Pattern joint origin

[R4SMEs]

Is there any way to pattern a joint origin?

I have a post with a pattern of holes, N vertically x 2 horizontally.
(N is currently 18, which is 2* numberOfShelves since a shelf bracket involves 2 pegs.)

Shelf brackets having protruding dowels have to joint to these holes in the post.

The joint origin disc is required at the bottom point of contact between the shelf dowel and the post hole, perpendicular to the axis of the dowel.

The joints are a little complicated since the hole is intentionally oversize relative to the dowel diameter and the angle of the hole differs from that of the dowel, refer graphic

Unfortunately, it seems that it is not possible to select a joint origin as a feature to be patterned.

The bottom left JO is easily defined based on the sketch line.

The bottom right JO can also be easily defined based on its horizontal offset from the other.

Defining other JOs as we go up the post will be more complicated.
The first JO is not horizontal, being perpendicular to the dowel axis (not to the hole axis).
The offsets in the JO definition process are not simply vertical, refer graphic.

This is do-able by applying some simple trigonometry since the vertical and horizontal spacings together with the angle are defined as parameters.
However, I am wondering if there is an easier way to go about doing this?!

Manually defining each JO is also unsatisfactory since the number of holes may be increased or decreased.
So, ideally the model should automatically update.

Another approach would be to pattern the line in the "Left" sketch. i.e., the construction line which defines the lower dowel in profile.
Then to define each JO up the post based on that line (as was done for the first JO).
That approach avoids the need for use of trigonometry to define JOs having relative offsets from the first.

However, I don't like using patterns in sketches.
If I subsequently decide to change the number of holes, I assume that the sketch would not realise this change had occurred.
Maybe, if I define the pattern in the sketch using the parameter 2* numberOfShelves, would the sketch respond to the parameter change (yet to try this ...).

Before someone suggests it, a fudged solution based on simply vertically patterning the brackets and shelves is not really satisfactory.
It may be desired to place particular shelves with different spacings between them - i.e., not all holes may be used.
Also, other things than shelves may be involved such as small cabinets - so there really needs to be a joint origin for each hole providing a convenient reference to hang each particular item at the desired location.

The model of the post is attached.

[etfrench]

There is only a 'Post' in the attached thread.  You'll need to include the shelf and/or dowels to create joints.

One way  to solve the problem is with hidden components.  For example:

1. Create the dowel as a component (horizontal on the origin).
2. Create a dummy dowel (on the origin) the same size as the hole and at the desired angle between the dowel and the hole.
3. Use an As-Built joint between the dowel and the dummy dowel.
4. Create a Rigid joint between the dummy dowel and the hole ( Select the center joint origin of the dummy dowel and the face origin of the hole.)

See attached file.

As far as patterns go, you may be able to pattern the dummy dowel and the dowel and just suppress the unneeded clones.

[R4SMEs]

Yes, there is only a 'Post' in what I provided.

My approach is to have each component of a modular shelving system as its own model.

Then a particular instance (e.g., system with 3 shelves 300 mm depth + 1 cabinet 400 mm depth/ double width (3 posts)) is a model which simply imports the various components and joins them together.

I haven't looked at the details of your "hidden components" approach yet - but shall certainly do so shortly.
I am not sure whether or not it would work with my multiple model "kit of standard parts" approach?

Another approach would be to have the JOs of the post flat on the surface of the post, rather than at the dowel angle.
Then in the system model, I would join the flat vertical face of the bracket to the JO of the post.
(As distinct from the present approach which involves joining the bracket's dowel to the post.)
The vertical height will be known from the various parameters.
That approach should at least avoid the need for getting into trigonometry!

I made a previous posting entitled "Joining to imported components"
https://forums.autodesk.com/t5/fusion-360-design-validate/joining-to-imported-components/td-p/951201...

The main subject of that posting was quickly resolved - after I learnt how to ground a sketch.

However, it is still an open issue as to how to best go about managing parameter usage for the "kit of standard parts" approach.
This was briefly discussed as a second subject of that previous posting.

Each component needs to "know" many of the parameters of other components.
A particular component such as the post may have some detail changed such as hole spacing and it is desirable that the change will be known to other components such as the bracket.

At the moment I am duplicating parameters in each model.

Another approach could possibly be based on importing a component, using it as a tool in some way and then suppressing it. For example, if the 'Bracket' is done first, the 'Post' design could perhaps import the 'Bracket' and use it to cut the holes in the post?
Is that a possible approach instead of the 'Post' design having to duplicate the parameters of the 'Bracket'?
Is this a somewhat similar idea to your "hidden components" approach?

[R4SMEs]

Here is the model for a complete "system", i.e., a particular instance resulting from integrating components from the "kit of standard parts".

It can serve as an example for experimenting with approaches for how one model can know about parameters of the other models!

The present approach is not elegant - it simply duplicates parameters and also some sketches between the models.

[etfrench]

Instead of a dummy dowel in my file, you can use a Plane at Angle with a sketch to locate a joint origin at an angle to the dowel.

I think you only need one joint between a shelf and the post.  The other shelves or cabinets can use a rigid joint between the first shelf and the new at a fixed distance with the distance a User Parameter in the Post file.

[R4SMEs]

Having spent more time with this, I have reached the following conclusions:

To facilitate placement of each bracket (or whatever other component) at any desired hole position it is most convenient to have joints defined for each and every of the 18x2 holes on the post.

Otherwise, in the "system" model I should need to position using offsets from a single particular hole position that has been provided with the joint origin (JO).

For this higher level model it is much more convenient to be able to position the items simply by matching up JO discs without having to get involved with any mathemetics!

It is easier to have JOs being defined as a disc on the flat front surface of the post, rather than at the dowel angle.
Then in the system model, I can join the flat vertical face of the bracket to the JO of the post.
(As distinct from the previous approach which involved joining the bracket's dowel to the post.)
The vertical height between JOs is known from the various parameters.
This approach avoids the need for getting into trigonometry!

In this way I can easily define 18x2 JOs on the post, although the process is tedious.
The result is shown after defining 4x2 of the JOs:

A conclusion is that it would be useful to have support added in Fusion 360 for patterning joint origins.

Alternatively (ideally, additionally), it would be useful to be able to turn on or off sketches of an imported model (without breaking the link to the model).
If I could see the horizontal construction line that defines the intersection with the bottom edge of the hole then I wouldn't need the JOs, having instead the possibility to joint to the end of that construction line.
[I could pattern the construction lines in the post model so that 18x2 lines were available.]

The subject of parameter usage/ maintaining consistency of parameters between associated interdependent models remains open ...