Fretting Corrosion: Resistance of Materials & Accelerated Life Test

[B.delong]

Fretting Corrosion

I am researching a fretting corrosion problem and I am looking for resources on the relative resistance of materials to this phenomenon as the material on this subject seems very limited. The application is a rolling ball switch, which impacts the tube sidewall due to vibration. The materials must be conductive, able to withstand the stress of the vibration impacts, yet minimize oxides breaking off which cause abrasion.

The historical metal used in this application was brass. Calculations using a Hertzian model indicate the yield strength is being exceeded and the formed oxide particles are causing high resistance and abrasion.

Plated materials such as chrome and nickel meet the needed yield strength but I have not found any information on the surface oxides of these materials in regards to fretting resistance so proper selection of material is impossible. If I were to use an empirical approach I would need insight into an appropriate accelerated life test.

Any thoughts would be appreciated on this subject.

 

[Astronuc]

Is the rolling ball switch like these?

http://www.mouser.com/catalog/specsheets/MS-100910.pdf

http://www.sparkfun.com/datasheets/Sensors/Ball-Rolling Switch AT.pdf


Is the tube not sealed? Which is typically brass - the tube, the ball, or both?

Are the contacts typically copper?

A thin layer of gold on nickel or brass would probably suffice.

What kind of environment inside the tube? The tube hermetically sealed?

 

[B.delong]

Yes the part is very similar to these parts you list. Unfortunately the part is not sealed from atmospheric oxygen. Both the tube and ball are brass but the ball has a nickel plate with a gold flash. The ball rarely hits the contacts so this is not the wear problem area. In the attached picture, you can see the lovely fretting oxide buildup and the channel worn by the ball in the tube sidewall.

Nickel is a material I have considered but once both parts are plated with nickel will the nickel oxide protecting the surface now break off causing similar erosion? This is why I am curious on relative ratings of materials to fretting corrosion. Currently, the nickel plating remains intact but I assume this is mostly due to Nickel being much harder then the oxides of copper and zinc fretted from the brass. Thus, the current weakest link is the brass tube.

 

[Astronuc]

IIRC, Cu and Zn oxides are more friable than oxides of Ni or Cr. In steels, the Cr oxides is more tenacious. Ideally one has a thin layer of oxide - nm's rather than microns.

Of course, Ni and Cr plating can various in quality.

At what temperature does the switch function. Variations and cycling in thermal expansion can be problematic.

 

[B.delong]

The switch temperature will very on a daily basis by about 30C.

I have calculated under worst case conditions that the ball’s vibration energy may exert a force of 52 N on the wall during impact at a 4.5e-6 m penetration depth calculated from the Hertzian contact model. This may disrupt the oxides as I assume they won’t like this flexure.

The fretting process is mysterious as I see once source stating that temperature may not greatly affect the process as it is driven by the mechanical impact energy. Also, the same source indicates moisture may impede rather then facilitate the process as a possible lubrication action. Both seem contrary to normal corrosion processes. This article, however, was discussing iron oxides.