The authoritative public forum
  for Metal Finishing since 1989
-----

Does Cosmetic Anodizing Affect the Fatigue Characteristics of Aluminum

A recent helicopter crash was caused by a fatigue failure of an aluminum control tube that had been anodized for appearance. I believe that if this type of anodization contributes to a reduced fatigue life, the amateur built helicopter community for which we publish a magazine, needs to know.

Stuart Fields
experimental helo magazine - Inyokern, California
2007

---

2007

Yes, it most certainly does adversely affect the fatigue characteristics! Although chromic acid anodizing is performed on critical parts by the aerospace industry (with due allowance for the effects), decorative sulfuric acid anodizing is never done on critical or structural parts. The anodic film created by decorative sulfuric acid anodizing is much thicker than the chromic acid anodized film, which means the loss of aluminum is much greater. Plus the aerospace industry needs to be cautious about inadvertently trapped corrosive materials like sulfuric acid.

Although I don't personally know exactly what topographical features the anodizing might leave on the aluminum substrate to function as stress risers, the loss of material alone is a very significant issue. So, yes, the amateur-built helicopter industry needs to know that flight critical items must never be sulfuric acid anodized.

Ted Mooney, P.E.
Striving to live Aloha
finishing.com - Pine Beach, New Jersey

Is there any test data which would provide numerical measurements indicating the extent of the metal removal in this particular anodizing process? Data indicating the extent of stress riser generation would also be of interest.(I'm an engineer by training and tend to love numbers)

Stuart Fields [returning]
experimental helo magazine - Inyokern, California
2007

---

2007

The thickness of decorative anodizing may vary but it is usually at least .0002". When it is to be dyed, the anodizing must be thick enough to allow saturated colors, so anodizing which will be dyed black is usually .0005" or greater. The coating is oxides and hydroxides of aluminum, and the aluminum that is in the coating comes from dissolution of the substrate. So a general rule of thumb is a 50 percent loss in aluminum thickness. Thus an anodizing thickness of .0004 inches consumes about .0002 inches of aluminum. If the appearance on the first go-round doesn't cut it, the anodizing is stripped. The stripping will not consume aluminum if proper acids are used, but people often use caustics which consume aluminum. Then the second anodized layer will consume another .0002" of aluminum.

There is plenty of data in the rather expensive "Finishing and Surface Treatment of Aluminum and its Alloys", but the short answer remains that even the major aerospace companies, who have access to tightly controlled processes and the most sophisticated testing equipment, never sulfuric anodize components of this nature, and amateurs surely should not.

Ted Mooney, P.E.
Striving to live Aloha
finishing.com - Pine Beach, New Jersey

---

Ted: Thanks for the data. It does raise another question. If I have a 5/8" diameter aluminum tube and I anodize it black, as I understand it, the new outside diameter, not counting the anodized coating is approx. down from 0.625 to 0.6245. This changed the inertia only in the 5th decimal place. This doesn't sound too dangerous to static loading of tension and bending, but I suspect there is a devil that I'm missing in the details. And that is a factor that contributes to the fatigue failure. Are there serious stress risers introduced in the 0.0005 anodize coating?
Yipes the cost of the source you referenced. I just bought a book : "Fatigue Design of Aluminum Components and Structures" Do you have an opinion of this text?
Thanks

Stuart Fields [returning]
experimental helo magazine - Inyokern, California
2007

---

I am not familiar with that book but it would be hard to believe that a book with that title would not include coverage of the effect of anodizing upon aluminum fatigue, Stuart. The anodized coating is brittle and will crack to the raw aluminum, causing stress risers, but calculating for fatigue loading of this sort is something I did only back in school, not the real world, so I don't want to offer an opinion on what factors to apply.

Ted Mooney, P.E.
Striving to live Aloha
finishing.com - Pine Beach, New Jersey
2007

Ted: I got the book I referred to and did some digging and found a chart indicating a significant reduction in fatigue life of anodized aluminum. I also ran into the Endurance Limit concept that I had forgotten. The endurance limit for aluminum was listed as 8-18 ksi. The aluminum control tubes in the helicopter could never see even 1/10th of those kinds of loads. The failed tube was loaded thru rod-end bearings on both ends. Bending moments were not possible. And a 50#, 17 hz cyclic compressive load was estimated. Based on the Endurance Limit concept, the fatigue life should have been infinite in spite of the anodizing. Lab tests showed that there were no pre-anodizing surface defects of any moment, yet the report said the tube failed thru fatigue. There must be something I'm overlooking.

Stuart Fields [returning]
experimental helo magazine - Inyokern, California
2007

---

My field is metal finishing rather than aircraft engineering, Stuart, and I am quite some distance from the problem, so I just cannot comment on those issues except to repeat for the third time that structural components in aircraft must not be sulfuric acid anodized, and this track of trying to calculate safe loading for sulfuric anodized aircraft components is the wrong one.

Ted Mooney, P.E.
Striving to live Aloha
finishing.com - Pine Beach, New Jersey
2007

---

2007

Hi Stuart,

Sorry for not posting sooner as I said I would when you contacted me - I hope the information is still timely.

Although it is not my area of expertise (my company specializes in Type II anodizing - fatigue effects are MUCH less of an issue with this type of coating compared to Type III Hardcoat) I may be able to help point you in the right direction.

The research I've seen seems to indicate that the fatigue loss is caused by the fact that the anodic coating (aluminum oxide) is MUCH less flexible than the aluminum metal itself. When bending and cracking occur, the aluminum oxide cracks more readily than the underlying metal and due to the very tight adhesion of the coat to the substrate - being formed out of the underlying metal the adhesion is excellent - the cracks are able to propagate down into the aluminum at much lower fatigue levels than they would be able to otherwise. The thicker the coating the more likely it is to crack at lower levels - hence a thinner type II coating (done for corrosion resistance and appearance) has less of an effect than the much thicker (5-10 times as thick) hardcoat done for abrasion resistance.

I am unaware of any specific studies addressing this matter, but it is a recognized issue and is addressed in some of the bigger reference books on anodizing - I'm almost certain for example that it is described in "The Surface and Treatment of Aluminum and Its Alloys" by Wernick, Pinner, and Sheasby as well as others.

Try contacting the Boeing company, I know that they have a patented Sulfuric Acid/Boric Acid anodize just to prevent this issue. Also, I believe that Type I anodize (Chromic Acid Anodize per MIL-A-8625 / MIL-PRF-8625 [on DLA] ) was developed to help in situations where fatigue strength was critical as well as areas where chemical entrapment was nigh unavoidable - unfortunately using Chromic Acid to anodize parts has been under heavy attack for environmental and health reasons so it is harder and harder to find someone who can do this process. If you look into the matter deep enough, it wouldn't surprise me to find that the original specification called for a chromic and that the requirement has been waived due to unavailability - this wouldn't be the first time that performance (even safety performance) has been sacrificed due to legal issues regarding environmental effects; for a similar situation look into the effects of outlawing (my understanding is that this has been effectively the case in many situations) cadmium coatings or removing lead from the soldering for electronics - both have been casualties in this way.

Also, the question has come up in the past here on this forum as well -- use the search engine to sort through the archived letter -- you may be able to find something more.

Good luck!

Jim Gorsich
Compton, California, USA

---

2007

We recently saw a fatigue failure in a part we make that is Type III (hard) anodized 6061-T6. Stress in the part according to FEA is in the 12-15 ksi range. The part failed after approximately 25K cycles. In this case, the part needs the hard anodization for wear.

My first concern is with the anodizer. Is it possible/likely that a poor anodization process could be a significant contributor to this type of failure? Too much time in the etching/cleaning process? Our local anodizer does not give me a lot of confidence based upon the way his shop looks (I don't know a lot about the anodization process).

After reading all this, my bigger concern is with the hard anodizing process itself. I assume that Type III is the best process for producing a hard wear layer? Are there some other reasonably priced options out there for hardness and wear resistance?

Thanks.

Al Youngwerth
motor sports - Boise, Idaho