heinzboehmer's 2002 Topaz 6MT Coupe

[heinzboehmer]

Clearances looking good, ready for a test fit:

And finally, ran some quotes and the one that made the most sense was right around $950 all in for 6061-T6 (materials, labor, taxes, shipping, etc.) and $1000 for 7075-T6. $50 extra for 7075? Uh, yes please!

Couple things I should also mention:

Some fab shops (the type that seem to sponsor every single electronics-related youtube video) are way, WAY cheaper than others. Like $300 for everything, it's insane. However, none of these cheap vendors specify T6 for their materials, so who knows what they're using. I think I'm not gonna risk it and instead I'll go with a vendor that tells me the exact material composition of what I'll be getting.

Also, I ran some quick quotes to compare cost for the laser cut + machined version of the windshield piece I briefly discussed above. Bottom line is that I don't think it's worth it, unless you're good at welding aluminum.

For 7075-T6:

• Machined in one piece: $641.42
• Multiple pieces: $409.49 (machined main body) + $40.25 (laser cut bottom sheet) + ~$100 (welding + machining bottom flat labor) = $549.74

So roughly $100 less for the windshield piece if you go the multiple piece route. Even less if you use a cheaper vendor for the machining. I'm personally gonna stick to the fully machined version. Yes, it seems a little crazy to carve away that much material for the flat part, but the extra overhead involved with the other approach is not worth the $100 saved for me.

All this is to say that you could technically get this all made for <$300, but I'm not going to.

[heinzboehmer]

Did what George Hill suggested and ran the analysis on the earliest version possible. Note that this version is actually fairly recent, but unfortunately, I hadn't modeled the brace attachment point in any previous versions. Previous ones did not have the fasteners in line with the load paths, so they would have definitely performed worse.

I'm not gonna show all the cases like I did before cause it's gonna cause a bunch more clutter. Instead, I'll show the worst one, which turned out to be the "one tension, one compression" case. Material was set to 6061-T6 for all of the following.

Stresses:




Piece that attaches to the windshield is still alright, which isn't super surprising, as this was one of the overkill versions. The other piece is definitely not alright, though! Max stress is higher than the yield strength of 6061.

Safety factor:




Same story here, windshield piece is fine, attachment point is not. 0.87 is not good.

Definitely cool to see the progress!

[heinzboehmer]

Got the full SimSolid analysis going to check the stresses on the adhesive. Largest load was with both braces in either tension or compression:



Since everything is either gonna be steel or aluminum, I think I'm gonna use 3M 07333. Datasheet says that ultimate tensile strength of the adhesive is 35 MPa, so we're good! Will run this latest design through all my mechanical engineer friends just as a final check, but it's looking pretty final!

[Bry5on]

Clearances looking good, ready for a test fit:

Couple things I should also mention:

Some fab shops (the type that seem to sponsor every single electronics-related youtube video) are way, WAY cheaper than others. Like $300 for everything, it's insane. However, none of these cheap vendors specify T6 for their materials, so who knows what they're using. I think I'm not gonna risk it and instead I'll go with a vendor that tells me the exact material composition of what I'll be getting.

Also, I ran some quick quotes to compare cost for the laser cut + machined version of the windshield piece I briefly discussed above. Bottom line is that I don't think it's worth it, unless you're good at welding aluminum.

For 7075-T6:

• Machined in one piece: $641.42
• Multiple pieces: $409.49 (machined main body) + $40.25 (laser cut bottom sheet) + ~$100 (welding + machining bottom flat labor) = $549.74

So roughly $100 less for the windshield piece if you go the multiple piece route. Even less if you use a cheaper vendor for the machining. I'm personally gonna stick to the fully machined version. Yes, it seems a little crazy to carve away that much material for the flat part, but the extra overhead involved with the other approach is not worth the $100 saved for me.

All this is to say that you could technically get this all made for <$300, but I'm not going to.

Important to note that if you’re going the weld route, absolutely do not use 7075, as it is prone to cracking in the heat affected zone! 6061 for welded, 7075 or 6061 for machined. Agree that the small premium for large piece of mind (strength) is worth it here. You could also machine this from various steels and thin the walls to mitigate some of the mass hit. But the machined design should be mega stiff so there’s not much benefit to the stiffer steel material.

[heinzboehmer]

Important to note that if you’re going the weld route, absolutely do not use 7075, as it is prone to cracking in the heat affected zone! 6061 for welded, 7075 or 6061 for machined. Agree that the small premium for large piece of mind (strength) is worth it here. You could also machine this from various steels and thin the walls to mitigate some of the mass hit. But the machined design should be mega stiff so there’s not much benefit to the stiffer steel material.

Oh that's good to know! I wasn't aware that 7075 was prone to cracking when welded.

[heinzboehmer]

Last piece of the validation puzzle arrived:



Time to run through an entire mock install, just to make sure there isn't some small detail I overlooked.

Started by printing exactly what is going to be sent out to be machined and installing the timeserts:




Car then got disassembled and piece installed:




I also printed out a cutaway of the piece to check (for the last time) that the profile of my design matches the chassis:



Everything seems to check out!

I forgot to take pics, but I also validated that there's enough tool clearance for the cap head screws that hold the brace attachment point to the main windshield piece. Regular hex bits are a little long, but stubby ones fit just fine!

Was a bit nerve-racking, but this thing has officially been sent out to be machined! ETA is about 30 days, which gives me enough time to finish up the strut tower brackets.

Also worth mentioning that I did a couple of small maintenance items while the car was taken apart:

1. Replaced the main throttle actuator pull rod just for preventative maintenance reasons.
2. Found a leak coming from the timing chain tensioner, so swapped that crush washer and retorqued. Was as enjoyable of a job as it always it. Let's see how long it takes before it starts leaking again.

[maupineda]

Heinz, are the studs and nesting surface at an angle? Never noticed the bars ends were slanted. Looking good!

[heinzboehmer]

Heinz, are the studs and nesting surface at an angle? Never noticed the bars ends were slanted. Looking good!

Yep, they are. The bars themselves are totally flat, but the mounting point is angled to tuck the bars up as much as possible. There wouldn't be enough clearance for tools and the moment trying to rip the mount out would be greater if it wasn't angled. Makes the machining more expensive, but no good way around that, unfortunately.

[bigjae46]

Just opened the box



I am trying to get the mold done before the end of the month. I’m going to mold in threaded inserts - a first for me. I’m excited to get this done!

[heinzboehmer]

Hell yeah! Excited to see the process

[YoitsTmac]

This thread is epic. I can't wait to seee the progress on this bar modification - I would totally follow in your footsteps. Even just the firewall mount opens a lot of doors. A quiet thought of mine is you could probably repurpose the hardware on this kit to get carbon fiber mountings. I don't know the dimensions of that piece, but if you did a custom top mount, you'd have a lof of flexibility. you can likely replace the ends with whatever you wanted and just use the carbon tubes. Just a thought. Subbed!

[heinzboehmer]

A quiet thought of mine is you could probably repurpose the hardware on this kit to get carbon fiber mountings.

While I wasn't aware of that specific kit, I did consider similar approaches. This project would have been way simpler had I just made my own braces or adapted aftermarket ones, but I really want to keep the crash safety aspect of the OE E86 braces. I don't trust that aftermarket pieces were put through rigorous crash testing and definitely don't have the skills/resources to do that with a design of my own. E86 uses the same suspension design as E46, so that gives me confidence that the forces seen by these braces will be similar on both chassis.

Chassis rigidity is cool and all, but hitting a wall head on and ending up with a brace going through my chest (or worse, a passenger's chest) would really, really suck.

Also, custom strut bar end brackets are coming! I need to machine the stock cast aluminum pieces first before starting on the design, but I'm still waiting on the certification class to get access to the machine shop at work. Those pieces are gonna be designed for laser cutting + bending, so manufacturing should be very quick.

[George Hill]

Chassis rigidity is cool and all, but hitting a wall head on and ending up with a brace going through my chest (or worse, a passenger's chest) would really, really suck.

If the bar moves that far to impale your chest I think you are dead either way lol

But I appreciate your concern.

[heinzboehmer]

If the bar moves that far to impale your chest I think you are dead either way lol

But I appreciate your concern.

Ha, you're almost certainly right and it's very likely that I'm being overly cautious about an improbable scenario. However, I really hate the idea of having something I designed injure someone, so playing it as safe as possible.

I'm particularly scared of a front overlap crash. Note how much the strut towers can move in one of those scenarios, especially compared to how little (relatively) the passenger compartment deforms:

(yes, I spent way too long looking at crashed E46s at the start of this project )

[George Hill]

OH I know, I did this a long time ago. Walked away without a scratch surprisingly. The engine moved back far enough the starter broke off the block. I think the strut tower would have to be past the firewall to be an issue and I don't think that is likely a survivable crash, but maybe? IDK. Either way, I appreciate that it is a concern for you.

\