socket weld gaps before welding dmw's

By - Date 04-29-2005 14:55

Mr. Nelson,
The P-22 pipe (2 1/4Cr- 1Mo) has a recommended preheat of 300F for thickness less than 0.5" in. and 350F for thickness of 0.5" and above. Higher preheat temperatures should be employed if the carbon content is greater than 0.15%, which is not the case for the P-22. ENiCrFe-2 is a good choice for welding P-22 to stainless steel because the nickel alloy filler metal acts as a "plastic hinge", absorbing most of the strain and thereby reducing the risk of cracking in the Cr-Mo base metal. But, in answer to your question, I would say that you experienced excessive, but not detrimental, grain growth in the weld and HAZ of both metals, which is not uncommon during welding of stainless steels or Cr-Mo steels. Also, the preheat, which I feel that 450F is really more than you need, is really only required when using a low-hydrogen filler metals. Less preheat temperatures may be used if the GTAW process is used. The reason for the preheat of the Cr-Mo steel is to prevent hardening and cracking, but the use of the NiCrFe-2 (also called Alloy A) compensates for those two conditions. Again, I believe you experienced excessive grain growth, but it is not detrimental to the integrity of the welded joint, rather, it is very common.

Chuck

By MBSims

Date 04-30-2005 01:37

Chuck,

Is ENiCrFe-2 considered "low hydrogen"? I've seen conflicting opinions on whether the electrode coating for stainless and nickel alloys is considered "low hydrogen" and should be stored in a rod oven. Obviously hydrogen would not be as much of an issue when welding austenitic alloys, but could be an issue on the low alloy steel HAZ in a dissimilar weld.

By - Date 04-30-2005 02:49

Hi Mr. Sims,
Good to hear from you. I guess I wasn't very clear about the "low hydrogen" scenario. I was actually referring to a low hydrogen type filler for welding the Cr-Mo steels, such as the E8018-B3L, for the 2 1/4Cr-1Mo. But since he was joining the Cr-Mo to a stainless steel, the recommendation for preheat is somewhat different. Using the NiCrFe-2 nickel alloy and using it to weld a stainless to a Cr-Mo alloy, I can't see where hydrogen cracking would be an issue. If both base metals were Cr-Mo, P-22 in this case, one would probably use either an E9018-B2, or the E8018-B3L. IN that case, the preheat would be of more of a concern. As you and I know, either would work in most cases.
Usually, not always, but usually, hydrogen cracking is not an issue in the nickel and austenitic grades of stainless filler metals. In stainless steel SMAW electrodes, the chemistry is in the flux, not the wire itself. The wire itself is plain old vanilla 304 stainless usually. Then all the alloying agents are in the flux. So really the reason for keeping them in the oven are for moisture control. I totally agree with your last sentence. To be honest, I've never heard the NiCrFe-2 specifically be considered a low hydrogen alloy, and the same applies to the austenitic grades of stainless stick rod fillers. You might encounter a delayed hydrogen cracking under some circumstances, but even that is rare using the nickel alloy and the stainless materials. Regardless, I really recommend a -17 flux coating on the stick electrodes when welding stainless to stainless or stainless to carbon steel.
I hope this isn't too confusing. Sometimes I even confuse myself. <smile> Anyway, it was good hearing from you. Have a good weekend.

Chuck

By - Date 04-30-2005 03:26

Mr. Sims,
I re-read my comments and always leave out something I wish I had said.
Usually, the only grade of stainless prone to hydrogen cracking is the martensitic grades, and that is normally when using the SMAW or SAW process where the associated flux would tend to introduce hydrogen. In the case of martensitic filler metals, there are SMAW types considered low hydrogen, just as the low hydrogen covered electrodes of the low alloy steels. This is important because the martensitic grades are prone to hydrogen induced cracking, unlike the austenitic grades of stainless steel. That is one reason that a lot of times an austenitic grade of stainless steel is used to weld a martensitic steel.
Well, I think I'm through this time. Again, it was good hearing from you.

Chuck

By GRoberts (***)

Date 05-01-2005 04:47

Hi Chuck,
I was wondering about one issue in your response. Since the original question was essentailly about weld shrinkage, and you discuss excessive grain growth in the weld and HAZ of both metals, I am curious how the two are related. I know you can get excessive grain growth from too low of a cooling rate in the coarse-grain HAZ from too high of a heat input (or a high heat input coupled with high preheat) for the given material, but am not familiar with how grain growth relates to shrinkage or distortion.

By the way, how was the AWS show?

By - Date 05-01-2005 13:54

Good Morning Mr. Roberts,
The original post did not mention shrinkage or distortion, only movement. Respectfully, there is a difference in them all. The two you mentioned (weld shrinkage and grain growth) are related due to the fact that both are contrubutors due to the high heat of the welding process. When welding, the molecules and grains grow, they do not shrink, or contract, until the cooling process starts. In the austenitic materials you will get grain growth regardless of the cooling rate. The "grains" will not shrink back after it is cooled. Excessive grain growth will result in loss of strength and toughness, but this grain growth is not uncommon in austenitic stainless steels. Now, grain growth is detrimental to ferritic stainless steels, though, but that can be controlled by using a stabilized grade of ferritic steel. In answer to your question about "how grain growth relates to shrinkage or distortion" can be answered by saying that all are "related" due to the high heat from the welding process. When welding stainless steels, they (the molecules) will expand during the welding process and will contract during the cooling period. That is where the "movement" comes in to play. Have you ever measured a 20' joint of stainless steel in the heat of the day? Then, measure it again after the sun has gone down and it has cooled down. You will very likely get 2 different readings. Without the heat, you will not experience "grain growth" or "shrinkage or distortion". That is how they are related. One last thing, that is why one has to be careful when restraining stainless steels during welding. The constant elongation/contraction during welding can contribute to weld solidification cracking if not careful. Restraining the weldment, one of the topics I think you contributed comments to, is one of many ways to help control shrinkage or distortion, but it is also can have negative results if not careful.
The AWS show went great. My technical paper was given on Monday. It was titled "Heat Affected Zones in Stainless Steel Weldments". Ms. Mary Ruth Johnsen, the Sr. Editor for the AWS journal ask me if I could prepare the presentation for pubication in the July edition, but I need to have it to her by May 20, which might be hard since I travel quite a bit and may not be able to have a written version by then. If I can get it to her by then, maybe you can read it in the Journal and see what I'm referring to about these posts. Have a good day.

Chuck

By ssbn727 (*****)

Date 05-01-2005 22:00

Hi Chuck!

If you you ca'nt get the written version by the 20th of May, I'm sure Mary Ruth will wait another month...
I personally cannot wait until the written presentation comes out but heck, I can also wait an extra month if I must!!! :)

On another note, do you recommend that welded pipe joints of schedule 40, 316L, 1-1/2" diameter, be cooled down after welding with just shop air (no filtration or drying?) in order to lessen the amount of time one has to wait before the weld & adjacent areas have cooled down to the recommended interpass temperature in order reduce the amount of "down time" in between passes?
To me, it sounds like "throwing caution to the wind"...:)

Respectfully,
SSBN727
Run Silent... Run Deep!!!

By - Date 05-01-2005 23:16

ssbn727,
Goodness, what do I owe to my lucky stars to be able to communicate with you twice on the same day?
I hope I can answer your question without sounding too confusing. By the time a 316L, 1 1/2" diameter, sch. 40 weldment has reached the recommended interpass temperature of 300F., the phase balance has done all it's going to do. Actually, it has happened long before 300F is reached. In the 316L, and at about 2550F., the weld metal is virtually all austenitic and as it cools down some ferrite forms around the grains of the austenitic structure. All of this happens long before ever reaching the recommended interpass of 300F. Probably one of the main reasons to wait until the interpass temp. reaches 300F. is to allow the metal to cool down so that any "movement" (distortion or contraction) is complete. Another reason for waiting is to let the material cool down is to lessen the amount of alloy loss in the weld arc during welding. This is much more critical in welding the Duplex grades of stainless steel than the 316 grades. Still, allowing the steel, regardless of the stainless grade, is to try to avoid staying at a high temperature so long that the base material encounters sigma phase. Letting it cool down to 300F. can help maintain a lower cool down rate, where sigma is not a concern.
I have read many of your comments in this forum, and have really enjoyed and agreed with all you have said (for what that's worth). Thank you for your replies, as I and others in this forum, can appreciate your concerns.

Chuck

By ssbn727 (*****)

Date 05-02-2005 01:04

Hello again Chuck!

I could'nt agree with you more regarding your response and it's almost word verbatim in my response to "mi jefe"... You've also included other reasons I may not have mentioned to "mi jefe" at the time we last spoke.

However, my question to you is whether or not one can - (provided that one waits until postflow is complete, the torch is removed and another 30 seconds have passed) - cool the deposited weld pass faster and in a safe & sound manner by blowing "shop" air - (not appropiately filtered and/or dried which should be used for this application) - in order to reduce the "down time in between weld passes" resulting in completing the weldments in less time.
I apologize for not being more specific with regards to my my original question.

My "jefe" does'nt think that using clean air is necessary because, it might cost too much... I told him that it would cost way too much more in repairing or replacing the defects caused by residual oils and moisture due to the use of unfiltered/undried shop air in order to cool down the weld in between passes, not to mention what this potential contamination does to the arc characteristics and the Wulfram or is it Wolfram? My point is not to disagree with him in implementing this idea of finding an appropriate method of cooling the welds in between passes. I just disagree with his method of implementation.
BTW, we already use chill bars, etc. in dissapating as much heat as possible but, "mi jefe" still gets "irked" about the welder having to wait in between passes for the appropriate interpass temperature...

I suggested for him to enroll in one of your company's welding seminars which covers much of the needed info and provides the necessary knowledge for one to make correct decisions in planning mfg. operations when one is working with 316L & 316Ti stainless steel or any other stainless steel ofr that matter... Needless to say, my suggestion was'nt well recieved - Oh well, he cannot come back to me later and complain about me not advising/cautioning him about this :)

Anywho, I appreciate your previous compliment and your explanation.

Respectfully,
SSBN727
Run Silent... Run Deep!!!

By - Date 05-02-2005 03:24

I will try to be as unprejudicial as possible in my answer as to not offend anyone. We have already discussed that maintaining an appropriate interpass temperature is vital for the welding of stainless steel. But, equally important is the amount of time it takes to get to that temperature. With the austenitic grades of stainless, chromium carbides can precipitate on the grain boundaries of the steel if maintained in the range of 800-1500F. for any amount of time. This is known as sensitization. Once you get down to the sensitizitation range, your phase balance has pretty much done all it's going to do. So, it is desirable to get through the sigma and sensitization phases as rapidly as possible within reason.
"Forced" cooling, or "accelerated cooling" is a term that can mean different methods. In a paper once written, and I can share that with you, the 3 most important things to consider when welding stainless steels are 1) cleanliness, 2) cleanliness, and 3) cleanliness. When welding stainless steel, all else is irrelevant if you do not start out clean. If the shop air is contaminated with oils from the filtration system, I don't think that is an acceptable way to cool down. In these shop air hoses, moisture can gather from the humidity in the air, dust from the activities of the every day workings of the fab shop, and oils. As you mentioned, the Wolfram can become contaminated and leave tungsten inclusions if it gets to that point. If the air is filtered and dried so as to not deposit residuals of the surface of the steel, then you are OK. Another way to speed up the process is to quench the weld with a spray mist of water. Now, that opens up another concern. Most tap water contains chlorine and calcium, neither of which is good for the stainless. It is not always practical to use deionized water, but al least you are keeping the chlorine and calcium off of the weld zone. Chill bars are another safe way to expediate the cooling process. I guess the most important thing to consider if forced cooling is implemented is using the method that introduces the least amount of contamination. That is probably the bottom line. I agree with all you said in your response to "mi jefe" about the cooling.
Again, it was great communicating with you, as you seem to be well versed in the welding of stainless steel.

Chuck

By Jim Hughes (***)

Date 05-03-2005 14:06

Good morning Chuck. I have been trying to get info on this coolig down subject for some time now. You mentioned a paper that you wrote about this subject. Is that something I can tap into. Are you saying that using demin. water for cooling the weld down is acceptable? For instance a spray bottle with demin. water or placing wet rags on the surface of the weld.

By - Date 05-03-2005 14:56

Hi Jim,
When I was at the South Texas Nuclear Power Plant, when the welders finished with their 8" Sch 160 stainless steel weld coupon, it was thrown into a 55 gallon drum of demin. water. Of course, if you did this with carbon steel, the world would cave in. Virtually the only time, using stainless steel, that it is not possible to use water to quench stainless steel is when welding the Duplex grades and that is because of the austenite/ferrite phase balance. Quenching it with water will retard the austenite and leave you with an excessively high ferrite phase. Look at it this way...The only recommended heat treatment for stainless steel is a complete soulution and anneal. This is done at approx. 1900F. followed by a solution (water, or in some cases oil) quench, or rapid cooling with a fan. Quenching the austenitic grades of stainless is not detrimental. Of course, you must be careful if you quench the weld with water while it is still red hot as you can have distortion if not done carfully. I have seen procedures, especially of heavy walled S.S. pipe where quenching is specified to bring it down through the sigma, chi, and alpha prime (885 embrittlement) phases due to the high heat input and slow cooling rate. I will see if I can find that paper you are referring to. A month or so ago, my computer caught a virus and our IT people were not able to save a lot of my documents, so it mght be one of the casualties. Yes, I'm saying that cooling down your welds of the austenitic grades, no Duplex grades, is acceptable if it does not violate the PQR and WPS. The thinner the material, the more chance of distortion, so use the quench discrimenantly. Let me say that I'm not "recommending" you quench all your welds, but saying that quenching does not have a detrimental effect, if done properly, on the more common grades of stainless steel welds.

Chuck

By GRoberts (***)

Date 05-03-2005 02:28

Hi Chuck,
I understand the relationship between heat input and grain gowth, but I'm still not understanding in your original post how you concluded that excessive grain gowth had occured since the poster had not mentioned what heat input was used. Was it only on the basis of the preheat? Also, you mention that there is a difference between shrinkage, distortion, and movement. Are they all not caused the the contraction of the base metal and weld metal during the cooling cycle of the weld? I am interested to know what you consider the difference between them.

I'm glad your presentation went well, and will be looking forward to my next Welding Journal!

By - Date 05-03-2005 12:33

Mr. Roberts,
In chrome moly and stainless steels, grain growth will occur during the welding process. The poster did not have to specify what heat input he used. Apparently he used enough heat to melt the steel. That is when grain growth occurs. In my experience, this will not happen during preheat because it does not, usually, reach high enough temperature for this to occur. So no, my comment was not based on the preheat.
Shrinkage, distortion, and movement....
Shrinkage is just what it says. It happens during the cooling process.
Distortion pretty much is just what it says. Because a weld shrinks, or moves, does not necessarily means it is distorted. You yourself can make 5-20' joint welds and if welded properly have a straight line of pipe. If you allow for the movement and shrinkage, you will have no distortion.
Movement can happen independent of shrinkage or distortion. Granted, when a weld shrinks or is distorted, the joint "moves", but that was not what I was referring to. When enough heat is applied to stainless steel it will elongate immediately, then start cooling (contracting). This is movement. I'm sure that you've applied a rose bud to a weld or piece of pipe to maybe straighten it due to out-of-alignment. As you apply the heat, the stainless elongates, then when you chill the heated area (immediately), it contracts. allowing you to pull the joint to whatever you wish it to be.
So, if you understand the relationship between heat input and grain growth, how can you not understand that if enough heat was used to make the successful weld, why did the post need to specify exactly what the heat input was?

CM

By GRoberts (***)

Date 05-03-2005 19:24

I think I may know where the disconnect is now. In your original post, you said that the fellow had experience "Excessive grain growth". I took that to mean- since it said excessive- that the grain gowth was above what you would normally expect, which would normally mean that heat input was also above what would be optimal, and that this excessive grain growth caused the large "movement" of his part (referring back to the original question in the post). So I guess I'm not clear on why you called it "Excessive" grain gowth.

I understand your definition of distortion now though, in that you are differentiating distortion in that it is detrimental, while shrinkage and movement may or may not be. This is often the problem with welding communication- that there are many definitions used throughout the industry for the same terms. (remember the post about chrome-moly steel, which many people refer to as high strength steel containing small amounts of Cr and Mo.) Also, I noticed that even AWS A3.0 does not have a definition for distortion. Most places I have worked, shrinkage and distortion have been used interchangeably.

By - Date 05-03-2005 19:51

Mr. Roberts,
When I used the term "excessive grain growth" is because that is how it is used industry wide, at least in my line of work. I use that term because we always try to avoid grain growth, but it is very common, so any grain growth is considered "excessive". Maybe it is a bad choice of words, but that's how it is used. So, when trying to avoid grain growth, any growth can be considered excessive. First, we have to determine just exactly what is excessive. To you, just for an example, excessive may mean one thing. To me, it might mean a different thing. In this case, and most cases I deal with, excessive means any more than none.

Chuck

By pipewelder_1999 (****)

Date 04-29-2005 19:26

Thats what the gap is for. To allow for contraction. It;s supposed to do that.

GA

By MBSims

Date 05-04-2005 02:50

OK, who can estimate the total shrinkage due to weld metal contraction and HAZ grain growth/refinement? The givens are 3" schedule XXS, grade P22 pipe, 2500 lb, A182 grade F304 socket welded flange, ENiCrFe-2 weld deposit, fillet size is 1.09 times the pipe wall thickness, socket depth is 0.750 inches, pipe insertion is 0.625 inches prior to welding. Obviously the gap was measured at 0.125 inches before welding and 0.000 inches after welding. But which material had the greatest impact on closing the gap, the F304 flange or the P22 pipe? Did the flange bore contract, did the pipe inside the bore lengthen, or did both occur? Who's up to the challenge?

Giovanni - good homework problem for the kiddies.

By GRoberts (***)

Date 05-04-2005 23:40

Well, Blodget says transverse shrinkage on a butt weld (A socket weld is basiclly a lap weld, so it doesn't correlate directly, but it's the closest I've found so far), is .1 X the average width of the weld. On .600" wall thickness, a 1.09 fillet would be .654. The average width of a right triangle is .5 X the leg, which would be .327. .327 X .1 = .0327, which only accounts for 1/4 of the shrinkage experienced on this socket weld. The dissimilar metals should play a role too, in that austenitic stainless flange has a higher coefficeint of expansion than the P22 pipe, so when linear shrinkage occurs during cooling, the flange will contract more than the pipe, decreasing the gap some. However, I think the only way to know for sure what moved how/where/when would be FEA.