The zinc bath is maintained around 850F. The weldment is usually placed in the bath until it is heated to the same temperature as the bath. So in essence, the qestion is, can a weld crack if it is subjected to a 850F PWHT. Also take into consideration the heating and cooling rate. Did the galvanizer preheat the part before it was put into the zinc bath, and did the galvanizer water quench the part as some do?
I haven't personally seen any welds fail from galvanizing, but that doesn't mean it couldn't happen. It would seem that if a highly stressed or complex part were galvanized, it would be a lot more succeptable to cracking than lightly stressed parts. If this is the problem, it seems like a stress relief PWHT before galv could help, or even lessening the thermal shock on the part during the galvanizing process.
Also, what kind of weld test are you referring to? Is this a procedure qualification test, or inspection of existing welds in a weldment? If it is a fabricated item, the contract documents should specify when inspection is to take place. In my experience, inspection of welds almost always takes place before coating, as the coating generally interferes with the inspection process, although if a weld is cracked before galvanizing, it will show up nicely afterwards, as the acid from the pickling bath will seep into the crack, and vaporize when it is dipped into the zinc, casuing a holiday or discoloration in the zinc coating (same thing with pinholes or overlap).
Hope this helps.
G Roberts
I have seen welds crack as a result of hot dip galvanizing. We fabricated some W36 x 328 beams with full length 6 x 4 x 1/2" angles stitch welded along the top flange toes. When these were dipped, the angles thermally expanded faster than the beam flanges did and many of the stitch welds "unzipped". The stitch welds had been MT'd after welding so we felt it was unlikely they were cracked before galvanizing.
Our problem was solved by having the galvanizer dip the beams with angles up and at a slower dip rate. (Kind of makes me wonder how many other problems could be out there without knowing about them, at least for heavy sections.)
I am not a UT expert but my opinion is that UT could be useful in checking welds afterwards. Keep in mind that there can be blisters between the steel and zinc, dross inclusions, and rough surfaces in places. But the zinc coating is actually alloyed with the steel, not mechanically held in place like a paint coating is, so I would think you'd get a sound path into the member. I'm sure the UT people in this forum will know.
CHGuilford
CH,
I am aware that various plating processes can, when gone awry, cause cracking. In all humility this is at the frontier of my experience and well beyond my expertise, but then again this site is supposed to help broaden that, I think. My curiosity has been aroused.
I am under the impression there are a variety of issues caused by the acids used, and I am not sure the pickle could not, under the wrong circumstances, be at fault. (I dont even know what is used as a pickle bath) Over my head at this level as I am, I remain curious about the possibility of perhaps Hydrogen being introduced into the steel of the weldment, which I understand to be a common failure mechanism when acid polishing, cleaning or plating at incorrect parameters. I know nothing about the Zn Galvanizing chemistry and process however (except it uses zinc), and very little more about chemistry in general.
One footnote: I don't understand how Zn could be (precisely and technically) alloyed with the steel as you suppose without melting the steel surface as when welding; I was under the impression the mechanics was more akin to soldering or brazing. In any event I am getting cloudy about various atomic, chemical, and electrostatic bonds.
Fundamentals from any source necesary to clarify either of my questions would be appreciated
Regards,
D
dee,
I am not sure what failure mechanism causes cracking in the galvanizing process. Many parts are hot dipped everyday with no problems at all. Other times, cracks appear in welds or base metal. Possibly there is liquid metal embrittlement combined with high thermal stresses as Niekie3 said.. Possibly h2 is introduced from the pickling baths (sulphuric and hydrochloric acids from what I was told). What I do know is some weldments that were MT'd and/or UT'd prior to HDG came back cracked. The example I related was the most extreme that I can recall, and was caused by something associated with galvanizing and a change in technique solved that problem.
I believe AISC is currently trying to determine why many structural beam copes crack after HDG. So I guess the main point of all this is that while galvanizing is a good corrosion prevention process with a long history, items can and will break and it is best to be aware of the possibility. Your friendly local galvanizer can give you more information on good practices to avoid problems.
Regarding your Zn/Fe alloy footnote- 2 publications I have in my office are "Corrosion Prevention by Protective Coatings" by Charles G. Munger and the NACE TPC9- "User's Guide to Hot Dip Galvanizing......." and both describe HDG in the same manner.
TPC9 describes the coating as "a layer of zinc with iron reaction compounds and an outer covering of zinc of the same composition as the bath." It goes on to say "the usual galvanized coating consists of 3 distinct iron-zinc alloy compounds (zeta, delta, and gamma layers) formed by the metallurgical reaction between iron and zinc and covered by a layer of free zinc (eta)." It shows the gamma layer as 75% Zn-25% Fe and adjacent to the steel, with other layers have progressively more Zn and less Fe. From that I would say that galvanizing is very similar to soldering and brazing in that dissimilar metals are metallurgically bonded together.
CHGuilford
CH
Thank you for the information. I am sure the short answer to the original question concerning HDG as a cause for cracks in weldments is "yes, it can cause cracks" to which I add that my research (discrete from experience) suggests many kinds of cleaning and plating processes can also contribute to damage of the weldment.
One side note is that as I read back what I wrote I find I did not communicate very well. I appreciate your courtesy and the depth of your answer, which clearly took some effort. I was a little shocked that I asked you essentially to quote a source... this is not typical of my nature, and was not what I intended to say. I was trying to ask anyone who knew anything about the topic to add it in. It seemed to make sense (if only to me) around midnight, but the point is that it seems we may be expected to be better artesians (thats close to the word was looking for) than writers, and that a certain amount of professional courtesy and latitude between us all is certainly called for and can be expected of me
Regards,
D
Zn can cause liquid metal embrittlement (LME) of highly stressed steel or austenitic S/Steels. While this should not happen under normal galvanizing conditions, it is a well documented failure mechanism of e.g. galvanized components following fires on chemical plants.
I recon that under properly controlled conditions this should not happen in a galvanizing bath. Should the temperature however be too high for any reason, coupled with keeping the component too long in the bath, you could potentially get this failure mechanism occurring.
The explanations regarding the uneven expansion between different section thicknesses does not sound hell of a convincing to me, because during the welding operation these differences would typically be even much greater than would be achieved in the Zn bath.
If you want to be 100% certain, get a metallurgist to have a look at your failure. LME typically resembles intercrystalline stress corrosion cracking.
Hope this helps
Niekie Jooste
W36 x 328 beams have 1.875" by 16.625" flanges with 1" thick webs. The angles were fitted with heels overhanging the edge of the flanges by about 1/2" as I recall. One angle leg was outstanding and one leg flat against the flange with staggered 2" on 24" stitch welds. The angles were intended for attachment to a block wall. The angle lengths was around 30 feet. All material was ASTM A36. The beam assembies had been dipped upside down so the full length ot the angles entered the zinc bath first.
The "popular" opinion was that the relatively small mass of the angles allowed them to reach peak temperature before the heavier flanges did. (Reportedly these were not dipped slowly) If so, they would expand sooner than the beam did. Even though 30 feet of angle would expand in length the same as 30 feet of beam, it did so first. Considering the small size of the stitch welds they would break relatively easily.
I cannot prove to you that the problem was not Liquid Metal Embrittlement. I can say that the thermal expansion caused by a 2" long weld is pretty small, but 30 feet is significant. That would highly stress the stitch welds, so maybe localized LBE was the mechanism for failure as you say. However, the job was completed a couple years ago so I wouldn't be able to have testing down now.
Further, when the galvanizer started dipping the beams at a slightly slower rate and positioned the angles so they entered the bath last, we had no more problems with broken stitch welds.
CHGuilford
Forum
