50ksi W21x44 over 60' long Welding 2 together? 4

[crozroz0000]

I need a W21x44 about 70'long. They come in 60' lengths. What is important to consider when specifying how to weld them together? Also is there a special inspection I must do (xray the full penetration weld)?

 

[metengr]

From a metallurgical view, you would want to have a full penetration weld joint to assure complete fusion of the weld filler metal with the base material (I-beam ends). This can be done using a weld prep that requires welding from both sides of the beam. A full penetration weld joint detail can be found in AWS D1.1 Structural Welding Code - Steel. Follow all required preheat and post weld heat treat requirements for the base material. You have several options regarding welding process – SMAW, FCAW-S or FCAW-G.

Regarding nondestructive testing of the full penetration I- beam weld joint, you have several choices for a volumetric examination - radiographic (x-ray) or ultrasonic examination. I would also make sure you have blend ground the weld cap across the entire web and flange surfaces to assure complete removal of all weld toe undercut and high/low spots. In addition to the volumetric examination, I would perform a final surface NDT – wet fluorescent MT.

From a Code view, you need to review AWS D1.1 carefully to make sure all welding requirements are met. Also, I would review this idea with a licensed Structural Engineer regarding placement of the full penetration weld relative to the load carrying capacity of the I-beam.

 

[metengr]

carpenters1son;
One other important point, I just noticed that in the title of your post you stated 50 Ksi for the I-beam. If this is what I think it is you are dealing with a high strength low alloy (HSLA) structural steel in lieu of A-36, and are going to have to watch your heat input during welding, and use lower preheat temperature requirements to avoid increased heat input which could adversely affect the mechanical properties of the I-beam adjacent to the weld joint.

This material is readily weldable but you need to use caution, and use low hydrogen electrodes (H4). Post weld heat treatment is probably not advisable because of deleterious effects on toughness – you will need to confirm. In addition to qualifying a weld procedure I would obtain some material for welding coupons to optimize the process.

 

[UcfSE]

There may be a prequalified weld you can use. You can also have the fabricator submit a weld process for approval instead of you trying to do it. X-raying is a good choice for inspection. If fatigue will be an issue, you will need to specifically take a look at your weld type and location versus the fatigue capacity of your beam as detailed.

 

[gtaw]

This post doesn't say whether the application is for static or cyclic loading. As "metengr" suggests, grinding the welds flush is a good idea and if fatigue is a consideration, AWS D1.1 would require any backing bars be remove if the fabricator elects a prequalified weld joint that utilizes backing bars.

The flanges are thick enough to consider UT for volumetric examination. The web is too thin to UT per AWS D1.1. X-ray is expensive when compared to UT. Verify the individual performing the NDT is certified per SNT-TC-1A, CP-189, or ACCP. Like the welder, the techician should have a written procedure to follow. I like ACCP certification because the certification process is handled by a third party, i.e., the American Society for Nondestructive Testing.

I also like "meteng's" suggestion to use the wet fluorescent magnetic particle test method to check for surface discontinuities. It is sensitive to small discontinuities open to the surface or slightly subsurface. Fluorescent penetratrant could be a second choice in place of MT if this is a field application and electrical power is not readily available.

If this is a critical application, and a 70 foot span should fit the bill, you might want to consider having a certified welding inspector witness and monitor the welding to make sure the welding procedure is actually followed.

Insist on a written welding procedure specification and make sure the welder is certified for groove welds.

Good luck - Al


Best regards - Al

 

[dik]

Can you use the 21" beam as a castellated (sp?) beam? There should be some automatic/semi-automatic processes for cutting and welding...

Dik

 

[crozroz0000]

You all have answered my question fully. Thank you bunches: By the way, I have no cyclic/fatigue issues. Thanks.

I have a few questions, I'm a little green:
1) Is UT the same as Ultrasonic?
2) What is a castellated beam? Sound's like it hurts

 

[metengr]

Here is the answer to (1)
UT can mean ultrasonic test or in Code language is it referred to as ultrasonic examination.

 

[WillisV]

here is the answer to (2)

A castellated beam is a beam which has been cut in a "zig-zag" pattern down the center of the web, picked up, slid over, and welded back together to form a deaper section with wholes in the web. They are useful for passing ductwork through etc. They are generally produced as proprietary products by different manufacturers. An example would be the "smart beam"

http://www.cmcsteelproducts.com/thumb10.html

 

[WillisV]

err...holes.

 

[crozroz0000]

Thank you metengr and WillisV.
The castellated beam is neat.
You have all been very helpful.

 

[dik]

The advantage of hexagonal holes is that the web only has to be cut once. The beam is used for long spans with minimum shear.

Dik

 

[crozroz0000]

dik,

The castellated beam may work for my application. I'll keep it in mind. Thank you.

 

[HgTX]

Consider placing the weld somewhere other than right in the middle where the moment is greatest (assuming you're supported at both ends).

Check the AISC Steel Manual for appropriate weld access hole shapes and finish.

Metengr--ASTM A 992 is now the standard for rolled structural shapes, rather than A 36. It's basically A 572 Gr. 50 plus a limit on yield point and yield-to-tensile ratio. It's not anywhere as rare a specialty material as you imply (most structural material has been met both A 36 and A 572 Gr. 50 for years even when the spec says "A 36"), and doesn't need such extremes as H4 electrodes or qualified procedures--most typical weld details will be prequalified per D1.1, where A 992 is listed in Group II in Table 3.1 for prequalified base and filler metal combinations.

Hg


Eng-Tips policies: faq731-376

 

[metengr]

HgTX;
My statement regarding A 572 Grade 70 was to imply more of a cautionary approach related to welding, not rarity of material. This material is a HSLA steel, and I would still recommend H4 electrodes, especially for material thickness above 3/4".

 

[dik]

Sorry... I was thinkin' (or NOT) that the 21" was a little shallow for 70'... if it's just a matter of stretching the beam... just weld a 10' chunk on the end. Splice is well away from max moment...

Dik

 

[HgTX]

Even the AWS D1.5 Bridge Welding Code, with all its concern about fatigue and fracture, doesn't require any optional H or R designation for non-fracture-critical applications, and requires only H16 for fracture-critical applications.

The use of H4 electrodes is NOT standard practice for A 992, A 572 Gr. 50, A 709 Gr. 50 or 50S, etc., or else pretty much all structural welding would be done with H4 electrodes.

Low-hydrogen electrodes, procedures, and practices are recommended, but "low-hydrogen" doesn't mean H4, or even H16. "Low-hydrogen" SMAW electrode classifications are listed in AWS A5.1 and A5.5; the H designations are optional further restrcitions. (For 50-ksi material, the low-hydrogen classifications would be E7015, E7016, and E7018. I'd be stepping outside my comfort zone if I tried to tell you which one of those to use and why, but the bridge welders mostly seem to use E7018.)

AWS D1.1 has lots of info about low-hydrogen practice.

Not that the H4 will hurt, but if the shop has a hermetically sealed canister of regular E7018 on hand, there really shouldn't be any problem with that.

If you want to be as cautious as you reasonably could be, follow the fracture-critical provisions of the AWS D1.5 welding code, though the codewriters' intent is really more for the cyclic loading provisions of AWS D1.1 to be used for your application.

Hg

Eng-Tips policies: faq731-376

 

[ornerynorsk]

Some people use UT interchangeably with Ultimate Tensile also, although UTS is more proper.

 

[unclesyd]

Trial version of Beam Splice Software with a relative inexpensive site license that might be of interest.

http://www.steel.org/AM/Template.cfm?Section=Home&template=/CM/HTMLDisplay.cfm&ContentID=9160