Update to Beams feature

NeilCooke

A new version of the Beams feature is available (Official V7). If you have it on your toolbar you will see a blue link icon - right-click to update:



Note: this will only affect new features you create. Any existing features using a previous version of Beams will also show a blue icon. There is NO NEED to update existing features unless the blue icon is annoying you - however one of the profiles was incorrect (as pointed out by a user - thank you):

ANSI Inch HSS Square Tube 2 x 2 x 1/4 was actually ANSI Inch HSS Square Tube 2 x 2 x 1/8, so the 1/4" thick profile is now correct and a new entry for the 1/8" thick profile has been added.

This means that if you have previously used ANSI Inch HSS Square Tube 2 x 2 x 1/4 in old designs, the geometry will update to the new (correct) profile.

You can do 1 of 4 things:

1. do nothing / don't update and it will remain how it was
2. update to the correct profile
3. update then edit your beam features to the 1/8" thick profile if that was the one you really wanted
4. update and if it all goes wrong, restore to a previous time in history

Also new in this version is the addition of part names and cut lengths (no aggregated cut-lists as yet).


Here is the Document: https://cad.onshape.com/documents/e15c2c668d138f01242d0c80/v/055326ad4b507c49a81b40ae/e/7a7d925dcd7374e1a0b41cea

Remember, you don't need to make a copy of this Document, just add it to your toolbar. To add custom profiles, use the Custom tab in the feature.

owen_sparks

Very nice, thanks Neil

Owen S.

Cris_Bowers

@NeilCooke Something is off with the length measurements...for curved parts at least.

The two parts I have highlighted should be the same. I'm not sure what you are using to calculate the length, because no matter what geometry I use, I can't come up with either of the lengths shown.

NeilCooke

@Cris_Bowers - sorry, but that is expected since a bounding box is used to calculate the length, so it won't work on curved beams. It uses the beam profile plane as the direction for the bounding box to determine the length, taking into account any trimming/coping/mitering etc. For both of your curved beams the profile plane may be at different ends and therefore the calculated bounding box will be different. I could have used the underlying sketch geometry to calculate the length, but that would not have taken any trimming into account. Ironically, if you create one and mirror it, the results will probably be the same (but still wrong). Best I can do is to test for a curved beam and not report the length so the results won't be reported wrong (or return the length of the sketch arc whichever you think is better). This is all good feedback for when we eventually have built-in weldment capabilities.

Cris_Bowers

@NeilCooke no problem. One of the things my shop guys complained about when I switched over to Onshape was that I wasn't giving them cutlists anymore. It's a step in the right direction though. I really appreciate the work you've done on this.

Konstantin_Sh

NeilCooke said:

It uses the beam profile plane as the direction for the bounding box to determine the length, taking into account any trimming/coping/mitering etc. For both of your curved beams the profile plane may be at different ends and therefore the calculated bounding box will be different. I could have used the underlying sketch geometry to calculate the length, but that would not have taken any trimming into account. Ironically, if you create one and mirror it, the results will probably be the same (but still wrong). Best I can do is to test for a curved beam and not report the length so the results won't be reported wrong (or return the length of the sketch arc whichever you think is better). This is all good feedback for when we eventually have built-in weldment capabilities.

I beleve that taking the length of the beam without trimming is more correct if you not going to make a drawing for each of them, at least it could be cut so that it ends may be adjusted in place after bending, but if you give a shortened length of the bounding box we can't cut of the beam based on it.

NeilCooke

V8 is live.

Here's what I changed:

1. For beams containing arcs, I measure the length of the underlying sketch entity and report that instead, since it should be longer than the trimmed beam. 
2. For merged beams like tubes, I measure the straight sections using a bounding box and curved sections using the sketch entity - if the loop of edges has lines at the start and end, the reported length is very accurate - if one of the ends is an arc it will be more approximate. 
3. I also populated the beam description with the same info.

So in this example the lengths are spot on:



Any approximation is indicated by * so you should probably manually measure those items to be sure.

https://cad.onshape.com/documents/e15c2c668d138f01242d0c80/v/63de3fd6ac09766dfc16a8a0/e/7167292bf121482e9d73044c

Cris_Bowers

In my case I am trimming out a flat plate with some 1" tube. The center section of the tube along the top arc is cut out with a 90 degree "V", which results in two shorter tubes. The method you are using to get the length attaches the full length of the arc used so I end up with two (60.643) tubes. I should actually have two (27.5) tubes. I was using the edges of the plate for my profiles, but that won't work if that is the length I'm going to get.

I first tried to get around that by creating a new sketch on the face of the plate and breaking up the arc into 3 sections. But this resulted in a length that was shorter than needed because the bottom edge of the tube extends further than the top.

In order to get the right length I had to over extend the top of the arc to match where the bottom edge ends up.

NeilCooke

@Cris_Bowers yours was the only scenario I couldn't account for. I did come up with a way to calculate it accurately, but it involves some heavy looping which doesn't work well in FS - just end up with a long spinner. I will keep at it.

Cris_Bowers

I understand. I just wanted to document what I did for anyone else that had the same scenario.