heinzboehmer's 2002 Topaz 6MT Coupe

[bigjae46]

Now, the big question is fatigue. No good way to model this, but the mounts will eventually fatigue because they'll be made of Aluminum. It's just a matter of how fast that happens. Hopefully not until the end of the cars service life!

I think it's time to stop messing around and just get these parts ordered. I'm running analyses with made up loads, which is unlikely to get me anywhere useful. Also, I've been thinking and, surprisingly, I don't think I care if these parts fail? They're dead easy to swap out, the price of manufacturing is super affordable and I've been dying to give metal 3D printing a shot. If they do fail, there's always CNC 7075 or printed Titanium

Carbon fiber bro...

[Obioban]

You know, a friend was just telling me the other day that it's hard for her to comprehend why I would spend so much time designing something just to make it look like I didn't change anything at all. And to be honest, she has an excellent point.

But there's just something I find so cool about the stealthiness of this. It's a significant upgrade over stock, but it's invisible to anyone who doesn't know exactly what to look for.

Same reason why I painted my 996 calipers matte black and went with no decals. Guess my brain is just wired weird

100% agreed.

IMO the best mods are the ones where someone who isn't intimately familiar with the car can't tell what is stock and what is aftermarket.

[heinzboehmer]

Busy week, so slower progress, but I did manage to find the time to get the weight savings features modeled in.

Had to learn some basic blender skills for the gyroid infill version. Hollowing the part out in parametric CAD was easy enough, but modeling the infill proved to be too much. Here's some Blender screenshots:





As cool as this seems, I don't think I'm gonna go forward with this as the final design. Seems like a bit of a pain to post-process and, crucially, I can't run FEA on meshes. I'm super tempted to lump one of these into my real order and run it just as an experiment, but we'll see.

So with the gyroid version off the table, pockets were next. Thankfully these were parametric CAD-able:




Just like before, the analysis on the version with the bosses does not close with the brace-buckling load. And just like before, not concerned. Analysis on the other version looks good.

Weight for the two no bosses, pocketed brackets made out of AlSi10Mg should be around 730g (+ weight of the studs). Stock brackets are 612g all in, so not bad at all. Muuuch better than the 1869g (!) for the sheet metal versions.

Worth pointing out that the pocketed versions are specifically made to be 3D printed. A 5 axis CNC might be able to get in there, but they are not designed for that, so it would likely be expensive. Any manufacturing method that involves a mold will for sure not work, since the draft angles are all wrong. Should be fairly easy to fix the drafts, but I didn't want the extra weight.

Just do pockets. The powder is packed in there and very hard to remove.

Also, stealthy is the way for sure. Cover carbon stuff in paint too

Yeeeah, I was stubborn with the gyroid version, but ultimately reached the same conclusion.

IMHO it's easy to make something that looks different, especially when adding/changing functionality. It's much trickier generally to add functionality while keeping the same packaging. That's exactly what I admire so much about this solution, as you say, unless someone was really looking they'd never notice it wasn't original, to do that, while also achieving the necessary functional requirements, is very cool.

I ordered a pair of second hand Z4 braces today

Hell yeah! That's exciting. I'll get these 3D printed brackets ordered and stress tested very soon.

It's great work. I was just curious as functionally it's fairly simple part but becomes quite complex with the OE cast (off chance that it's forged? I haven't looked that closely) design.

They look cast. Funnily enough, getting the angles and interfaces between the two brace pickup points correct was more work than recreating the stock bracket geometry.

Carbon fiber bro...

You know, I think this would be a good use case for the material. I'm still a bit worried about it shattering under high impacts, but that might not be a real concern with the load that these will see.

That being said, I really want to try the 3D printed metal thing, so gonna try aluminum first. And as mentioned earlier, I'm a sucker for maximum OE-ness and the texture of the 3D printed parts should match the OE parts quite nicely.

[Obioban]

Even if you go through with an infill, I’d think not gyroid? It’s optimized around strength while not shaking the printer— which is of no value here.

[bigjae46]

You know, I think this would be a good use case for the material. I'm still a bit worried about it shattering under high impacts, but that might not be a real concern with the load that these will see.

That being said, I really want to try the 3D printed metal thing, so gonna try aluminum first. And as mentioned earlier, I'm a sucker for maximum OE-ness and the texture of the 3D printed parts should match the OE parts quite nicely.

Carbon fiber in general has better material toughness than aluminum...not sure about forged carbon. Send me a set of 3D prints and I can make some! Would need a base piece that the part will sit in, has a flat bottom with a slight draft angle on the side to aid in release.

Or let me know how much chopped carbon you need to DIY it.

[heinzboehmer]

Even if you go through with an infill, I’d think not gyroid? It’s optimized around strength while not shaking the printer— which is of no value here.

I choose a TPMS surface as a proof of concept because the negative volume in there is continuous, which means the powder can be removed post-print. Specifically went with gyroid because it's used everywhere in FDM printing and the wide availability of 3D models online (even the wikipedia article has one) made it easy to implement in Blender.

If I were to actually go this route, I would likely choose something more like a skeletal, D type TPMS surface. Stiffness to weight is better than gyroid and it uses slightly less material for the same volume. Skeletal would make post processing easier as well.

There's other surfaces with continuous negative volumes, but the TPMS surfaces seem to have attracted the most research for use in materials (NASA has done a bunch), so there's less uncertainty with them. This route involves sitting down and figuring out how to mathematically generate the surface in Blender/Python though, sooo I didn't spend any time on it

[heinzboehmer]

Carbon fiber in general has better material toughness than aluminum...not sure about forged carbon. Send me a set of 3D prints and I can make some! Would need a base piece that the part will sit in, has a flat bottom with a slight draft angle on the side to aid in release.

Or let me know how much chopped carbon you need to DIY it.

Down! You think the pocketed version would lend itself well to being made in CF? Not sure if the features are too small/pointy to get a nice part from. I can add a small draft angle to the ribs to aid in mold release, but maybe it's simpler for you to work with the non-pocketed part.

[bigjae46]

Down! You think the pocketed version would lend itself well to being made in CF? Not sure if the features are too small/pointy to get a nice part from. I can add a small draft angle to the ribs to aid in mold release, but maybe it's simpler for you to work with the non-pocketed part.

The pockets just need to be less than 90 degrees to aid in the release. Maybe 87-88 degrees? It is a compression mold so the pockets are fine except for the smallest pocket near the bolt hole. I can still work around it but that one would be a pain.

[Bry5on]

3 degrees is the good rule of thumb for mold release draft angles. It’ll be cool to see both options! You’re making me want to swap out the Slon brace to make maintenance easier.

[Slideways]

Never knew about draft angles until I watched this video:

[heinzboehmer]

Never knew about draft angles until I watched this video:

That's exactly the scenario I want to avoid

3 degrees is the good rule of thumb for mold release draft angles. It’ll be cool to see both options! You’re making me want to swap out the Slon brace to make maintenance easier.

Hell yeah! It is quite nice to be able to remove only a few components to get to the engine. Still a bit finicky to deal with the cabin air filter plastics, but I don't think there's a good way around that with how tightly everything is packaged back there.

Removing the Slon firewall mount point sounds like a huuuuge pain though.

The pockets just need to be less than 90 degrees to aid in the release. Maybe 87-88 degrees? It is a compression mold so the pockets are fine except for the smallest pocket near the bolt hole. I can still work around it but that one would be a pain.

Had a go at this and I think all of the pockets on the brace pickup points are gonna have to go for the carbon version. It's kinda hard to show in pictures, but the issue comes from the fact that the three mounting faces are at completely different angles to one another.

Hopefully this is appreciable in the following pics of the 3D printed aluminum version.

Viewing normal to the strut tower mounting surface:



Viewing normal to the stock strut bar mounting surface:



Viewing normal to the E86 brace mounting surface:



For the CF version, I modeled the pockets so that they're all normal to the strut tower mounting surface, since I think that makes the most sense. Unfortunately, after making the pockets at the brace pickup point normal to the strut tower surface AND modifying them so that there's still enough wall thickness around the fastener holes, they become too small to be practical for manufacturing.

I guess I could make ALL the pockets normal to some other surface (e.g. the stock bar mounting surface), but probably not worth the hassle. The volume of the pockets around the brace pickup points in the aluminum versions is ~15 cm3, which would roughly translate to tens of grams in CF. I bet that would be even less once you take into account the now-crooked pockets in the strut tower mounting surface.

Anyway, here's what I came up with for CF. 3 deg draft angle added to all the surfaces that are normal to the strut tower mounting surface (hard to tell, but trust me, the angles are there):



Let me know what you think of the manufacturability of the design bigjae46

Incidentally, I noticed this small oversight in the design I sent out to manufacture:



Min wall thickness there is 0.6 mm, which is not ideal. Fortunately the orthogonal surfaces are more than thick enough, so it doesn't actually impact the strength of the part:



I've since fixed this problem area in the CAD for both the aluminum and CF versions.

I did also try reaching out to the manufacturer to get the fixed version manufactured instead, but I was too late. Oh well. Again, strength of the part isn't affected, so as long as it prints out nicely (i.e. no holes), I'll be happy.

[bigjae46]

Let me know what you think of the manufacturability of the design bigjae46

The three pockets around the bolt holes would need to go. The threaded insert may not fit and probably want some material there to take up anu torque stresses. The others aren't an issue as far as releasing the part.

[heinzboehmer]

The three pockets around the bolt holes would need to go. The threaded insert may not fit and probably want some material there to take up anu torque stresses. The others aren't an issue as far as releasing the part.

Yep! Those three pockets are gone in the carbon version.

But cool, I'll get these printed in plastic and sent over to you. No rush or anything, just think it's cool to have some material choices.