembly must be able to make. This makes it difficult to determine just how the gripper must hold the object and how firmly the grasp must be. For example, can it simply scoop the object up in a shovel and dump it out, or must it pick it up in one orientation, carry it, and reorient it before placing it at the final position. How fast does it have to function; will dynamic loads be a consideration?

It has to work in numerous such conditions and be able to adapt.

2. It appears that you have not made use of any of the features of the object to help secure the hold of the gripper on the object; shaped jaws could certainly help, but the closing/opening motion must not allow for a jam.

3. Your gripper design looks really bulky. Often the gripper needs to insert a part where there is not a lot of room for the gripper. You might want to look at ways to stream line the design. (Make your two operating wheels coaxial, so that they appear one over the other in the side view, for starters. I think you might think you might want to look at other mechanisms as well.)

I agree, it's awfully bulky. Although, in reality, it would only open up 30% larger than the object it needs to catch (from each side), in my solidworks video it opens up really wide. It was an initial sketch idea.

4. In your second figure, the on labelled Fig 1, what appears drawn as a spring is labelled "cam." I cannot see it being anything other than a spring.

I don't see a spring in our design. You might be confusing the "Cam" from the "CamStudio" software I used to record.

5. Remember that done properly, the grip/release motion is really very small. What you have drawn generates a very large motion which is easy to see, but far too big to be serviceable. Think about the sizes that are really needed, and the motions, and corresponding cycle times that could be achieved with them.

As I said my video was overdoing the grip. We did think of a less clumsy mechanism, which I'd like to present:

http://www.youtube.com/watch?v=zKtcwkMuL3k&feature=youtu.be

Any thoughts?

 

[berko1]

Both designs are based on rotary movement converted to linear motion. You may think of using a linear actuator (Pneumatic or Hydraulic) to drive your design in some sort of scissor motion.. Just a thought..

 

[OldEngr63]

It has to work in numerous such conditions and be able to adapt.

This is really not a design spec, and there is no way to say whether a particular design meets it or not. This is especially true with respect to the dynamic loading/cycle time considerations.

The use of a rack and pinion, with the rack perpendicular to the main system axis, is going to make it very difficult to ever achieve a compact design. Why not a rack parallel to the axis, driving a wedge to clamp onto the object (much like a Jacobs drill chuck)?

 

[Femme_physics]

Both designs are based on rotary movement converted to linear motion. You may think of using a linear actuator (Pneumatic or Hydraulic) to drive your design in some sort of scissor motion.. Just a thought..

The idea is indeed to use a linear actuator in our design to drive both gears :) We discounted scsissor motion because direct linear motion for grabbing seems more accurate, don't you think?


Our part weighs 7.3 gr. Not sure about using Pneumatic or Hydraulic. We need to grab the part on the area marked in red, so we figured an electric one would best serve this purpose. No?

This is really not a design spec, and there is no way to say whether a particular design meets it or not. This is especially true with respect to the dynamic loading/cycle time considerations.

Well, it has to be designed to grab that SPECIFIC part, in that specific area marked in red. The part weights 7.3 gr and made out of AISI steel 1040. It has to be able to pick it and place it elsewhere.


I wasn't sure about your comment at clause 1, I'll try reanswering it:

1. The problem statement apparently does not specify the direction that gravity wills act on the combined gripper/object assembly

Doesn't gravity always act down?

not does it specify the actions that the assembly must be able to make. This makes it difficult to determine just how the gripper must hold the object and how firmly the grasp must be. For example, can it simply scoop the object up in a shovel and dump it out, or must it pick it up in one orientation, carry it, and reorient it before placing it at the final position. How fast does it have to function; will dynamic loads be a consideration?

Oh, I see what you mean.Yes, it's for pick and transfer.

The use of a rack and pinion, with the rack perpendicular to the main system axis, is going to make it very difficult to ever achieve a compact design. Why not a rack parallel to the axis, driving a wedge to clamp onto the object (much like a Jacobs drill chuck)?

How can you grab the object with a wedge exactly?

 

[I like Serena]

How about a sliding gripper?
Something like this:

 

[Femme_physics]

Well, that's essentially what we've made so far in the last solidworks animation (imagine that at the end there are 2 jaw grippers like in your pic. The difference between my idea and your pic is:

A) Mine drives the end-effectors directly. Suppose I could mount on the rails grippers like yours.

http://img638.imageshack.us/img638/8798/mounting.jpg B) Yours doesn't show internal design

 

[OldEngr63]

Do you have a drill press, or an electric drill with a keyed chuck? This is called a Jacobs chuck after the original patent holder. If you look inside, you will see that the "teeth" that hold the drill bit are in fact three sliding wedges that come out (toward the user) as they close. When you turn the key to tighten the chuck, you are simply tightening the force driving the wedge into place.

I don't think I had previously picked up the part about having to pick up on the red surface if that had been mentioned. That is quite significant.

As to the direction of gravity, yes, it is always down. But the part, as shown in the drawing looks like the symmetry axis is horizontal. The point is, will it remain in that orientation, or will it be necessary to rotate it into a new orientation, possibly orienting it into several orientations before finally depositing it in its new location. In that regard, gravity then acts in a variety of directions with respect to the gripper. If all of this must happen quickly, then there are rotational inertial loads that can be significant. It is generally considered bad form to drop the part!

Do you know how it is to be oriented in the final position? Which end is finally accessible? This makes a lot of difference.

 

[Femme_physics]

As to the direction of gravity, yes, it is always down. But the part, as shown in the drawing looks like the symmetry axis is horizontal. The point is, will it remain in that orientation, or will it be necessary to rotate it into a new orientation, possibly orienting it into several orientations before finally depositing it in its new location. In that regard, gravity then acts in a variety of directions with respect to the gripper. If all of this must happen quickly, then there are rotational inertial loads that can be significant. It is generally considered bad form to drop the part!

Basically our teacher told us to not limit the consumer, if the consumer want to put the part from a high region to low, or vice versa, it should all be possible. The idea though is basic pick and place movement.

Do you know how it is to be oriented in the final position? Which end is finally accessible? This makes a lot of difference.

No, we don't. Again, it's about not limiting the consumers.

Do you have a drill press, or an electric drill with a keyed chuck? This is called a Jacobs chuck after the original patent holder. If you look inside, you will see that the "teeth" that hold the drill bit are in fact three sliding wedges that come out (toward the user) as they close. When you turn the key to tighten the chuck, you are simply tightening the force driving the wedge into place.

Well, I understand what a drill press, and how it works. But how will it help me grab the part at the section I need, that I can't relate.

I don't think I had previously picked up the part about having to pick up on the red surface if that had been mentioned. That is quite significant.

You're right,I shamefully forgot to mention it. Mind you the first time I made this thread I got an error that deleted my entire post (where I mentioned it), so I'd chalk that to impatience. My bad.

 

[OldEngr63]

@ Femme_Physics: You said, "Our part weighs 7.3 gr. Not sure about using Pneumatic or Hydraulic. We need to grab the part on the area marked in red, so we figured an electric one would best serve this purpose. No?"

Now as I go back and look at the original spec sheet, I see Mass = 7286.12 grams. Has there been a factor 1000 lost on the mass of the object? This will make significant difference with regard to holding forces required.

 

[OldEngr63]

Well, I understand what a drill press, and how it works. But how will it help me grab the part at the section I need, that I can't relate.

You have not really looked closely at the drill chuck. What the drill chuck does is precisely what you need to do here - grab hold of a cylindrical surface, move it, and then release it. To understand, you have to look closely at how the inside of the chuck works.

 

[OldEngr63]

"Basically our teacher told us to not limit the consumer, ..."

Sounds like your teacher must of been a great sales engineer, the sort that wants to be all things to all men. The truth is, it is not possible to design machines (of any category) with unlimited capabilities. They always turn out to be finite in every aspect. (If you tell him what I said, he will probably hit you right between the eyes and swear at you! Faculty get very sensitive about such things!)

He gave you all of those things, mass, mass moments of inertia in all three axes, cm location, etc. more or less implying that he expects you to use them to calculate reaction forces and moments on your mechanism, but then he gives you nothing at all about the motions that the mechanism will experience, not even max accelerations, max angular accelerations, nothing. Talk about a red herring! Shame on him.

 

[Femme_physics]

@ Femme_Physics: You said, "Our part weighs 7.3 gr. Not sure about using Pneumatic or Hydraulic. We need to grab the part on the area marked in red, so we figured an electric one would best serve this purpose. No?"

Now as I go back and look at the original spec sheet, I see Mass = 7286.12 grams. Has there been a factor 1000 lost on the mass of the object? This will make significant difference with regard to holding forces required.

You're right, it's 7.3 kg. Sorry, excited. :)

You have not really looked closely at the drill chuck. What the drill chuck does is precisely what you need to do here - grab hold of a cylindrical surface, move it, and then release it. To understand, you have to look closely at how the inside of the chuck works.

You're right, after a closer look...the drill chuck looks much better. Now I just have to understand the mechanism the drives the entire drill chuck to make it work. I'll consult with my project classmate to see what he thinks, also.

Sounds like your teacher must of been a great sales engineer, the sort that wants to be all things to all men. The truth is, it is not possible to design machines (of any category) with unlimited capabilities. They always turn out to be finite in every aspect. (If you tell him what I said, he will probably hit you right between the eyes and swear at you! Faculty get very sensitive about such things!)

Actually my professor is very nice old Ukrainian man, I doubt he's ever been involved to deeply in sales. I'm pretty sure he's aware of the finite stuff--perhaps I do a poor job representing him. If in our design we have no choice but create limitations, then we do it. His main concern is really that we document everything :)

He gave you all of those things, mass, mass moments of inertia in all three axes, cm location, etc. more or less implying that he expects you to use them to calculate reaction forces and moments on your mechanism, but then he gives you nothing at all about the motions that the mechanism will experience, not even max accelerations, max angular accelerations, nothing. Talk about a red herring! Shame on him.

Well, to my understanding yes, we need to calculate a lot of stuff, shows advantage and disadvantage in certain scenarios, basically do a lot of documentations.

As far as max accelerations, max angular accelerations...does he really need to define us that stuff? Perhaps he expects us to choose it ourselves within reasonable margins. It appears to me like quite an open-ended project, indeed not many conditions have been predefined, but we do have to account to what we choose in calculations, diagrams, and etc etc...

 

[OldEngr63]

"As far as max accelerations, max angular accelerations...does he really need to define us that stuff?"

That's a pretty good question. I don't see how you are going to determine the required gripper forces without some estimates on these things. In the end, it may be up to you to assign some values for these, but you might want to try to have some basis for them. I see a couple of ways to do that:

1) Look for product data on a gripper that you think is similar to yours. Automation magazines would be a place to search.

2) Define some plausible motions and calculate the loads that result. Then take the worst combination as your limits. For automation work, cycle times can be very, very short, much less than a second to pick and place an object, so things really have to move. Unfortunately, you do not have control over the entire system, particularly the robot arm making the move. If it is jerky, the loads on the end effector, your part, go up seriously. So, some things to think about ...

 

[Femme_physics]

Thanks a lot OldEngr, these will be important tips I can work with down the road to make a more comprehensive work.

I have a trouble though finding a detailed product description about chuck grippers. I want to see basically the method of operations, mechanics, etc. We're still beginners in the field of mechatronics, with only some solidworks, basic technical mechanics and material strength behind us, this is our first big project. Then again, I don't shy from hard work and a chuck gripper does seem ideal. But, there are so many options in this field, and so little information about it that I can find. I even downloaded 2 PDF's about grippers. None seem to have a special section about the chuck grippers. I'm basically looking to understand the operational mechanism of the chuck gripper itself, and what's more favorable in what conditions for the chuck gripper - electric, hydraulic...etc...

A quick search shows me that it's more popular in the pneumatic field.

I'll refer to my professor and classmate tomorrow with questions about the acceleration, part positioning (even though I suspect his answer is "adapt to conditions"), and what does he think about a chuck gripper.

Thanks :)

Damn, our grabbed part is quite challenging.

 

[OldEngr63]

As you said, there are so many options, and that does indeed make it difficult to choose the optimum for any particular situation, particularly when you are supposed to be flexible. You are going into a very interesting field, and one that has a good future, so hang in there and don't be at all discouraged. To be sure it is challenging, but it would not be fun if it were just all cut and dried! Cheers!

 

[I like Serena]

Oh okay, so I didn't know what to give you, so here goes:

Happy birthday!

 

[Femme_physics]

hehe..thanks ILS.

And thanks, OldEngr. I guess we have time with that, yet, so I'll take my time do more research and see if I can come up with anything interesting. :) Appreciate it.

 

[OldEngr63]

Please keep us posted on what you decide to pursue for this project. There are so many possibilities, with no one of them being "right" with the others 'wrong." I'm very interested to see what you will come up with on this.

Have you taken a course in the kinematics of machines, something often called Theory of Machines? That would be a big help at this point if you have not. You might look for a ToM textbook for some guidance.