Assembly - insert sphere bearing into hole

[craynerd]

First to say I'm new to inventor but this one even confused my buddy of mine who is quite good with it.

I've created a 5mm hole in a 4mm thick rectangle of metal as one part and drawn an 8mm bearing as another part. In assembly drawing, I want to insert the bearing into the hole as far as it will go until the edge of the smaller diameter hole meets the bearing. Any constraint we use either holds the bearing to the front face of the hole and not into the extra depth it should be able to go into!! I could do a calculation to find the depth it should go but is there no way of saying "place the ball bearing as deep into the hole as possible until it hits the edge of the hole?" Other methods I have used have made the bearing going into the material of whole!
Help appreciated

[admaiora]

Hi Cray,

i am not sure on which results do you want... anyway i show you same way..hoping that there is what you mean.

[craynerd]

No sorry, that is a bearing race assembly, I'm talking about a single spherical bearing.

I know it sounds strange, but I want to place or sit a sphere bearing on top of a smaller diameter hole. The hole is 5mm the bearing is bigger. I just want to rest the sphere on the hole and let it drop into the hole as much as it can... Obviously it can't fall through as the diameter is too small on the hole. I can't sit the sphere on the hole edge.

[admaiora]

Ok Cray,

There no direct constraint that allow this particular position

more or less something like this?

Nothing more came to me better than this at the moment.

[JDMather]

This is easily done with the Insert Constraint and a "trick".

Attach your assembly here or at least indicate what version of Inventor you are using (I don't want to work up an example, only to find out you can't open my files.)

It seems like some variation of this problem comes up here every week.

To see the "trick" edit Sketch1 of the Sphere.

[admaiora]

With trick do you mean additional sketch geometry? i Am  curious

[admaiora]

No Insert..but is it near?

[JDMather]

---

@admaiora wrote:

With trick do you mean additional sketch geometry? i Am  curious 

---

[admaiora]

Nice!

Thanks JD for sharing 

[-niels-]

You could also do it like this, using only constraints:

With a fair warning that this has 2 solutions, if you drag the sphere it can go to the other side of the hole.

So you might have to ground it to make it stay in place.

I think JD's method is the most robust for removing degrees of freedom, just wanted to show another method.

[wimann]

I would take a similar approach to what JD has suggested except I might just use work geometry on the line that represents the hole in your plate instead of splitting the solid. That's my personal preference so that the part doesn't appear to be split on a drawing. However, it all depends on how you move forward with it and what works for you. I do agree to adding the extra geometry in your sketch though.

People at my work are alwasy asking me how I can get something to bend a certain way or how I can get complex geometry to solve itself without having adaptivity or iLogic. The answer is really simple. Sketch out your requirements in the part file that requires the flexibility. If done correctly, you should have a part that works out it's own complex geometry.

Anyway, hope this helps.

[jeanchile]

I don't have time to read all the posts, and I can't seem to locate the specific blog post that I am referencing, so I apologize if this doesn't offer much help but....

Curtis Waguespack has a neat little trick for this using a simple axis-mate constraint and the contact solver. It's in his Mastering Inventor book or on his blog. You mate the centerline of the hole and an axis from the ball bearing. Make the two parts a contact set. Activate the solver, move the bearing so that it is touching the hole. Add another mating constraint but check the "predict offset" option. Maybe someone not on their phone can track it down?

Here's the blog: http://inventortrenches.blogspot.com

[wimann]

Jean,

I've seen that method but I feel like using sketch geometry to fully contrain it is still the more reliable way. It's more flexible. Using the method you describe, any change to the size of either piece means you have to repeat the process. Using the sketch method, changing the size of the ball means you have zero processes to repeat, and changing the size of the hole only means you need to change that parameter in the sketch in the ball part file.

That's my 2 cents.

Thanks,

[jeanchile]

Yeah, you're right. Plus, if you link the two parameters (the hole and the sketch circle) you only have to change the hole. Both good ideas I hadn't thought about. I just immediately thought of Curtis' method when I saw the post.