Folks,

I'd really appreciate some metallurgy input on the fastener pictured below. It's a Rolok self threading fastener. The application is a bus basement bulkhead. The bulkhead is at the rear of the basement and is made up of 1.5"x1.5"x0.125" thin wall box tubing and 14 gauge sheet steel. At the base of the bulkhead there's a 2"x2"x0.25" angle iron member running transversely. There are 15 Rolok bolts clamping the angle iron, at the base of the bulkhead to the bus basement floor. The bulkhead serves as a fixture for the drive axle articulating control arm pickup joints. We're experiencing shear of these bolts. The buses have been in service for 10 years or more, but with infrequent use, typically less than 150K miles. Thanks.

It appears there's trouble with the link to the image. Here's a photo album of the image: https://picasaweb.google.com/dmb2000/RolokBolt?aut...

The fracture surface of the bolt shown is very rusty and at this point cannot be commented on. You should first try to look at the fracture as it is under a stereobinocular microscope using both a ring and an adjustable fiberoptic light to see whatever you can and document it. Then put the bolt in an SEM and get semi-quantitative composition by EDS analysis on the possibility (and I consider it a low one) that contaminants relevant to the fracture are present. After you have done these steps, use a mild cleaning agent to clean the fracture and reexamine it under the scope. This will tell you what kind of fracture mode (fatigue, overload, intergranular fracture, etc.) you have and will provide you with lots of clues as to how the fracture occurred. The next steps will likely be based on that optical examination but will typically include general characterization of the bolt to determine if it was what it was supposed to be (hardness and composition are most common). You also will likely need to cross-section through the bolt fracture for structure and to see condition of threads adjacent to fracture.

Good luck!

I would suggest taking some of these assemblies completely apart and replacing all of the fasteners (carefully checking that they are tightened correctly).
You can look at bolts that have not broken for signs of the start of failure.
Bolts could be over tightened leading to tensile failure.
They could be under tightened causing either fatigue or overload of the remaining bolts.
Or it could be a corrosion fatigue mechanism.
Only partially failed bolts will tell you for sure.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

I wouldn't be too concerned as to why they are failing.

They are breaking and therefore not fit for purpose.

You say that they have infrequent use - only 150k miles!!!! What is the design life you expect?

You can carry out as much analsyis as you like but they will still shear if they are not strong enough.

It may be difficult and costly to prove but stress corrosion fatigue probably has a hand in these failures.

As they hold in drive axles the application does seem somewhat 'Safety Critical'

I suggest you use either larger screws or rivnuts and bolts.

I would also think that an A4 Marine Grade fastener might not be a bad plan.

The alternative is that if they have only failed after 10 years replace them every 5 years.

I think it is very important to know why these bolts are failing. After all, failure has the potential for causing loss of control of the bus and having the potential for injury. This fastener system is intended for one-time installation. Bolts should never need replacement prior to end of component life provided they were properly installed in a properly designed system with an appropriate bolting material utilized.

I have a couple of additional comments on such an investigation. First, I neglected to mention performing SEM fractographic examination after cleaning surfaces. This step is probably not necessary if the mechanism turns out to be fatigue as you should have all the clues you need from optical examination. The other point is the necessity to attempt to identify which bolt failed first if several have already broken in the assembly. The reason is the first bolt failure will automatically increase static applied load of the remaining bolts (and increase amplitude of cyclic loading if that applies), making those bolts more susceptible to subsequent failure. You will need to compare all of the fracture surfaces with each other to find that first failed bolt; that will be the one you need to subsequently focus on and characterize. (A recent example for me involved an assembly with 8 bolts; my ground zero bolt exhibited reverse bending fatigue and almost no final fracture area, indicating it had been insufficiently tightened).

As for corrosion: you should ask yourself if bolts are supposed to see a wet corrosive environment. If you determine mode of fracture as fatigue, you will also want to determine how much, if any, effect corrosion would have had. Just because the mechanism is technically corrosion-assisted fatigue does not mean failure occurred primarily from corrosion exposure.

Thanks guys, given the seriousness of this issue I now realize that I need to spend the bucks for a profession consultant. Thanks again, David

Have you pursued this further? We're having similar problems with Roloks in the same application.