Subframe Dropped and cracks found

[VinceSE2]

It’s actually TiAg, but the lighting does make it look silver grey.

I used an epoxy paint that is super close to the color of the factory e-coat.
First the epoxy paint, then seam sealer, the epoxy paint again, and finally I tattle canned it TiAG because it was already rattle canned TiAg before I installed the VinceBar kit. Otherwise I would have strived to replicate the e-coat with slight overspray of TiAg.

Towards the end in this folder you can find lots of pictures of the surface treatment process

https://www.dropbox.com/sh/wvpyxrfuna41216/AADqQnvifiUH71JhXlGAB3L7a?dl=0

Sent from my iPhone using Tapatalk

[pawelgawel]

all you guys going that extra mile and do the bars up front and rear with plates... impressive, honestly! You guys must be tracking and beating on these cars like no tomorrow... I can't imagine having issues after doing the plates and the welds up top with structural foam in the cavities for normal spirited driving once in a while. I think, in reality, the car will be driven to the ground before the floor rips after having that done.

[VinceSE2]

all you guys going that extra mile and do the bars up front and rear with plates... impressive, honestly! You guys must be tracking and beating on these cars like no tomorrow... I can't imagine having issues after doing the plates and the welds up top with structural foam in the cavities for normal spirited driving once in a while. I think, in reality, the car will be driven to the ground before the floor rips after having that done.

You should absolutely do what you feel is right, but the issue is NOT limited to track cars or massive beating. It’s a matter of time before the damage occurs, and how you drive it determines the time. A jerky driving style where you are frequently and abruptly on and off the throttle kills the RACP much sooner than a smooth driving style.

The damage is the result of metal fatigue due to a design that allows for way too much flexing. The flexing is in turn the result of a design flaw (production/cost related I believe) that has the floor panel (RACP) very poorly connected to the chassis legs. The load path from the subframe to the chassis legs is long and passes through several weak/hot spots.

The good news is that it’s fairly easy to address and eliminate this design flaw by coupling the subframe mounts directly to the chassis legs. This is however NOT done with just plats and structural foam (the foam was BTW just a cheap fix BMW did to get out of the class action law suit in 2008/2009).

To pretty much unload the RACP some sort of cross bar is needed. And at the top of the front mounts there needs to be both vertical and lateral additional support.

That said, I agree that there are some solutions that are crazy over dimensioned, and way more clunky and space claiming than necessary.

Please have a look at these videos (particularly the first one) for more details.

https://youtube.com/playlist?list=PLg2TArrteKd4ttsu_Nr1SznnEw9N81cN7

Sent from my iPhone using Tapatalk

[VinceSE2]

Following up on the previous post, for anyone wanting this in text and pictures rather than the videos.

I’d like to share a few pictures that I believe very clearly illustrates the underlying design flaw that causes the dreaded so called ”subframe failure” on the E46’s.

In the first picture I’ve exposed the topside of the tear left subframe mount. And this view of it shows a couple of the key spot weld that have “popped”.

The aforementioned spot welds are to the left and right of the subframe mount (which in turn has the “T” cut and partly flattened on top of it) The popped spot welds have been drilled out in this picture.

In the second picture you can see how the load (the downwards pointing arrow) is transferred to the chassis leg (the thin red lines), and what a detour that path is.
Furthermore the last part of that path is taken through a single layer of sheet metal.

Needless to say there’s a lot of flexing going on there, and the RACP i unnecessarily exposed to a load that could/should have gone directly to the chassis leg where theres just a large gap. (This gap is BTW closed by the end plates of the VinceBar)

So, the load is transferred to those poor spot welds in front, and behind, the subframe mount. The front ones are also just connected to the inner wall of the chassis leg.
The outer wall of the chassis leg (the wheel well) is reached via the part of the RACP that runs in under the inner wall of the RACP as seen in the third and forth pictures.

In those last pictures please note how the RACP profile slats not only downwards, but also upwards (look at the left side) ending in a (you guessed it) single layer of sheet metal thats spot welded to the wheel well. Again resulting in a fair amount of flexing.
This is where the separation shown in the fifth picture occurs.








Sent from my iPhone using Tapatalk

[VinceSE2]

And here are two sketches that I made back in 2015 when originally explaining the issue and the VinceBar design. This is viewing the left rear subframe from the back, looking forward





Sent from my iPhone using Tapatalk

[pawelgawel]

I understand the issues and theory behind all of this... if i had the space and time to let the car sit on jacks for weeks, Id probably do this as well just for the sake of doing it and knowing I have a "brace" vs anything else. In all honesty, what is the life span of the structural foam? if done properly, it fills in the cavity and joints both sheets and the sides together creating a single solid entity. How strong is this in long term? has anyone recorded failure after doing this properly? and what type of abuse did it go through?

[poss]

I understand the issues and theory behind all of this... if i had the space and time to let the car sit on jacks for weeks, Id probably do this as well just for the sake of doing it and knowing I have a "brace" vs anything else. In all honesty, what is the life span of the structural foam? if done properly, it fills in the cavity and joints both sheets and the sides together creating a single solid entity. How strong is this in long term? has anyone recorded failure after doing this properly? and what type of abuse did it go through?

My understanding is the foam goes in the cavity around the gusset - which would deal (at least to some extent) with the common primary failure point around the gusset (see Hot Spot No 1 in Vince's diagram), but would not address the secondary failure point (Hot Spot #2) where the RACP panels join together before attaching to the wheel arches (an inherently weak design that allows the RACP to flex up and down). As you can see from Vince's sketches, the Vincebar concept addresses that weak design by connecting the RACP directly to the rail before the top and bottom sheets join together.

In short, the foam only delays RACP failure IMHO. I wouldn't go near it - particularly as it makes proper topside reinforcement more difficult (the foam puts off toxic fumes when welding). I strongly suspect it was simply the cheapest way for BMW to make an RACP with minor cracks last out the 10 year period agreed to when they settled the US class action.

[VinceSE2]

I understand the issues and theory behind all of this... if i had the space and time to let the car sit on jacks for weeks, Id probably do this as well just for the sake of doing it and knowing I have a "brace" vs anything else. In all honesty, what is the life span of the structural foam? if done properly, it fills in the cavity and joints both sheets and the sides together creating a single solid entity. How strong is this in long term? has anyone recorded failure after doing this properly? and what type of abuse did it go through?

As Poss explains in his response above, the structural foam does nothing to address the underlying design flaw. It’s simply a bandaid (similar to just welding on traditional plates) that addresses one of the symptoms of the design flaw.

Simply put, the RACP will still pull away from the rest of the chassis when foamed. So the lifespan of the foam really is not of any great significance.

The foam was a cheap way for BMW to push the problem in front of them long enough to get out of the class action lawsuit. Nothing more, nothing less.


Sent from my iPhone using Tapatalk

[cobra]

I would like to see a solution that can be added without making such extreme modifications to the existing sheet metal, other than maybe a few drill holes. Is it possible?

[poss]

I would like to see a solution that can be added without making such extreme modifications to the existing sheet metal, other than maybe a few drill holes. Is it possible?

You could try pumping epoxy foam into every cavity - but I wouldn't recommend it

I wouldn't class this as an extreme mod. It's less effort than doing the bottom-side reinforcement - you're only cutting away a strip of non-structural boot floor sheet metal to fit the bar in. If you use Vince's epoxy solution it only requires a few drill holes and rivets plus glue after that.

If you want the car to last this is a mandatory op unfortunately. As both Vince and CMP say, its not a matter of if but when...

[D-O]

I have received my rivet/epoxy VinceBar and VinceSkinz. I am now gathering the rest of the parts to refresh the entire rear suspension while it is apart.

The previous owner did the BMW foam injection. How much trouble is the foam going to cause for me when installing the bar and skinz?

Jesse

[heinzboehmer]

I have received my rivet/epoxy VinceBar and VinceSkinz. I am now gathering the rest of the parts to refresh the entire rear suspension while it is apart.

The previous owner did the BMW foam injection. How much trouble is the foam going to cause for me when installing the bar and skinz?

Jesse

Not much. Just don't light it on fire when cutting up the trunk floor.

[pawelgawel]

As Poss explains in his response above, the structural foam does nothing to address the underlying design flaw. It’s simply a bandaid (similar to just welding on traditional plates) that addresses one of the symptoms of the design flaw.

Simply put, the RACP will still pull away from the rest of the chassis when foamed. So the lifespan of the foam really is not of any great significance.

The foam was a cheap way for BMW to push the problem in front of them long enough to get out of the class action lawsuit. Nothing more, nothing less.


Sent from my iPhone using Tapatalk

Mybe Im playing devil's advocate here, id like to see the extent of damage after plates are done and foam. I haven't been able to pull up any examples mysefd, maybe someone can post some links?

[martin.oconnor]

Mybe Im playing devil's advocate here, id like to see the extent of damage after plates are done and foam. I haven't been able to pull up any examples mysefd, maybe someone can post some links?

I'm also deeply skeptical of the need for such elaborate solutions. When doing my car I decided to foam all four mounts and leave it at that. (You can find my write-up here: https://nam3forum.com/forums/filedata/fetch?id=78463.)

I do not think any post-foam mount failures have been reported.

However, the theory seems to be that foaming the mounts simply moves the failure points to the RACP-chassis attachment points - and that the RACP will eventually separate from the chassis. I could not find any convincing evidence that this happens with any regularity in the real world.

[lemoose]

Mybe Im playing devil's advocate here, id like to see the extent of damage after plates are done and foam. I haven't been able to pull up any examples mysefd, maybe someone can post some links?

I had the welds in my wheel arches completely fail on me after installing plates and the entire RACP was pulling itself out. the root cause has already been well explained to you so i have no idea how anyone would arrive to the conclusion that plates address the issue but you do you boo