Full pen cracking

Date 07-30-2010 19:42

Jpardo,

First, WELCOME TO THE AWS WELDING FORUM!

Second, some questions, type of process? size of electrodes? type of weld- fillets, bevel groove, etc? material grade? When you say "we rotate and weld the other root" are you referring to the other plate or to the second side of the plate the first weld was done on? What size weld are you using for a root pass?

Could it be that the depth of weld is not right for the gap and width of weld and the stresses are pulling it enough as the first one is cooling to cause the second to crack?

Some of the other questions may help better determine the cause.

Have a Great Day, Brent

By jpardo Date 07-30-2010 21:26

Details, details, Thank you for responding. The process is semi-auto FCAW, 1/16" wire, single bevel groove (45 deg), A36 mild steel. We are doing these with ceramic backers, so a one sided weld. We rotate to weld the second plate. We have spoken to several very reputable sources for help, but we are not having any luck, and we are actually getting the cracking on every weld. We all seem to agree that because we are having to pool the weld to achive the nice ceramic backed finish, we are creating a concave weld that is too thin in the middle and as soon as it begins to cool (at a much faster rate than the surrounding material) the stress is causing a complete failure. So our next thought was to slow down and deposit more metal, but no luck. You can only deposit so much with 1/16" wire before you run into premature bridging of the gap and slag inclusion. So now we are thinking of stepping to 5/64" wire (more metal at a faster rate_ to try and build a thicker root. Any thought?

By welderbrent (*****)

Date 07-30-2010 23:09

I would even be tempted to consider 3/32. Are we talking gas shielded or self shielded. A shop I do many TPI inspections at uses 3/32 on everything. It really does the trick on heavier sections. And you are using heavier sections.

How far through does your single bevel go? Most of the way, then you have your 1/4" gap and backing? So after the root, it gets welded out the rest of the way from one side.

Hopefully some of the others will start posting some ideas soon as well.

Have a Great Day, Brent

By L51174

Date 07-31-2010 12:25

Ive seen this before, our solution was to weld the root and at least two more passes before it cooled, and to hold the interpass temperature until the weld was complete, using electric heaters. Is it critical to your process that you get both roots in first? I also agree with the suggestions for higher deposition, such as using a larger wire, but an increase in wire size may require requalification of your WPS?

By 803056

Date 07-31-2010 17:42

I like many of the suggestion I've read, but I would look at the root opening of 1/4 inch to be a bit wide for ceramic backing.

Best regards - Al

By jpardo Date 08-02-2010 01:14

Thanks to all for your suggestions. We stepped up to 5/64" wire and were able to deposit enough metal to stop the cracking. We are doing a qualification and wps for the procedure. As far as the ceramic goes, we are still having a problem with achiving a uniform root appearance on the back side, we are contacting the supplier to see if we should be using a different profile.

By jwright650 (*****)

Date 08-02-2010 16:01

Can you use a ceramic backing that has a concave surface to deposit on? This gives you some reinforcement once the backing is removed and some extra material to dress up with a grinder.

(refer to the 3rd image from the right)

http://www.airgas.com/browse/popup/enlargedImage.aspx

By GRoberts (***)

Date 08-04-2010 23:23

jpardo,
I have seen this problem using FCAW with ceramic backing before as well. With ceramic backing (as opposed to metallic backing), there is no heat removed from the back of the weld, which changes the solidification pattern. The thicker welds you describe that you are achieving with the larger wire is one way to help address the problem because it makes the centerline of the weld thicker and more restitant to cracking. Another way to attack the problem of centerline cracking is to switch to a weld metal with lower impurities. During solidification with the ceramic backing, the impurities like sulfur and phosphorous segregate to the centerline to help cause the cracking problems. The company I was with at the time did not want to switch to a larger wire as you are doing, so I changed to a basic slag FCAW wire, such as E70T-5 (I assume you are using either a E7XT-1, 9, or 12 if gas shielded). Just be aware that if you try this, your welders will most likely complain because the basic slag wires produce more spatter and fumes. However, you may like the back profile better with this approach than the larger wire approach.

By Skaggydog (**)

Date 08-02-2010 19:29

If you are welding to D1.1 specs, welding from only one side with ceramic is not prequalified. I've done this type of weld with ceramic many times and you need more than just the root pass befor allowing it to cool.

By L51174

Date 08-02-2010 20:37

Its not? Just for the fun of it, can you guide me to that? We are switching to mostly ceramic backing, so this might be important lol. Thanks

By Skaggydog (**)

Date 08-03-2010 02:05

In Response to L51174
Sounds like your going to get a kick out of D1.1:2010 2.18.1

By L51174

Date 08-03-2010 11:56

No doubt lol. It makes sense though because we are finding that the stuff melts away if youre not careful. We are currently working from 2006, so I figure by the time 2015 rolls out, we'll be working from 2010. I wish i could bring myself to laugh about it....lol.

Question though, do they specifically exclude ceramic backing?

By jwright650 (*****)

Date 08-03-2010 12:08 Edited 08-03-2010 12:16

BTW, He's quoting out of "Part C" of that Section/Clause....which deals with specifically nontubular cyclically loaded designs.

[quote]Question though, do they specifically exclude ceramic backing?[/quote]
No, see Clause 5.10
Roots of groove welds may be backed by copper, flux, glass tape, ceramic, iron powder, or similar materials to prevent melting through.
EDIT:
He is right though about not being prequalified....I just looked and the 2010 edition has changed the wording in 3.13.2 to mention that the backing must be steel, if welded from one side. You would have to grind or back gouge the second side and reweld to remain prequalified.

By L51174

Date 08-03-2010 12:45

I see, thank you.

By waccobird (****)

Date 08-03-2010 13:47

John
AWS D1.1 2008 only prequalifies steel backing fully fused to base metal only for a CJP anything else is PJP unless it is Qualified according to Clause 4
Marshall

By Skaggydog (**)

Date 08-04-2010 01:31

jwright650BTW said, "He's quoting out of "Part C" of that Section/Clause....which deals with specifically nontubular cyclically loaded designs."

Dang, it got me again. You're right, 3.13.2 says the same thing but the way I meant but different. Sorry I don't keep my mouth shut, but I'm learning things better by throwing them out here. Good for me, hopefully not too bad for you.

By jwright650 (*****)

Date 08-04-2010 11:40

hehe....yeah this code keeps changing and the additional phrase in 3.13.2 changes the way we have to address one-sided welds backed with ceramic backing. (Not sure why the code committee did that to us though, we never had any problems in the past as long as we at least ground the back side flush which got rid of any trouble areas and still passed UT)<--BTW-this wasn't on cyclic material.

No worries, keep digging through this ole code, it's good for all of us, and keeps us on our toes. You never know what they might have changed, if we don't dig back through the new revision a little.

By CHGuilford (****)

Date 08-04-2010 16:25

jpardo,
How did you make out with this? Problem solved, I hope?

By jpardo Date 08-05-2010 00:50

Thanks to all for the responses, this forum is a great learning tool. Well we have changed the direction we were heading with this project by altering the design to accommodate a metal backing. We realized we would have to qualify the proceedure but still wanted to test the ceramic for possible future use as well. I was lucky enough to get in contact with Duane Miller at Lincoln and he was able to provide us with a great deal of insight into the issues we were having. The problems have been solved for the most part and the first stage of the project has been sucessfully UT tested. Here is some of what we learned: The larger diameter wire did solve the cracking problem in the root when using the ceramic backing, but we also learned that ceramic backing is not ideal for a T joint in the flat position. It would not give a clean backside appearance consistently, and still required gouging and cover passes in most instances. To any thinking of using ceramic for a joint such as this, I would recommend trying several sample profiles to find the one that works for your welders. Another helpfull tip was to go with straight CO2 (we typically use a 75/25 mix) and weld with a lower voltage and wire speed while moving slower to help build up a more convex weld to gain strength in the center of the root (where the weld typically cracks), again these are all helpful if attempting the ceramic backing.

By L51174

Date 08-05-2010 10:53

Duane Miller at Lincoln, Im sorry, thats funny, need more coffee I guess. Glad to hear you got it worked out.

By aevald

Date 08-05-2010 13:57

Have to agree, Miller at Lincoln, now that's a bit or irony. Best regards, Allan

By trapdoor (**)

Date 08-05-2010 20:50

I agree with Al on this one. Tighten up the root and mabye use a steeper bevel angle (staying within the allowable tolerances of course). That is where I would start before I did anything more drastic. I would also consider not using ceramic backing I've never really seen it work well consistantly.
just my opinions.