Alibre Atom assy problem

Does anyone now how to assemble a sphere into a through hole, in a flat plate?(The sphere being a larger diameter than that of the hole in the plate). I can get the axies all lined up, but when I apply the "Alibre Quick Command" to the the sphere surface and edge of the hole it should nestle in, the sphere refuses to move into the hole.

I hope I have put this thread in the right place.

Regards

Mark

If you extrude the plate "mid plane" this will give you an axis through the plate. You can then use that axis to set the sphere middle of the plates thickness.

Sphere won't move into a hole if as you say the sphere is bigger than the hole so surface and edge can't match

Edited By JasonB on 20/01/2019 13:25:45

Right click on sphere -> show reference geometry. You can then see the planes/axes that the sphere was built on - use these with the constraint tool to place the sphere relative to the plate.

Size of sphere won't matter to the software, even if it can't be done in real life.

You should be able to align the cylindrical surface of hole, with an axis of the sphere (handy of hole isn't placed to coincide with an existing axis in the plate).

Edited By David Jupp on 20/01/2019 13:36:00

I had a play with this, got a slightly surprising result. In the image, there is only a single Align constraint between surface of sphere and cylindrical surface of hole.

I wasn't surprised that the centre of sphere is lined up with hole axis, BUT I didn't expect it to be fixed at mid-thickness in the plate. I can 'tumble' the sphere about its centre, but can't drag it away from location.

Edited By David Jupp on 20/01/2019 13:53:14

Have just experimented and got a slightly different result ! Sphere 'on axis' , but fixed some way from the plate

Conclusion - use the reference geometry of the sphere for reliable constraining.

Thanks very much for the feed back chaps, but unfortunately I didnt explain the problem too well.

I have a fixed size ball bearing (6mm dia) that I want to trap between two plates that both have a through hole of 3.5 mm dia.

I was hoping to use the "Quick" command to nestle the ball bearing between the plates, ie let the software work out the geometry, but I've tried using the ball surface and the edges/sides of the through hole as constraints, but when I apply the command it doesnt want to move despite all the ball and hole planes and axies being lined up)

The geometry is simple enough but I was hoping Alibre would handle it rather than me have to do the trig manually and set up a plane of intersection.

Hope I've made things a bit clearer.

I could not get it to do that but you can use the software to work out the gap during the sketching stage and space the two plates that distance apart in your case 4.873397mm and then position the ball half that from one face of the plate.

I've managed to do it but it's a long winded way and my method above will be a lot quicker. If you want to let the assembly tools do it for you then this is how I did it:

1. Add a small chamfer to the edge of the hole 0.1mm will do

2. In assembly align the axis through the middle of the hole with the axis through the ball

3. Using standard Constraint tool select the face of the chamfer and the face of the ball and then the dot "tangent inside"

When you click apply it looks like it has done the job but if you section the parts and zoom in you can see there is a gap

If you zoom in further you can see the ball is actually sitting where lines projected from the side of the hole and face of the plate would meet and that is where you say you want the ball.

4. Anchor the two parts, save and close the assembly as it stands

5. Open up the plate part and delete the chamfer, save and close

6. Open up the assembly again and now things are where you wanted them.

Jason - standard (traditional) constraint tool is not available in Atom (except perhaps when editing existing constraints). So if this approach is possible in Atom, it'll require even more steps to achieve.

OK David though when using the Quick version Tangent does come up along with Align, you can click OK but it does not work.

Fiddling about does bring up the question as to whether Mark can actually make the plate with a perfectly square edge to the hole, more than likely to need the burr removing which will create a small chamfer so the ball will drop further into the hole rather than just touching a theoretical edge. Again easy to get the depth with a chamfer while sketching.

Edited By JasonB on 23/01/2019 09:07:40

Thinking about it - a 45 degree chamfer on a particular hole size will only work with specific ball size range. Smaller ball will sit on inner edge of chamfer, larger ball will sit on outer edge.

Unfortunately there isn't a constraint type that works universally here.

Pro works when the chamfer is larger by using tangent outside.

Yes - should work if sphere size allows it to sit on the chamfered face. If the ball is larger, the tangency condition moves beyond the chamfer, and sphere will have to 'hover above' to fulfil it.

Of course the chamfer angle could be changed to suit, but it all gets a bit involved...

I typically resort to creating a cross section view (so I can see what is going on), then adjust a constraint by eye to fix sphere distance from plate. Low tech, but usually adequate.

Sorry to butt in,chaps. Instead of using the chamfer tool to create the edge, couldn't you revolve a cut-out with an arc to form the "chamfer" the same radius as the ball, which would then fit exactly to the sphere? (To answer my own question - yes, you can, I've just done it!)  Sort of inverse fillet?

John

Edited By John Hinkley on 23/01/2019 11:41:58

Yes - and you could extend that idea to use an arc in a sketch in the plate part, to position a 3D reference point, which could then be picked up in the assembly to position the sphere. That way you wouldn't even need the cut on the edge of the hole.

There's usually several ways to tackle the problem.

Just calculate the distance from the plate to the centre of the sphere. Got to be easier than faffing around with model geometries.

Andrew

That does seem to be the simplest way, I just drew a line, set it as 3.5mm long and measured the offset on the sketch as per my first post today.

Only down side of using a dimension if that if you later alter the size of the hole or sphere it will not automatically adjust the positions of the parts

That's cheating, do the maths!

Andrew

Sorry Sir.

The other thing I don't do anymore is work out trig, far easier to just draw two lines and the angle if that is what you know and read off the missing third item in Alibre than try and remember your opposites and adjacents