'Sticking' components together


#1

I am trying to draw up a bowl which features staves in it’s construction. However, having drawn one of the staves, I’ve been trying to ‘assemble’ the bowl sides without too much success. Whilst I have used components for a while, I must confess having read through the tutorials and explanations, I can’t say I’ve really grasped how to specify, import correctly and generally use the component function. In this example I have been laboriously ‘sticking and rotating’ each component together, hence the additive error that has crept in to the assembly. Is there a quick way to accurately ‘stick’ the mating faces together?
Stave.skp (19.7 KB)


#2

Use the Lathing function of the Follow Me tool.

-Gully


#3

It seems you’re trying to develop the dimensions and angles of the stave for a given size of bowl.
Try the reverse. Model the staved bowl and then erase all except one stave.

Stave Component.skp (79.6 KB)


#4

Oops…
I notice my method of scaling made the interior/exterior faces of the stave non-parallel.

Here’s a fix for that:
Stave Component II.skp (83.5 KB)

Think of a Group or Component as a ‘container’ which serves to isolate geometry.
The geometry within the container is separate and protected from inadvertent alteration.
A Component in the model space is an instance (copy) of the Component Definition in the model file.
The power of instancing is when you edit one instance in the model space; SketchUp automatically updates the definition and all instances in the model.



#5

I think your problems have little to do with components and a lot to do with how you drew the stave. You don’t say how you arrived at the compound angles or made the miter cuts, but it would be easy to introduce small errors at that stage. Plus, you are trying to create the bowl shape itself at a compound angle, which makes things difficult. It would be a lot easier to create it aligned with the model axes and then move or rotate it to final position as necessary.

One quick way to make an accurate component would be to use Gully’s technique and then “slice” the shape by drawing the top and bottom lines of the miters and then erasing the rest of the staves to isolate just the one. Afterward you can use a rotate/copy with a repeat of 12 to regenerate the bowl from the single stave.


#6

Perhaps it’s worth mentioning that if you want a model of the finished bowl but you don’t necessarily need each stave represented as a separable component, you can just turn the shape from a profile as @slbaumgartner and I have demonstrated and then just draw in one radial seam on the top (and bottom) and copy it around using a radial array. Also, if you want a bottom for the bowl, as I presume you do, you don’t have to construct it separately: you can include it in the original profile. Thus:

-Gully


#7

First, I would second all the advice given above. That said, I was fiddling around with the OP’s model just seeing if I could use his stave to make the item. Take a look at the gif and see if I’m missing something really obvious (a distinct possibility). It seems we should be able to capture endpoint inference. Maybe it doesn’t because it’s in the same component?

Shep


#8

I think the problem is that you have opened the selected instance for edit and are moving the Face within its context, i.e. modifying the ComponentDefinition. The adjacent instance is simultaneously going through the same edit (watch it move too in the gif), so SketchUp is unable to determine where it is located before you finish!


#9

HI Gully, and others!

I have used the follow me tool to generate the bowl, and then have isolated the lath. This is the part that has to be produced on the compound mitre saw. However, when I ‘measure’ the lath, using SU, I don’t get the same angles I have found in tables for compound mitres.
For example, a 12 sided bowl with a 30 degree angle, from tables, requires cutting angles of 14.511 and 3.841 degrees respectively. However, SU seems to leave the bevel angle at 15 degrees as if it is still a parallel sided bowl). Am I measuring incorrectly, or does SU simply average the sizes out. I was hoping to draw the bowls in SU, and use the dimensions from the resulting lath to set up my saw.


#10

Check Window->Model Info->Units and see whether you have angle units precision set to 0. That will cause SketchUp to display whole number angles even when it actually has finer detail. You can set up to three digit precision.


#11

If you want to make the corner points meet there is a great inference-locking method that I learned from Alan Fraser (has created great human models for FormFonts) in the original @Last forums:

  • With no objects selected, select the Move tool
  • hover it on the top edge, when the “on edge” tooltip appears, press and hold down Shift
  • click on the endpoint with the tool, point to the other top edge, click and release Shift
  • move the endpoint of the other face to the endpoint you just moved.

Anssi


#12

Thanks for that, it did improve matters for the angle of the lath giving 7.4 degrees, but the mitre angle still ‘measures’ 15 degrees, this being the angle of a 12 sided figure with no side slope, which isn’t correct.
Any ideas?


#13

You might not be using the protractor perpendicular to the cut. Try click-drag along the edge you want to measure.

Shep


#14

Shep is correct. For a miter joint you have to measure the angles in the planes of the board face, miter gauge, and blade - which aren’t flat to the x,y plane. In a flat top view, you just get the angle of a twelve-sided polygon.

My plugin “Rake and Splay” (available on the sketchUcation plugin store) can help to see this as well as to set up compound joints and chair legs (though currently it is only for 4-sided figures).

Steve


#15

Hi Steve

I understand the need to ensure that the angle is measured in the plane of the face. What I have been doing is selecting the protractor tool, and then positioning it on the face to be measured. I then press and hold the shift key, which I thought held the protractor in that plane, and then measure the angles. Is this correct?
I’m afraid I didn’t quite understand your click-drag method, sorry!

Dave


#16

Click Drag lets you orientate the rotate and protractor tools along an edge rather than a face.
Select the tool, then click drag click along an edge and it will be perpendicular to that edge.
The gif is the rotate tool.

Here is a simple prctical application of click drag with the rotate tool.
Not really related to the thread but a way of understanding how the tool works.


#17

It is very difficult to understand these angles well enough to measure them correctly. That’s why most people refer to tables or calculators. In your case, there appear to be two problems, neither of them SketchUp’s fault!

First, you don’t say what tables you used for looking up the angles, but I believe they are either wrong or don’t mean what you think. For example, using the online calculator at

http://www.pdxtex.com/canoe/compound.htm

for a 30 degree slanted 12-sided joint, I get angles of 13.06 degrees and 7.435 degrees.

After setting up to measure these correctly on SketchUp, I get the attached drawing which matches the calculator’s values. The bevel angle is between the normal to the Face and a line lying in the mitered Face perpendicular to the Edge. The miter angle is between the miter Edge and the top Edge. Most miter gauges mark the right angle as 0, so the smaller (complementary) angle is what matches the calculation.

Hope that helps

Steve


#18

I suppose it would be more useful if I showed you what I meant by “after setting up to measure these correctly” which is the second issue I mentioned ! :wink: The key is to make the measurements in a plane that is normal to the edge of the miter, as follows:

The attached GIF shows one process using just the protractor tool. First I dragged along the miter edge as Box described to make the protractor snap normal to that edge. I chose a point on the far edge of the outer face as the reference for the protractor so that it is measuring from the plane of the face and chose an angle of 90 degrees to create a construction line normal to the face. Then I repeated the process but used the normal line as the reference and measured the angle to a point on the other edge of the miter face.


#19

Steve

Thanks, I’ve not seen this technique before, but will use it in the future!!
Agree about the differences in the figures for the compound mitres. I have found a few sites and they all seem to be a bit different. The link you sent was one I had found, but was using figures from another. I guess I picked the wrong one!

Dave