Any metallurgists on here? Input about breakage

[BluDestiny]

So this is going in this board because this component was actually on my subaru. This is a rear camber arm, specifically the drier side. I bought the arms in march/april, so it's only been about 6-7 months.

 

I have done 4 track days on them and several autox sessions. Also about 8k road miles going to a from the track and around town. This break happened when I was going up a friends driveway where I lifted the inner passenger rear wheel (all the weight then going on my driver rear wheel).

 

The owner called me about 30 min after I submitted an email to them about the snap. He said he's been seeing a lot of problems specifically with that batch and he is thinking it's from an improper heat treating cycle. I believe he said this is chromoly. What he has done is send me one that has not been heat treated yet (just so I can get to and from work since this is my warm car), and will be sending out another pair once he gets tham back from heat treating. So all in all I'm satisfied with the customer service. 

 

This picture shows how it snapped on the car

 

and here are some closeups. I have sent the part back after I got a new part

 

The break happened right at where the lock nut was set, and all the lock nuts were tight as I have just checked them 3 weeks ago after my last track day. Again, I'm just looking for input as to whether this part was improperly abused (which means I'll need to upgrade) or if the design is not optimal (it does have a longer adjustment bolt than most other kits which I think puts more stress on that specific area)

[NewZed]

Looks like a terrible design.  A bolt ot threaded rod, essentially designed for tensile loads, used to handle a side load, with about 20 stress risers to choose from for a crack to start,  Once the crack starts fatigue will kill it eventually.  You can actually see the rusty spot where things started.

[Leon]

Wow, that's scary. Certainly a design flaw, that adjuster is undersized for the application.

[Chris Duncan]

Looks like the design is flawed, the material is weak and possible manufacturing problems

 

this piece should be an off the shelf rod end, one on the stronger more expensive end of the range. Probably a 5/8" dia thread at least.

 

It looks like a fabricated piece and it's too long so it has too much leverage.

 

If you use a rod end threaded straight into the end of the arm then it's hard to adjust so it looks like they made this one with opposite threads to ease adjustment.

 

Normally this is done with a opposite threaded sleeve and regular rod end so it doesn't have too much length and thus leverage.

[74_5.0L_Z]

If you look at the one on the opposite side of the car, I'll bet it is bent and may already be cracked and ready to fail.

 

Total crap design.

[BluDestiny]

Yeah the other will be coming out as soon as I get some heat treated replacements. Hopefully in two weeks or so. Just going to and from work on it now so I can't complain. 

 

My future option would be The cusco version. It has a much shorter shank on it. Only issue I have is that it has a pillowball/heim joint which knocks me up a class. These arms were the only ones I could find with that much adjustment with a urethane bushing to Keep me in street prepared. 

 

Edited November 1, 2016 by BluDestiny

[JMortensen]

We use threads in bending on Z's all the time. Look at just about any adjustable control arm, whether it be Arizona Z Car, Modern Motorsports, etc, they all have a threaded rod end that gets loaded in bending. Many of them use that kind of adjuster as well. Mike Kelly's old arms used that kind of turnbuckle. While it isn't ideal, I don't agree that it's a crap design or that it is destined to fail. I've got them all over my own suspension too. I do think there is something to be said for not having a lot of threads exposed, which is why I really didn't like Ross's rear arms for example, but he was using 7/8" threads IIRC, so I think his defense was despite the long section of threads in bending, the material was thick enough to take it.

 

One difference in this case is the poly bushings. The poly bushings cause a heck of a lot of stiction, and that combined with the threads in bending might be more problematic.

[NewZed]

I knew that somebody would bring that up.  That general design is ceryainly everywhere.  The location of the ARB end link seems closer to the stress point though in this case, than a comparable Z car arm.  Pulling and pushing more on the arm itself than the wheel hub.  The leverage factor is higher.

[260DET]

The OE design which loads up the middle of the arm and so tends to bend it is the starting problem, that means everything else has to be bigger than we are used to with Z cars. Just use best quality rod ends, check them after every race day and if there are still problems go up in size. It's a overall shit design suspension, you should stick with Neesans 

[BluDestiny]

Well my datsuns still outnumber my Subaru 2:1. They just aren't as comfy. 

[260DET]

My Cherokee diesel tow car is comfy, that's it, Z cars are for racing stuff.

[Xnke]

Yeah, that fastener was cracked pretty deeply for a long time prior to the break-the steel being used is NOT adequate for the job.

 

I'll bet it's 4140, and hardened up to about 40RC, which puts it into a brittle temper zone. Should be down to 34-36RC and my god, the threads...they're single-point cut but you can see where it's been cut with a manual threading lathe, and the safety net cut wasn't made deep enough. The threading tool was ground from a HSS toolbit, but the root radius wasn't cut large enough for this kind of loading. Also the shank of the fastener here, where you can see the threading was turned off to match the safety cut, needs to be cut smaller and with a larger radius in the corners, you want to completely clean up the thread root and do it with as large a radius as you can cut without chatter-chatter marks would simply induce the same stress risers you're trying to remove by undercutting the thread root.

 

There is evidence of quench cracking on the face in at least three places, fatigue damage over about 40% of the face, so yeah, his heat treatment is way off here.

 

Borrow/beg/buy a dye penetrant crack testing kit and test your newly heat-treated parts when they come in...if they fail, don't use them. Send them back with photos of the test and recommend he figure it out before he keeps selling dangerous parts.

 

I'll try to get your photos edited with crack progression markers so you can see how and where it started and ended, and it'll be pretty clear what happened.

Edited November 19, 2016 by Xnke