Mitre bend problems

[HDEAR]

I recently updated to PU1. I was reasonably impressed with the new sheetmetal features as 90% of my work is in sheetmetal.

 

I have struck problems with Add miter though and I wonder if it isn't buggy.

 

I was making a 4 sided box out of 1.5mm st st 400 long that would have 25mm internal return flanges mitred at the corners - refer 'Miter Problem 1' which shows the design being an open butt corner for tig welding.

 

The 1st bend was used to 'Add Mitre' ( refer 'Mitre prolem 0' ) but I could only get two sides to accept the mitre bend ( refer 'Miter problem 2' )

 

So I experimented getting around the problem by using 'Vertex chamfer' on the flat ends nearest the side that couldn't accept the 'Add Mitre' ( refer 'Miter problem 3 ) and the adding a sing bend on side 4 and modifying each end to suit ( a time consuming effort I might add ).

 

Before adding the last bend I performed an unfold and got an unusual result ( refer 'Miter problem 4' ). The unfold resembles nothing like the 3D folded model would have you believe.

 

So, I decided to tackle it another way by having a return bend on the last leg and use that for spot welding to the corner. The add mitre worked on all 4 sides except that it overlapped at the return fold end ( refer 'Miter problem 5' ). There was no way I could modify the straight ends using vertex chamfer ( refer Miter problem 6' ) as the the selection process won't allow that.

 

Is this a bug or am I tackling this the wrong way?

 

Mitre_problem_1.ics

Mitre_problem_2.ics

[Cary OConnor]

The Miter has some dependency on the order of the stock. So the selection or what it can create will depend on the history order of the edges it is trying to create on. In your example, I would tackle it a bit different.

 

For example: Ending the stock at the corner will not miter the ends of either side. So I would suggest to move it to the middle or different area. See Attached.

 

Mitre.ics

 

 

[rsaucier]

Pretty slick Cary. Also, and support may want to investigate if it is a bug, but the reason the last side would not work is the corner where you will weld is touching. If you take .001" off of it, you will see you can get all 4 sides. but still have the issue with the last corner not being mitered. So I like Cary's idea but wanted to add why the last side did not work in case a similar issues pops up in another area.

Add_Miter.wmv

[Cary OConnor]

Yes if it is touching it will be an issue. There needs to be a gap on a closed sheet metal box in general as you pointed out RJ.

 

Even so, the last leg in the corner will not miter since it is not a connected edge (there is a gap). There is a way to do it manually. Create the miter with the option set to "Create Miter Flanges only". This will create individual bends. Then you can edit the properties and manually add the closed corner information. It is more difficult but possible.

 

Cary

[HDEAR]

Thanks guys.

 

Both methods from Cary and rsaucier are helpful, especially the create the miter with the option set to "Create Miter Flanges only"

 

Cary's option of having the stock meet in the middle or anywhere away from the bend looks good but unfortunately isn't an option when making the product.

 

However I still can't do what I need to carry out without creating problems.

 

I tried using Cary's and rsaucier's method in the attached 'Mitre problem 3.ics

 

Refer 'Mitre problem 7.jpg' - the result on the spot welded configuration was clearly not usable.

 

'Mitre problem 8.jpg is the butt welded option with rsaucier's gap applied and this allowed me to put the right relief of the bends and create a vertex chamfer. I had to pull one bend back to achieve this.

 

However when I move the last bend towards each other ( 9.jpg and 10.jpg ) you can see that there is yet another problem created. Once the vertexes cross, it's WWIII.

 

I have attached a frame system I drew a while back using bends and add stock. Refer CHF66-90.ics. This worked fine but was quite time consuming and if the outer sizes of the frame changed, what a mission it was to get it working right again. The other problem I have is that I can't remember how I actually did it - a function of becoming an old fart I guess

 

Summary is that I was hoping the add mitre feature would reduce this type of work to a minimum amount of clicks and be relatively simple and for many applications this is the case. However for 'closed boxes with returns' it's still frustrating.

 

Finally, can anyone explain why the unfold referred to in my first post looked completely different from the 3D folded model? ( 'mitre problem 3 and 4 .jpg' )

Mitre_problem_3.ics

CHF66_90.ics

Edited June 15, 2016 by HDEAR