Best practice for welding 2 plates full pen. 1

[Java999]

Hello,

I would like to know what is the best practice for welding two 10mm thickness plates full pen as shown on the image.

Is it good idea to make a bottom plate say 10mm whider to avoid burning the edge?

What is the depth of melted metall of second plate?

 

[metengr]

Your option 2 weld joint detail is the best approach. Options 1 and 3 result in local stress risers and risk of weld defects.

 

[EdStainless]

Your weld should fully penetrate (your lower drawing does not show full pen), and have some positive dimension (no underfill or undercut).

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[Java999]

metengr, thank you for your answer. Could you please advise what is the depth of melted metal in plate number 2. Thank you!

 

[Java999]

EdStainless, thank you for your answer! Do you mean the weld will penetrate plate number 2 on full depth? In another words 10mm?

 

[metengr]

The weld detail should show full penetration where the weld prep on the vertical plate should be knife edge over the full thickness of 10mm with no land.

 

[KootK]

My take:

1) All three details are poor as they wold result in lamellar tearing potential. Bevel the horizontal plate instead.

2) Unless you require a full penetration welding, which is rare, partial penetration welds are generally more economical to produce.

3) In architectural applications, we do joints like this successfully all the time without the overhang.

4) If you're going to supply the overhang, fillet welding may be a more economical choice depending on your loading condition.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[EdStainless]

Good job Koot,
The upper right and lower right are the two details that I would like to see.
If you really need the strength then how about a fillet from each side?

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[KootK]

Thanks Ed. Looking at the OP's details, I've noticed that they're showing a gap between plates. If that's intentional and accurate, that may obviate the concern for lamellar tearing as shrinkage would no longer be restrained.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[weldstan]

Java999,
The depth of penetration on the 10 mm thick bottom plate is determined by the weld process and current used. It can vary from about 1/16" to about 3/16". One question should be, why do you care as long as full fusion is obtained?

 

[Java999]

KootK,

Thank you for your answer.

I am not arguing just thinking:

1) if the thickness of the beveled plate is not fully melted (fused) what prevents from lamellar tearing of remaining part of the plate below melted part?
2) Agree;
3) I agree, it is possible to weld without overhang, I was thinking of the best, convenient for fabricator/welder arrangement;
4) Agree;

Partial bevel and weld from one side of plate is not good for offshore applications, as the gap between plates good place for corrosion to start. We prefer where possible to seal all gaps.

 

[KootK]

I like thinking. And it's a fine question that you've asked (#1). The reality is that, even with the improved detailing, there is still some tendency toward lamellar tearing. That tendency is just spread out over multiple lamina and therefor a good deal less severe. Let me know if the sketch below doesn't clear up #1.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[metengr]

KootK
Sorry to throw cold water on lamellar tearing but the Option 2 weld joints are done quite a bit with no issues in other industries. The lamellar tearing requires a susceptible steel microstructure which included a banded structure with sulfide stringers as a preferred path for cracking. Steel cleanliness today and preferred orientation reduce the risk.

http://www.twi-global.com/technical-knowledge/job-knowledge/defects-lamellar-tearing-047/

 

[KootK]

Sorry to throw cold water on lamellar tearing but the Option 2 weld joints are done quite a bit with no issues in other industries. The lamellar tearing requires a susceptible steel microstructure which included a banded structure with sulfide stringers as a preferred path for cracking. Steel cleanliness today and preferred orientation reduce the risk.

This is interesting information but I don't see how any of it detracts from my previous statements. It a matter of what is better and worse detailing, not a matter of what is possible or impossible. No doubt, in some situations, the worse detailing is quite possible, as you say. Also, keep in mind that until this afternoon, I had no idea what the particular application/industry was.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[Java999]

weldstan, thank you for your answer!

 

[Java999]

KootK,

Thank you for the sketch!

 

[Java999]

KootK, metengr,

My understanding why option 2 is maybe better for full pen (I do not consider other types of weld as the full pen is what is required in my case): the stress in the weld area is spreading in the metall around. Bigger the area - less stress consentration in the joint. In option 2 overhang helps to spread the stress. Is what you ment metengr?
I have skethed out my understanding what will happen in the regards to laminating in the option 2. (Sorry for poor qualiy of the sketch) I think it is less affected in the case of overhang. I would like to know your opinion.

 

[hokie66]

KootK's sketch and illustrations show the lamellar tearing issue well. I would point out, however, that lamellar tearing is generally an issue only in thicker plates. I suppose it is possible, but lamellar tearing in 10 mm thick plates is not on my radar.

 

[Java999]

hokie66,
KootK's sketch explains his point wery well but does not show what will happen to the plate number 2 (option 2 with overhang) on my sketch. Combining what said by KootK and metengr, I sketched out my understanding and I would like to know whether it is correct or not.

 

[KootK]

On the issue of lamellar tearing:

1) This document (Link) is quite comprehensive and appears to be the source document for metengr's link above. I've included a snippet below which speaks to the 1970's line of thinking on plate thickness and steel cleanliness.

2) Here's a more modern discussion (Link). The Grand Coulee example is one of the rare instances of a 10mm plate lamellar tearing failure. It is also an example of exceptionally high level of restraint. As I understand it, thin plates perform better with respect to lamellar tearing because their flexibility can take up some of the induced strain in many cases. I'm not sure how that would apply to the case at hand.

My understanding why option 2 is maybe better for full pen (I do not consider other types of weld as the full pen is what is required in my case): the stress in the weld area is spreading in the metal around. Bigger the area - less stress consentration in the joint. In option 2 overhang helps to spread the stress.

From a lamellar tearing perspective, I would agree that your #2 would be better than my "bad" detail above. As you say, you would be spreading the tendency for delamination over a larger area. That said, you'd still only be delivering the load to a small number of laminations so I would not expect it to be anywhere as good as my "good detail". Interestingly, the link that I supplied above includes something similar to your #2 in the "bad" category. See the clip below.

With regard to stress concentrations, I suspect the issues are less with lamellar tearing and more to do with stress concentrators that generally accompany a difficult to accomplish weld. You know, starts and stops and surfaces that are difficult to keep smooth and regular. The detail that I recommended, in a architectural context, would often be ground smooth which tends to ameliorate some of that. If this is a field weld, there is no doubt that the overhang, particularly when combined with two sided fillet welding, would produce a welder friendly, higher quality weld.

I'll post-face my comments with the following: I haven't welded a single thing in my entire life. In this matter, I'm happy to defer to those who have.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.