NACA Technical Note No. 842

NACA Technical Note No. 842 - Tidewater and Weather Exposure Tests of Metals used in Aircraft - II was issued by the United States National Advisory Committee for Aeronautics in February 1942. It was an addendum to an earlier NACA Technical Note, which dealt with tidewater and weather-exposure tests being conducted on various metals used in aircraft.

Summary
TN 842 is an addendum to NACA Technical Note No. 736, which dealt with tidewater and weather-exposure tests being conducted by the US National Bureau of Standards on various aluminum alloys, magnesium alloys, and stainless steels used in aircraft. The exposures were begun in June 1938, and were terminated, for this particular series, in June 1941. The methods of exposure and the materials being investigated are described, and the more important results obtained up to the conclusion of the second year's exposure are reported.

Conclusions
The conclusions are pertinent to panels exposed for 2 years under extreme saline conditions, as exemplified by tidewater tests or weather exposure with the metals in close proximity to salt water. The flexural fatigue tests on corroded panels demonstrated that endurance limit losses were lower for the steels containing molybdenum or titanium (approximately 9,000 lb/in²) than for those containing columbium or no additional alloy element (approx. 14,000 lb/in²).
 * 1) The panels were, in general, somewhat more corroded at the end of the second year than of the first year, particularly those with dissimilar metals in contact. In most instances the rate of corrosion during the second year was not as rapid as during the first.
 * 2) Alloys Alclad 24ST and 52S-½H proved the most resistant to corrosion of the aluminum alloys tested and were but slightly attacked during 2 years. Alloys 53ST and anodized 24ST were somewhat more susceptible to attack, while the alloys containing copper, such as 24ST, 14ST, and Major metal were much more susceptible.
 * 3) Anodized Dowmetal M appeared more resistant to corrosion during the first year and anodized Dowmetal H, but during the second year developed considerably larger pits than Dowmetal H.
 * 4) Stainless steels containing 2.5% molybdenum were very slightly more susceptible to corrosion than those containing 3.5% molybdenum, as judged by the rust on panels exposed to the weather for 2 years. At the end of 3 years the stainless steel containing 3.7% molybdenum was much less rusted than steels with additions of columbium or titanium, or than those without additional alloying elements. A 16:1 chromium-nickel alloy was more susceptible to attack than any of the others and was practically the only one on which rust was present in the tidewater tests.
 * 1) Anodized 17ST rivets proved far better than 53ST or anodized A17ST rivets for joining aluminum alloy 24ST. All three were satisfactory for joining aluminum alloys 52S-½H, 53ST, or Alclad 24ST, but the 53ST rivet heads on these alloys, in the weather-exposure tests only, were somewhat more corroded and exhibited intercrystalline attack.
 * 2) AM55S rivets proved far superior to 53ST or anodized 17ST rivets for joining magnesium alloys. Anodically treated AM55S rivets were somewhat more resistant to attack and paints applied to them adhered somewhat better than on unanodized rivets. Anodization was not so effective in improving adherence of paints to AM55S as it was to alloy 24ST.
 * 3) The welds on alloys 52-½H, 53ST, or Alclad 24ST were anodically protected in the tidewater tests but were corroded in the weather tests. Gas welds were the least attacked, spot welds next, and seam welds the most attacked. Welds on 53ST alloy were more prone to attack than on the other two. The aluminum coating on the Alclad 24ST welds was sacrificially attacked and thus prevented deep penetration of corrosion.
 * 4) Anodized gas welds on Dowmetal H proved as resistant to corrosion as the rest of the sheet, but spot welds were severely attacked. Welds on painted panels were practically unattacked after 2 years.
 * 5) Spot welds on stainless steels were more rusted than the remainder of the panel. The rusting was superficial on welds of steel which contained molybdenum.
 * 6) The area ratio between any two dissimilar metals in contact proved very important and was frequently the determining factor in the amount of corrosion. The anodic metal was usually more corroded when its area was small as compared to that of the cathodic metal.
 * 7) Alloys 52S-½H, 53ST, and Alclad 24ST were slightly corroded when in contact with each other but all wee anodic to alloy 24ST and were attacked when in contact with 24ST.
 * 8) Alloy 52S-½H was the least attacked of the aluminum alloys when they were in contact with dissimilar metals. Alloy 53ST was usually considerably more corroded, while attack on 24ST and Alclad 24ST alloys was severe. This result does not necessarily reflect the true potential relationships involved, owing principally to inherent differences in the resistance of the various aluminum alloys to corrosion.
 * 9) The aluminum alloys were anodic to stainless steel, nickel, monel and Inconel, and were severely attacked when exposed in contact with them.
 * 10) Electrodeposited coatings of cadmium on SAE X4130 steel strips attached to aluminum-alloy panels were in excellent condition and intact after 2 years of weather exposure. Electrodeposited zinc coatings on the same steel were mostly corroded off when joined to Alclad 24ST and 53ST sheets. When joined to 52S-½H and 24ST sheets, the zinc was attacked but was not corroded off to the same extent.
 * 11) The magnesium alloys were very anodic to aluminum alloys, or to stainless steel. The adjacent aluminum alloys, especially 24ST and Alclad 24ST, were in turn severely corroded by a base produced during the formation of the resulting corrosion product, which was a basic magnesium carbonate. Dowmetal M proved anodic to Dowmetal H alloy. Painted panels exposed for 2 years to the weather were but slightly corroded.
 * 12) Corrosion products that accumulated at the faying surfaces of the dissimilar metals raised the stresses in some instances enough, with the combined corrosive action, to cause cracks to form in the strips. Such cracks were found on 24ST and Alclad 24ST strips coupled with nickel alloys or stainless steel, on Dowmetal H strips coupled with aluminum alloys or stainless steel, and on stainless-steel strips coupled with Dowmetal M.
 * 13) Painted anodized 24ST panels, with paint schedules utilizing good grades of aluminum-pigmented varnishes conforming to Navy Department Specification V10, V11 or 52V15b, were in excellent condition after 2 years of exposure.
 * 14) The magnesium-alloy panels, painted with good grades of aluminum-pigmented varnishes, were in excellent condition after 2 years of weather exposure, except for slight failures at the edges of and adjacent to those rivet heads from which the paints were off. Paint failures in the tidewater tests became advanced during the second year on three-coat paint schedules. Schedules involving two coats of P27 type (zinc-chromate pigments) primers and two additional coats of aluminum-pigmented varnishes of good grade usually remained in good condition, especially when the second coat of primer was also aluminum pigmented. Primers of the P23 type (iron-oxide pigments) reacted to accelerate attack on the mangesium alloys, after coating failures had occurred.
 * 15) Paint failures were considerably more advanced on the anodized (PT13a) Dowmetal M panels than on those given the chrome-pickle surface treatment and exposed to tide-water. On the Dowmetal H panels, after 2 years of exposure, no differences were observed in the amount of paint failure regardless of which method of surface treatment was used.