Jorge,
Bridges typically come under the fracture critical code don't they? Meaning that if a component fails, the entire structure could collapse. Not really an appropriate place for gambling with H2, considering that they are subject to dynamic and cyclic tensile stresses. Isn't that the reason behind the extra caution?
Isn't a martensitic microstructure more suseptable to hydrogen induced cracking than a ferritic or austenitic microstructure, because of the limited number of slip systems, 3, found in a martensitic microstructure, not counting mechanical twinning, as opposed to 12 slip systems, found in both ferrite and austenite? I am under the impression that the slip systems are the mechanism which allow for plastic deformation, and with fewer slip systems available, the martensitic microstructure is more prone to fracture. Not to say that hydrogen won't wreak havoc on a ferritic or austenitic microstructure, just that the martensite is more prone to experiencing problems with hydrogen. Is this correct?
I am also under the impression that, concerning carbon steel, martensite is a transformation product of austenite which has under gone rapid cooling, in which the carbon doesn't have time to diffuse from the austenite, forcing the face-centered cubical structure of austenite to transform by shear, into the body-centered tetragonal structure, known as martensite, and a perfect place for the formation of martensite would be the heat affected zone, with the associated large thermal gradients, potentially resulting in underbead and hydrogen induced cracking.
Would it be, generally speaking, true to say that a martensitic transformation is a result of:1, chemical composition of the base material, 2, initial temperature at the start of cooling, and 3, the rate at which said cooling occurs?
I apologize for the length and number of questions, but I'm still trying to get a handle on the metallurgical aspects of welding, and the potentially deleterious effects of welding. I'm finding it to be a very, very complicated subject. Someone told me that a metallurgist goes to school for 10 years. I don't know if it's true or not, but I wouldn't doubt it.
Dale Simonds
Hi Lade:
For what you said, Lade, In my past post I left the idea that the "white stains” are more critical than hydrogen induced cracking (HIC) wich is not true, because are two different phenomenes depending of the steel’s microestructure; I gonna try to explain me better (thanks for that Lade). However, the HIC represents one of the most common problems encountered when welding steel structures.
1. The atomic hydrogen has the following effects in steel: (1) embrittlement of the martensite (when the steel has this kind of microestructure) causing the risk of ‘hydrogen induced cracking”; and (2) embrittlement of the ferrite (when the steel has this phase) causing the risk of “white stains” if the weldment is under cyclic loads; these white stains (fragilizated zones) can serve like start points of fatigue fractures. The austenite phase it is not susceptible to embrittlement because the hydrogen.
2. Yes, I’m agree with you Lade when you said that some Codes, like AWS D1.5, has a Fracture Control Plan for several members of a bridge that AWS call them Fracture Critical Members (FCMs). Nobody wants any kind of cracks on these FCMs (HIC or white stains or lamellar tearing or any other) because they can compromise the structural integrity of the bridge and public safety. Then the code requirements are directed to avoid unnecessary hardening of base metal (HAZ) and avoid of hydrogen and other aspects that can lead in cracking. I think that both of us were correct in the reasons for hydrogen control.
3. I am agree with your correlation betwen number of slip systems/plastic deformation capability; I have my doubts regarding the martensite is more prone to fracture because you can have tempered martensite with very high toughness (resistant to brittle fracture). Perhaps you were talking about martensite as welded. I still think that both, martensite and ferrite, are prone to experiencing embrittlement with hydrogen but they can lead in different problems: HIC the former and white stains the second.
4. You explained very well the way to obtain martensite. Just one more thing: normally, the HIC is located in the HAZ but not always; depending of the relative hardenability among base metal and weld metal you can have HIC in the weld metal wich has too large thermal gradients during the welding.
5. I am agree with you in that the martensitic transformation on the HAZ is a result of (1) chemical composition of base metal and (2) the cooling rate. I am confussing with your second point: I don’t know if you are talking about preheat temperature or the máximun temp. achieved in a particular point during the welding; but I think that both of them affects the cooling rate, then may be included in this point.
Just like you, Dale, I am trying to get a handle on the welding metallurgy and its effects, benefical and deleterious but, I am agree with you again, it is a very difficult and wide subject. I think that we’ll have always something to learn about welding, then the “welding school” is for all of the live.
Jorge Giraldo
Medellín, Colombia
Jorge,
Could you explain further, or give some references that I can read, as I'm unfamiliar with the term 'white stain'.
In number 5, I was referencing the initial temperature for the formation of martensite, which varies with base metal composition, and has an effect on the transformation products produced during cooling. It is a point that I'm not still clear on. The martensitic transformation occurs only? from the austenitic phase, so if a steel were already in the ferritic phase, then you wouldn't have any martensite form. True?
Preheating is for minimization of the temperature differential between the weld pool and the base metal, so that we don't have excessively high cooling rates and the formation of unwanted martensite.
With the finish temperature for the martensitic transformation being sometimes well below room temperature, so how does one arrive at tempering for the martensite formed during welding?
I think this is relevant as our welding procedures call for the imediate PHWT directly after welding. If we do this, as I said, how do we ever get to temper the martensite, if it never got to finish it's transformation in the first place?
I hope I'm being clear enough that you get what I'm trying to say.
Dale Simonds
Hi Dale:
What a discussion!. Very interesting.
1. The atomic hydrogen into the ferrite microestructure is concentrated around any possible discontinuity (like an inclusion or a porous) actived by the effect of the cyclical loads (I don’t understand how this mechanism works). Then, this zone around the discontinuity is embrittled by the hydrogen concentration (the precise mechanism of embrittlement for H is still not fully understood) and grow until eventually may serves as nucleation point of a fatigue fracture. The fracture surfaces shown a cristaline (cleavage) sort of fracture around the initial discontinuity different to the tipical fatigue fracture texture; for this reason is called white stain. Take a look to this book: Fundamentals of Welding Metallurgy by Henry Granjon. I gonna try to find other references about this issue.
2. Yes, depending of your particular steel if you have a cooling rate fast enough for the avoidance of upper transformation subproducts (ferrite, perlite, bainite) you will obtain martensite from the initial austenite. The ferrite doesn’t transform to martensite directly.
3. The martensite start temperature (Ms), as you said, is a function of the alloy elements whose can serve us in order to determine Ms using several formulas; the martensite finish temp. (Mf) is not so clear and some autors said that is not possible a full (100%) transformation austenite-martensite, but we can obtain big porcentages (99%); then we have retained austenite. I don’t know what are your kind of steel nor the WPS designe considerations, but I think must be very alloyed if Mf is below room temperature. There are cases where the preheat and interpass temp. is betwen Ms and Mf (partial transformation) and the PWHT after welding in order to temper the martensite; depending of your steel and the tempering soaking time, the retained austenite can transform to bainite. In other cases is demmanded the execution of a second tempering in order to treat the fresh martensite formed from the retained austenite during the cooling of the first tempering. Depends of your materials, method used to avoid HIC, service conditions, etc.
May be if I can take a look at your WPSs can give a more particular concept, but I know that is not easy to access this kind of documents.
I recommend you the reading of a excellent book: Welding Steel without Hydrogen cracking by N. Bailey, F.R. Coe, T.G. Gooch, P. Hart, N Jenkins and R. Parteger. It includes different methods for avoid HIC in different kinds of steels.
Jorge Giraldo
Medellín, Colombia
Hi Jorge,
Thanks for the reference on white stain. I'll be sure to follow up on it.
I was speaking very generally, and referring to the wide range for the formation of Ms and Mf, dependent on composition. Not to say that the materials we generally use have a Mf below room temperature, just that the range is very wide, again, depending on composition.
You hit the nail right on the head, in your reference to the fact that the preheat and interpass temperature might fall between the Ms and Mf.
Again, if the transformation was never allowed to finish before the PWHT was started, what is accomplished other than the reduction of residual stresses?
Perhaps the reduction of the residual stresses themselves is enough in and of itself, as the residual stresses are one of the factors that invoke HIC, and combined with the H2 migration at these elevated temperatures(bake out).
Dale Simonds
Hi Dale:
OK, I got it.
I'll tell you my concept about it, that I think You already knows. If the preheat temp. fall between Ms and Mf we'll have a partial transformation from austenite to martensite whose percentage depends of the final preheat temp. before the PWHT, because the martensite transformation is athermal (is not time dependent - is not diffussional) in contrast with other transformations wich occur at a temp. (isothermally). When we do the PWHT the fresh martensite is tempered and the retained austenite may transform isothermally in other microestructure like bainite (if the soaking time at the temper temp. is enought) or is cooling again until room temp. (RT): (1) If the RT is above Mf, this austenite will transform partially (because RT is below preheat temp.) to martensite in a proportion wich is function of the RT level and the rest will be present as retained austenite (stable at this temp.); (2) If RT is below Mf, then all the initial austenite will transform to fresh martensite.
You touch something important: the tempering treatment looks for improve thoughness and ductility and, as you clearly said, for reduce the residual stresses that can affect the HIC susceptibility. I am agree with you.
Jorge Giraldo
Medellín, Colombia
Thank you Mr. Jorge Giraldo
I didn’t know anything about “white stain”. It’s a good reason to use low hydrogen electrodes for bridges.
1) I don’t have bridge welding code. Does it let to use E6010 for root pass?
2) Because of "white stain" even a small percent of all weld metal, just root pass, in critical welds such as bridge welds is important; I think. So, even root pass must be E7018. Am I right?
3) Does any test piece has taken from test coupon requires to check “white stain” as per bridge welding code?