Overlapping Flange for Seam Weld in Sheet Metal

[Anonymous]

I'm trying to create an overlapping flange from two adjacet previosly created flanges. The purpose is to create a closed corner to a box and provide for a lap weld rather than edge weld. I can't seem to creat an offset such that the new flange doesn't merge with the adjacet flange. It seems that this should be a common need, and maybe I'm missing an easy way to do this? Attached are edge on screen shots of the flange and error I get. I realize there is a miter that needs to happen at the corner. I've tried the overlapping flange just at the far edge and still it doesn't build.

[coreyparks]

Attached is a way I have worked around this in the past.  Just add a .001" extrusion to the edge of the sheet before you create the flange so that when the part is bent there ends up with a tiny gap between the sheets.

[Anonymous · Accepted solution]

coreyparks thanks for that tip. I also got it to work by first unfolding the model from the 'modify' commands, and then adding the second overlapping flange, and then 'refolding'. It works perfect with 90deg corners, and I used a slightly different trick, slightly increasing the radius of the bend in a less than 90 bend to get the overlapping to land at the correct height. See model.

I'm not sure anyone would sign up to make this part, but at least it now exsists in my model.

[DGodwin-XRG]

Still a problem with Inventor 2020. I'm assuming there's a patent or similar protection on the tool from a competitor's product, otherwise I would expect an "offset" field to extend the material before applying the corner, as it has existed in other software for as long as I can remember. (for this exact purpose)

[JDMather]

@DGodwin-XRG 

Can you Attach example *.ipt file here?

[DGodwin-XRG]

@JDMather  see attached, though I don't think it's going to teach you much, everything you need is discussed above. I've attached a part file with a handful of flanges, including one of which does work but is a very poor solution.

In the end, adding an extrusion on the end of the flange edges as described above will work, but it's cumbersome and prone to break with flange adjustments. The strategy that works for me but far too time consuming is like this:

1. Everything we do here depends on the relief shape. So I change my relief type (affecting the corner opening) then the rest of the settings need to be manually adjusted.

2. While editing the flange, we have to click this corner control button. Frustrating this isn't a universal control in the original dialog box, but ok, it lets us have custom styles each corner. To each their own.

3. Check the box then select the dominant corner

4. Then choose a relative ratio value (0-1, I chose 0.1), not an absolute offset controlled for tolerancing.

From here, I was able to add the final flange that resulted in what I wanted.

**but, that's frustrating because I know there's an alternate solution. See below**

I'm going to do some blasphemy here, because as I've mentioned before in other posts I switch back and forth between these two programs DAILY in my consulting work. No, I'm not picking favorites, I have many many more gripes with the other program, this screenshot is here specifically for the purposes of showing a clean workflow for this very specific purpose:

1. Choose outer or inner face for mating face priority

2. "Offset from Surface" - choose outside (or inside) mating face to reference

3. Add offset value for tolerance control of bends and ensuring flanges are not merged.

4. (as a result, common parametric edits to the flange will make the second flange move with it)

See the difference?

[johnsonshiue]

Hi! I am not sure if this has been mentioned. There is a workflow allowing this case to work in Inventor. But, it could be risky.

Here is what you need to do. Move EOP to right under Flange3. Create a zero-offset workplane on the outer face of Flange3. Next, use Split command to split the side faces (all around). In this way, the flat pattern can be created. Also, you can add flanges to the overlapped edges.

There is risk here though. The split changes topology, which is directly related to associativity. The workflow is kind of unusual. There could be issues with the downstream features built around the edges.

Many thanks!

[DGodwin-XRG]

@johnsonshiue I hadn't thought of the split method, thank you. In any case where these finishing flanges are added at the end of the part, this is probably going to be my go-to method because of the simplicity/speed of doing it. Won't need associativity below it, I've mostly needed associativity only above, due to adjustments in the primary flanges. Suppose this might also work using a sketched line instead of a plane, depending on the geometry affected. Great stuff.