Unable To Join Solid Bodies

[MaxHugen]

I've been trying to create a single solid body, of a complex curved fairing, as follows

1. Lofted sections of sketch = Surface Bodies
   Nb: The entire sketches for the fairing could not be lofted together as "The loft would intersect itself", however individual sections were lofted OK.
2. The Surface Bodies were stitched together = single Stitched Surface Body
3. Thickened each part of stitched Body as F360 could not do all together = individual Solid Bodies
   (Body57 to Body67)
4. Join Bodies - two Bodies won't Join others - Body59 and Body60
5. As per Forum-Unable To Combine ,
   a. turned off Design History, for direct modelling mode
   b. used Inspect -> Validate to check, reports AOK
   c. Re-enabled Design History - and found it was all GONE - and Sketches could not be edited!!!

I've spent many hours researching and trying to find a way to create a 3mm thick fairing from the sketches, and thought I'd gotten there until this last issue.  I have attached the previous version - before I went to Direct Modelling and wiped out the History - in the hope that someone might see what I'm doing wrong.

[davebYYPCU]

Taking your points

1. No mention of body edges and tangency to them. See point 4.

2. Zebra and Curvature map analysis needed about now to check for irregularities.  

3.  If won’t thicken - investigate why, see 2 above. But yes Thicken parts is a work around.

4.  Welcome to the land of streamlining.  Section Analysis, while your surfaces will be adjacent, thickened solids may have minute internal gaps / birds mouthing and or overlaps at various parts of the adjacent / intersecting bodies, overlapping all the way will Combine, gaps birds mouths can’t.

5.  You won’t do that again.

will the stitched outside surface offset inside 3mm.  If so, use boundary fill, or patch the edges and stitch.

To get you thinking, will check the file bit later.

Edit - just read that help page, says nothing about surface problems, and is presuming sketch / timeline errors, but your models haven’t got those.

Might help....

[jeff_strater]

thanks for sharing the model.  But, the model you shared does not look like the one in your description, there is no Body59 and Body60 in that version.

From the images, it could be a case where there is not clear overlap between the two bodies.  If they just meet at a vertex or edge, Combine won't work.

[davebYYPCU]

Checked the file and stands, without the thicken / combine steps.

Back to my

Step 1.  This fairing is built from sketches.  Building from sketches does not allow tangency.

Step 2.  Lumps and bumps are evident, the red sections should be well defined high curvature and the stop start streaking is always going to play havoc with solid Thicken or Shell requests.

I like to think of Thicken as a bit like Sheet Metal, will always build the edge perpendicular to the face.

So referring to these 3 loft panels you have tight concave profile at the front and a ess bend at the rear, somewhere along the the 2 internal edges the thicken edges will birds mouth, and not using Tangency across these panels is a huge source of the problem. (Thicken has to be small panels and they don't combine.)

I have not attempted to offset this at 3mm.

Might help....

[MaxHugen]

Hi Jeff, sorry about the file I sent. After ~16 hours and ending up with lost History, I just scrambled to find a "last good" version with the History.

I've tidied some things up, and 3 sketches are used:  3Dfairing, 3Dfairing2 and 3Dfairing3.  I labelled "panels" I used to Loft, although I found I could Loft a couple together OK ( S2+S3 and S4+S5).  

As @davebYYPCU noted, amongst other issues, is that Thicken is not the right tool to use, so I'm working on using Offset. My initial plan is to create a solid body from surface bodies S1, S2+S3 and S4+S5. An offset (S1-S5 Offset) was created without issue, and Patches added, but Stitch failed.

So I created a new temp component "Offset Test", made a simple 3D sketch using 4 curves, added Offset and Patches, and it Stitched AOK into a solid body. Using colured faces for clarity in this process, I did find some anomalies, with some surfaces that had to be "Reverse Normal"-ised.  I don't know if that affects the final Stitch or not.

So I tried again, with a new Offset (A5 Offset2) and Patches, but it still fails. The surface bodies and patches can be added to Stitch without issue, but adding the Offset always fails.

 Any suggestions please?

[MaxHugen]

Hey Dave, although most Thickened surface bodies Combined OK, I do see that it's the wrong tool - Offset extends perpendicular to the surfaces. So I'm trying that out on part of the model.

Been snowed under with memory issues on my new laptop, an MS service - Desktop Windows Manager - is using up to 2Gb of RAM, and may be causing some graphics probs in F360 as well. Wading through 7 months of similar issues on MS and Intel forums, it seems neither have sorted out this apparent bug.

As for F360, I'd really like to get on top of this Offset technique. Then I have to rethink some of the other issues you pointed out like tangency. 🙂

[davebYYPCU]

Nope, you won’t beat the Offset issue, before you get the Tangencies where you need them.

Check my front fairing, only the first Loft used sketched rails, then it was as little sketch articles as possible.  

Hide the sketches until you need the odd curve here or there, prioritise using Body Edges from previous Loft for the next Loft.  For this work, in the Loft dialogue box controls, Connected is not as helpful as Tangent.  (Edit My Lofts count the word Tangent in each)

Do you have a close up photo for this one?  Can’t remember if there was a canvas.  I have some long distance photos.

Might help....

[MaxHugen]

Hi Dave.

Hope they fix this Forum soon - I get constantly logged out , and to make it worse, when I log in again I don't get returned to where I was.

I have very little useful info about the aft beam and how it attaches to the hull etc. An old pic (2016?) which I didn't save showed a hull being fabricated, and it did not show any flat area for attaching the aft beam, unlike the forward beam.

It seems that they added a section that bolts on to the hull, as a flat area for connecting the beam to.

Most of the attached pics are from races in 2019. The beam looks flatter in 2021 where it attaches to the hull.  You can easily tell the difference as they also added an extension to the hull. Originally, the trailing edge of the beam lined up with the transom, now the hull extends further aft almost 0.5m.

Compounding the difficulty of creating this aft fairing is that I have a hull section there that is crap.  It sticks out topside where it shouldn't, a result of inexperience in "lofting" (drawing) the hull lines properly. So the outer edge of the beam had to be curved, whereas it's clearly not in the pics.

Hopefully, I will become more proficient as I gradually learn CAD techniques!

[MaxHugen]

Forgot to include this pic... shows the general profile of the extra support piece between the beam and the hull. It's a recent pic but doesn't yet have the new transom extension to the hull.