Clamping Cylinder Motion/Constraints

[phlyx]

We use clamping cylinders in some of our assemblies.  These are pneumatic linear cylinders that rotate 90° during a small portion of the stroke.  

For example the one we are looking at travels as follows.  With the piston fully down, it travels up 20mm without turning.  Then it travels and additional 9.5mm and during this part of the linear stroke, it rotates 90°.  This swings the clamping arm out of the way for loading and unloading.

So the challenge is constraining the piston so it travels straight for 20mm and then turning 90° while traveling another 9.5mm. 

Any help? 

[swalton]

I'd try using a transition constraint to a surface that represents the clamp arm path.

[JDMather]

Can you Attach an example cylinder here?

Is the cam in the mechanism?

If this motion can't be done with Transitional constraint - it is rather trivial with Dynamic Simulation - Input Grapher.

[phlyx]

These have a built in cam but the cam isn't in the model.  Just the center post extends and after extending 20mm, it extends another 9.5mm and rotates 90°

[j.palmeL29YX · Accepted solution]

If >>this<< is what you're looking for, see attached iam. 

Please mark "Accept as Solution" if my reply resolves the issue or answers your question, to help others in the community.

[j.palmeL29YX]

If you only want to SEE the movement (to demonstrate the functionality), you can also create an animation in the Inventor Studio environment.  (see attached video). 

And, of course you also can show and detailed analyze the mechanism in the Dynamic Simulation environment. 

The choice depends on the purpose of the animation.  We don't know what you need it for. 

[phlyx]

That's definitely it!  Thanks for the help.  Now I'm going to ask how the angular constraint value works:

min(( d12 * 1 deg / 1 mm * 90 ul / 9.5 ul + 90 deg / 9.5 ul * 24 ul );0 deg)

d12 is the distance from the end of the shaft to the top of the column but not sure how the rest of that works or what the syntax is.  Would like a lesson in formulas in constraint values, honestly did not know you could do that.

[j.palmeL29YX]

---

@phlyx wrote:

Now I'm going to ask how the angular constraint value works:

---

Tomorrow I will give you an explanation. Today is too late. I hope you can wait so long. 

cu

[phlyx]

No rush, I just like to know as many tricks of the trade as I can.... been that way since AutoCAD 2.5.  

[RNDinov8r]

So, I took the iLogic approach.

In my parameters I created these two constraints.

Then I wrote this iLogic Code.

If Stroke <= 20 Then
	RotationAngle = 0 deg
Else
	RotationAngle = (1 - ((39.5 - Stroke) / 39.5)) * 90
End if	

I applies a made constraint to my piston nad cylinder for the stroke, I contrained to an axis to hold the rotation, then I set an Angle Constraint = to the parameter...so wehn I updated my Stroke distance, it determines what the value is.  this formula easily updates for any travel lenght. the '39.5" is the max stroke. and the "20" is the stroke lenght at which the rotation begins to occur. So this works for any situation, just by replacing the "20" and "39.5" with updated values.

[j.palmeL29YX]

---

@j.palmeL29YX wrote:

---

---

Tomorrow I will give you an explanation. 

---

A first note: 

My suggestion above is too complicated, the dummy part (temp.ipt) and its relationships are not necessary. You can delete the part temp.ipt. (It would be only necessary if the movement is more complex). For the angle we can write a function depending directly on the parameter d0 (the Flash constraint). But the formula will be the same (only d12 substituted by d0): 

We can imagine the angle as a function of the linear movement. 

d21=f(d0) 

(see attached image, where  x-axis = d0, y-axis = d21)

While a move of 9.5 (-24 ... -33.5) mm the rod must rotate about 90°. 

So the slope of the line is 90/9.5. 

The angle is 0 where the movement is at -24 (and higher). 

So we can write a linear function 

d21= d0 * (90 / 9,5) + 24 * (90 / 9,5)

Check it: 

  d0 = -33.5  => d21 = -90

  d0 = -24       => d21 = 0

If d0 will be greater than -24 the calculated angle d21 will be greater than 0. Therefore I use the min-function to get always the value which is the smaller one of both, the calculated value or 0. 

d21= min((d0 * (90 / 9,5) + 24 * (90 / 9,5)) ; 0)

At last we need to clean up the units of measurement. d0 brings the unit mm, the result d21 needs the unit deg. 

Therefore we remove the mm with a division by 1 mm and add the degrees with a multiplication by 1 deg. The factor (90/9.5) is unitless (ul).

The complete formula:

min(( d0 * 1 deg / 1 mm * 90 ul / 9,5 ul + 24 deg * 90 ul / 9,5 ul);0 deg) 

Now we can drive the parameter d0 as shown in the video above. 

HTH

Please mark "Accept as Solution" if my reply resolves the issue or answers your question, to help others in the community.

[phlyx]

Thanks @j.palmeL29YX and Chad, good to have options.  I have done invisible parts before to drive things but couldn't see how to do this.  Was a little lost on the syntax but that clears it up.  Will have to study it a little more but can see a ton of opportunities for calculated constraints. 

The one thing that it still lacks is the ability to manually 'drag' the parts and get the same rotary motion.  Driving is fine but being able to drag and get the motion would be better.  But beggars should not be choosers. 

Thanks! 

[JDMather]

---

@phlyx wrote:

Driving is fine but being able to drag and get the motion would be better.  But beggars should not be choosers. 

---

You could model the internal cam slot and follower pin.

This one wouldn't be too difficult.

Then add a Transitional Constraint in Assembly Environment or a 3D Contact Joint in Dynamic Simulation Environment.  Motion then would occur on dragging...

[RNDinov8r]

 If you did model the cam, then in theory you would just set the two bodies to be contact bodies in a contact set. But, in large models, this isn't ideal. 

[j.palmeL29YX]

---

@phlyx wrote:

... being able to drag and get the motion would be better.   

---

The above question remains: What do you need such an animation for? Just for fun, to play and to enjoy how the parts move each other? Or is there a seriously background? 

[phlyx]

I can say from personal experience that modelling a linear slot in a part that then rotates while moving linearly so the sides are parallel (as if a mill cutter ran thru for a cam to follow) isn't the easiest of tasks, either.  Made parts that did that at a previous job for actual parts on a machine that acted like the clamping cylinders (just noting that me saying that will probably get me a load of replies on 'easy' ways to do that now... LOL)

[phlyx]

Not needing an animation, actually need to move the parts thru the motion.  Have some clamping cylinders that need long arms to clamp parts on the base of a purchased unit.  We have to clear everything with a new part delivered with robot EOAT.  So unless we can see where the clamp tool itself moves we cannot determine the actual relationship between it and other tooling and fixture components.  There are also four (4) clamping cylinders, two LH and two RH clamping all four corners of the part.

[JDMather]

---

@phlyx wrote:

 (just noting that me saying that will probably get me a load of replies on 'easy' ways to do that now... LOL)

---

The new Sweep (cut) with Solid makes it easy.