Positioning parts; struggling with joints

[MQCD_Daedalus]

Hi,

I'm trying to design a solar panel pergoda, and I'm struggling with positioning of parts. Advice would be gratefully received. This is what I'm doing:

1/ Draw a solar panel as a 'New Design'.

2/ Drag (I'm not sure of the proper term for this, but I drag it in from the project browser on the right) this into a new design and make a rectangular pattern (based on user parameters). Then make a rigid group of the array.

3/ (This is where I get stuck). I again 'drag' this into a new design, and I want to position it with one edge a parameterised height above the XY plane, and co(planar?, if a line and a plane can be coplanar) with the YZ plane (see edge below). Which will determine one of the two heights of the sloping (pent) roof.

And I want to position an opposing edge at a (slightly higher in this case) height above the XY plane, to define the highest edge of the pent roof.

Also, I guess after this it would have one remaining degree of freedom, so I'd need to put an edge like below coplanar with the XZ plane.

Here's what I'm aiming for:

I then want to design the frame around (referenced to) this 'floating' array, such that the frame adjusts as and when i adjust the h1 and h2 user parameters.

I've attached the fusion design into which the array was 'dragged'.

I'd really appreciate any help you can give.

Thanks,

[davebYYPCU]

Simple joint - need how high is H1, and what angle the panel will be.

In the pic I have H1 as 108 high.

Angle is 20 from horizontal, but dialogue box needed 110.

Imported Test 2.  Rigid Group all Panels.

Created top level sketch and drew a line on X axis from Origin.  Origin hidden for clarity.

Joint the Lower Right corner of the panel on the ZY face, to the end point of the sketch line at the origin end.

On the Dialogue box, lift the panel to H1 high, and rotate the panel 20 degrees or calculated angle from H2.

When this part is done just the corner panel is in place in preview, 

press Ok, and all panels will align.

Might help...

[MQCD_Daedalus]

Hi,

Thanks for your fast reply, that has helped and I think I'm on the right track. If I can i'd like to specify the two heights H1 and H2, rather than H1 and an angle. To that end, I've made a sketch (i never thought to use sketches to joint to), where I've made one 'revolute' joint on the end of a sketch line for the H1 side, which allows it to rotate around the axis I've shown in green. I now want to joint another corner to the higher line, which I've shown in red, and I'm not managing to achieve it.

Thanks again for the help so far.

[davebYYPCU]

Ok, much easier.

Fusion sends a warning if the second joint is on the same components, the warning can be ignored, but is not required if you start thinking outside the box.

Draw the end sketch.  Dimension the lines H1, H2, add a rafter.

Select the midpoint of the bottom edge of the middle side panel, and joint it to the midpoint of the rafter.

Might help...

[MQCD_Daedalus]

Hi,

Thanks, that sort of works, but it requires a level of knowledge about the dimensions of the panel array, (see image below with an exaggerated example of how it could go wrong if i don't constrain the length of the rafter to the overall width of the panel array (i.e. 3xpanel+2xgap); the effective heights h1 and h2 would be wrong). Is there a way to achieve it without manually entering the length of the rafter?

Thanks again for the help.

[davebYYPCU]

You don't want to use a known / unknown Angle from a front edge, 

That means you have dimensions to refer to, between posts, or length of rafter, 

after the 2 selected points, you can slide the panel wherever you want - away from the midpoint, 3 directions - above the roofing, back from the gutter, horizontally away from the rafter, all those offsets are available, before the OK, and later coming back to edit them.

Parameters H1 and H2 are your values, and can be updated, the panels will adjust as required.

Might help...

[MQCD_Daedalus · Accepted solution]

Hi,

I worked out how to do it. The first joint needs to be a revolute, which allows the array to 'hinge' about H1, and the second joint needs to be a pin-slot where it slides along the h2 line and can rotate around the plane that the sketch is on. Thanks again for your help, I would have been stuck without knowing that I can joint to sketches...

[davebYYPCU]

I think you missed the logic.

2 Joints are better than one, Ok.

[MQCD_Daedalus]

Hi,

It's highly likely I have missed the logic, but surely 2 joints are better than one if it means that the parametric design works? As far as I can see, using the method mentioned previously I would have had to enter information concerning the dimensions of the solar array (i.e. the rafter length, or distance between posts, or offset from the centre joint etc) which means if I change the array, I would have to manually renter the changed dimensions into this design, you said it yourself "coming back to edit them"? (I know I could use 'global' variables, but that seems like an unnecessarily complicated work around, for what should be a simple positioning requirement). If you can suggest a method that requires only h1 and h2 data, and isn't dependent on the panel array dimensions, that avoids having 2 joints (which I've inferred from the tone of your last post is frowned upon) that would be great, as I'm really struggling to get my head around the F360 paradigm of joints. I come from a SolidWorks and SolidEdge background where the positioning was far more intuitive to me, and F360 is completely different from these.

[davebYYPCU]

Let's go through that.  Object of the exercise, get the job done efficiently.

2 joints are better than one if it means that the parametric design works?

Setting up 2 joints - twice the workload.

As far as I can see, using the method mentioned previously I would have had to enter information concerning the dimensions of the solar array (i.e. the rafter length, or distance between posts, or offset from the centre joint etc)

Same for any model.

which means if I change the array, I would have to manually renter the changed dimensions into this design,

Nope Never.  Change anything in the linked file, will update here.

you said it yourself "coming back to edit them"?

I was talking to this destination file.

(I know I could use 'global' variables, but that seems like an unnecessarily complicated work around, for what should be a simple positioning requirement). If you can suggest a method that requires only h1 and h2 data, and isn't dependent on the panel array dimensions, that avoids having 2 joints

Attached...

(which I've inferred from the tone of your last post is frowned upon) that would be great, as I'm really struggling to get my head around the F360 paradigm of joints. I come from a SolidWorks and SolidEdge background where the positioning was far more intuitive to me, and F360 is completely different from these

I have no experience with those software packages but sounds complicated.

Joints in Fusion - require a common point in the assembly, that belongs to each component.  Align / orientate at time of selection, the faces of the 2 joint discs,  - at that common point, set any offset you wish.  Thats it.

In my file I have the Assembly file.

1.  Insert your Test 2 file into it. (Keep linkage)

2.  Make the array of panels, a Rigid Group.

3.  Secure the panel to the appropriate location, by making a Rigid Joint - selecting the midpoint of the lower edge of the middle side panel, to the midpoint of the rafter line. (as pictured previously.)

Edit parameters if you wish.

Report back if this is not as you expect the file to behave.

Might help...

[MQCD_Daedalus]

Hi,

Thanks for that, and thanks for sticking with me. I've done what you suggest, which to works out because 5000mm is almost the width of the array (4970mm):

But if I edit the inter-panel gap, for instance, it no longer works, the low edge is too low, and the high edge is too high. To make it work, i would have to change the 5000mm to be the new width of the array:

I want a method where the array can change, but the two heights of it in the assembly automatically remain correct.

[davebYYPCU]

When you change test 2 panel span, 

edit the 5000 dimension to match.  One update.

But if you build the sketch after the import, that panel span is a known and parameterizable value.

Might help...

[MQCD_Daedalus]

Ok, I understand.

Thanks ever so much for your help with this.