NDT for Alum Banana Arms

304065 · 07-08-2003, 06:50 AM

I just got a set of Al banana arms from a '74, going to convert my '71 over.

What is the best testing process to check for cracks? Magnaflux? Doesn't that only work on ferrous metals? Zyglo?

Also, I remember reading an article, I think in Excellence, that said that one of the european shops had the banana arms polished. As they are now, the surface is pretty rough with casting lines, does anyone have a view on this? I'm all about removing stress raisers.

Bonus question: who's the best shop to do such a procedure, on the east coast, NJ preferred?

724doorE · 07-08-2003, 08:07 AM

John, by polishing the surface you are minimizing the possibility of surface cracks forming by reducing the number of imperfections available. You are smoothing out the crevices on teh surface that could under fatigue stresses turn into cracks and propagate. You are also subjecting the material surface to a condition such that it maintains a residual comressive stress. This also makes it harder for cracks to develop.
Polish them and clearcoat, they will look great!

sammyg2 · 07-08-2003, 08:54 AM

No magneflux un non-metalic metals, doan worky.

I'd suggest you get a 3 part dye penetrant kit. They are relatively cheap. Comes with cleaner, dye, and developer all in rattle cans. takes 10 minutes.

304065 · 07-08-2003, 10:22 AM

Good suggestion. I just ordered a Magneflux Spotcheck system from Aircraft Spruce. I guess this is a three part system, clean, shoot on the dye which goes into cracks, then developer that makes the cracks show up as red lines on a white background. Neat stuff.

Jim Sims · 07-08-2003, 11:10 AM

If you find cracks, then what will you do? Attempt a repair or scrap the arms? Jim

724doorE · 07-08-2003, 11:38 AM

track their progress

Seriously, John, if you were close, we have some research going on with ultrasonic non-linear crack detection using both high and low frequency waves, nondestructive of course, It works great although we can't yet triangulate the position of the flaw. Ad this method does ignore flaws such as voids or porosity and only finds cracks.

Has anyone ever seen one of those al arms actually break?(before the wreck!)

Paul Thomas · 07-08-2003, 12:31 PM

I saw a pic of some Al arms polished on Gunnar racings website. It looked pretty darn cool.

Paul

304065 · 07-08-2003, 12:35 PM

If I find cracks I will junk the arm.

Dennis, that sounds like a cool process. I considered other forms of NDT like eddy current or x-ray, but I should be able to detect pretty small cracks. These arms came out of a running, straight car, but you never know: I also plan to check the geometry BEFORE I put them on the car.

I've never seen them break, they look pretty stout, but not knowing the specific history I'm inclined to check them while I can. On the steel bananas with rubber bushings I've seen 1.3g on the data logger (a number that will be probably be poo-poohed by the big fender crowd) so want to make sure they are up to snuff.