For a Steel to be susceptible to Hydrogen Embrittlement, there are several
things that must occur at the same time: First you have to have a
susceptible material; second, you have to have hydrogen present dissolved in
the metal; and lastly, you need a low strain rate. Static tensile loads are
appropriate.
In general, the all steels are susceptible. Cr-Mo steels, i.e., 4140 and
Cr-Ni-Mo (4340) are some of the most susceptible - if they are heat treated
to a tensile strength level above 200KSI. This means, in general, heat
treated to above Rockwell C 43. Practically all steels are susceptible
above HRC 43. The tower materials used are typically cold rolled, annealed
and pickled, or hot rolled, then galvanized AISI 1018 steel. The hardness
of this material is on the order of HRC 25 - not susceptible. The Cr-Mo
masts most people use today are also cold rolled and annealed AISI 4140 or
4340 at approximately 135 ksi (HRC 28) - again not susceptible.
When the steels are hardened above the HRC 43 or 200KSI threshold, then they
become susceptible to hydrogen embrittlement at about the 3-4 ppm level.
BUT to get a failure, it is absolutely necessary to have a nearly static
strain rate. This allows the nascent hydrogen atoms to diffuse thru the
matrix and collect at dislocations or other material imperfections.
There are several primary sources of hydrogen embrittlement. The foremost
one is plating, where hydrogen is generated on the surface of the steel.
Cadmium plating is the worst, requiring a hydrogen embrittlement bake at
375F for at least 24 hours, although several companies require 36 hours at
375F. Another source is from the welding process. However, the culprit is
wet welding rods or cover gases - not welding per se. This is why welding
rods are kept sealed and heated in an oven - it prevents water vapor from
being entrained. It will also drive off water. It is only a problem if the
hardness of the weld metal exceeds approximately 43 HRC, and nearly static
tensile loads are present. However, the presence of residual tensile
stresses can also contribute to the static loading. Another source is
environmental corrosion. This is similar to plating, where hydrogen is
generated at the surface. Depending on the material, hydrogen embrittlement
or stress corrosion can occur - a lot depends on the material couple, and
where the hydrogen is generated. However, for SCC to occur, you need to
have a susceptible material, a corrodant known to cause SCC, and a tensile
stress. This tensile stress does not have to be static - but can be
changing.
Based on the photographs, it is not possible to determine if it is SCC or
HE - to do so requires a SEM, and a confirmation of hydrogen content. If
the hardness of the material is above 43 HRC, and hydrogen is present at
levels above 3 ppm, and the fracture surface is cleanly intergranular, and
the loading is nearly static tensile, then it is likely that the failure
mechanism is HE. If hydrogen is not present, and the fracture shows
evidence of corrosion product on the fracture surface; and the fracture is
intergranular, then it is likely that the fracture is SCC.
HE or SCC fractures are insidious - often failures occur with no warning or
symptoms. Routine inspections using NDT such as magnetic particle is about
the only way to catch SCC prior to catastrophic failure.
However, many of the fractures occurred at the edge of the weld metal, where
the heat effect zone occurs. This is generally soft, and will show a
hardness similar to annealed steel (HRC 25 or so). Typical failures
associated with lack of penetration show similar features to those fractures
shown in the photographs. HE or SCC fractures would typically occur thru
the center of the weld bead because the hardness of the weld bead is highest
at that point because of the extremely rapid cooling rates. No evidence of
this occurred.
It is my personal opinion, based on examination of the photographic
evidence, that it is probable that that welds failed because of lack of
fusion, as a result of improper welding technique. It is not likely that
the welds failed because of HE or SCC. However, this does not rule out the
possibility with out closer (and much more expensive) examination.
Hope this clarifies things. BTW, check out my book, Analytical
Characterization of Aluminum, Steels and Superalloys, published by CRC
Press. Or my upcoming book on Failure Analysis of Steel Components, to be
published by ASTM. I also have another book in work on the Failure Analysis
of Aircraft Components and Systems (no publisher yet).
Scott
-----Original Message-----
From: towertalk-bounces@contesting.com
[mailto:towertalk-bounces@contesting.com]On Behalf Of Hal Kennedy
Sent: Monday, November 14, 2005 1:33 PM
To: towertalk@contesting.com
Subject: [TowerTalk] Hydrogen Embrittlement
The first person who fully understands hydrogen embrittlement (HE) will
win a prize of some kind - or at least write a book a few of us
metal-geeks will buy.
A Google search on the two words will yield up all sorts of background
information.
I've never seen the topic discussed in the context of ham radio however,
so here are a few thoughts:
- HE is a form of stress corrosion, both can lead to stress corrosion
cracking. For either to occur, the metal in question must be under
tensile stress and must have been, or be in, the process of being
corroded. Most of us encounter corrosion due to oxidation - typically
producing rust - but hydrogen is as bad or worse a corroder than oxygen
and does not produce rust (iron oxide) hence making it impossible to
see. My first guess for the failed tower legs in question was simple
stress corrosion cracking, but the complete absence of rust in the
photos makes me think it was HE.
- Various metals are more and less susceptible to HE. Medium strength
steels and stainless steels are susceptible to HE failures. This would
include tubular tower legs. This would also include the "mystery water
pipe" that is often used for masts, but masts are seldom subjected to
the main culprit for causing HE - welding. If you weld to a mystery
water pipe the point on either side of the weld will not carry the same
tensile stress as the rest of the pipe and your mast may fail from a
clean annular crack rather than bending. This is a catastrophic failure
- just like a cleanly cracked tower leg will typically be catastrophic.
Cr-Mo masts "the good stuff" are virtually immune to HE.
- The result of HE is a loss of ductility - i.e., embrittlement. A lot
of the performance (strength) of a metal in tension results from the
ductility of the metal - loss of ductility equates to loss of strength.
- The strength of a metal undergoing HE goes down with time. Sometimes
lots of time - like years. Both elements are at work - corrosion and
applied tensile stress. The exact role applied tensile stress plays is
not understood.
- For HR, we first need to avoid plain old stress corrosion at high
tensile stress points in our metal systems - like at the bottom of tower
legs. Tower leg bottoms often get their galvanizing knocked off or sit
in water - obvious problems. More sinister - the welds at the bottom of
tower legs may have pin holes in the galvanizing permitting corrosion
that is hard to see. Tower leg bottoms should be painted with cold
galvanizing paint (or some other oxidation inhibitor) now and then and
inspected regularly.
- The next thing we need to do is avoid bad designs. Welding at the
bottom of tower legs is asking for trouble. Rohn 25 and all its bigger
cousins are good designs because of three things: 1. The stress at the
bottom of the legs is well below the tensile limit of the unembrittled
metal (unless you make the tower higher than Rohn recommends), 2.
embrittlement of the entire leg cross-section is unlikely given the tack
welds of the cross members are very small, and 3. as guyed designs, the
stress at the bottom of the legs does not rise appreciable with wind
load. I have personally failed two Rohn towers, unfortunately, and in
both cases the legs failed by exceeding the modulus of the steel
producing bending, not cracking. Good job Rohn. Bad job N4GG - we
sometimes learn the hard way. The failed tower we are discussing has
significant welds at the highest stress points in the legs, and is self
supporting with reasonably significant wind load. The wind loading at
the top is applied to the weld-embrittled location on the legs
multiplied by the moment that is proportional to the height of the
tower. If the modulus of the legs had been exceeded they would have
crumpled, if the welds had cracked it would tell us we had bad welds -
the nice clean leg cracks tell us stress corrosion (embrittlement) was
the culprit.
- After a little reading, you can get convinced everything causes HE.
HE is most often caused, in my experience, by welding or plating. Most
ham stuff is not plated (galvanizing within this context is not
plating). Welding plasma contains hydrogen as do some welding rods.
There are special hydrogen-free welding techniques - these are typically
not available to hams or ham-related manufacturers, i.e., they are very,
very expensive.
- Poor welding leads to cracks at the weldment. Even perfect welding -
if there is such a thing - can lead to crack failure in the base
material (not the weld) immediately adjacent to the weld due to HE.
I hope this is helpful...
N4GG
_______________________________________________
See: http://www.mscomputer.com for "Self Supporting Towers", "Wireless
Weather Stations", and lot's more. Call Toll Free, 1-800-333-9041 with any
questions and ask for Sherman, W2FLA.
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
_______________________________________________
See: http://www.mscomputer.com for "Self Supporting Towers", "Wireless Weather
Stations", and lot's more. Call Toll Free, 1-800-333-9041 with any questions
and ask for Sherman, W2FLA.
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|