Replacement of W18x76 with a Built up Section

[tmgczb]

I have no experience nor knowledge of composite sections built up sections.
Now vendor for our project is suggesting that we replace W18x76 with a built up section.
The thickness of steel plates is 18mm and 12mm, connected by fillet weld 8mm at both sides of web plate.

 

[BAretired]

What is your question?

The built up section shown in your sketch requires a lot of welding, which tends to make it uneconomical. I would be surprised to learn that such a beam is more economical than a W18x76, but I am not familiar with prices in your country (Iraq).

Your vendor may be thinking of a section in which the concrete slab acts compositely with the steel beam. That would be the more usual interpretation of the term "composite section". It is quite conceivable that a composite beam would be more economical than a W18x76, but if you have no experience or knowledge about composite sections, you will need the assistance of someone more familiar with composite beam design.

Check the internet for information on composite beams. Below is a section of a composite beam showing the essential parts.

BA

 

[skeletron]

It's the welded connections that will be the difference here. Getting an 8mm fillet (2-pass weld) all across the top flange to web connection, both sides is sure to have some distortion effects + labour costs + quality control issues. It can be done, but it's not a straight swap without some extra precautions.

 

[BridgeSmith]

As BAretired noted, the OP says the "vendor for our project is suggesting that we replace W18x76 with a composite section", but the screenshot shows a built-up plate girder section. Maybe things are different in other industries, but in the bridge world, suggesting a change to a composite section wouldn't come from a vendor.

Assuming the substitution proposed is a plate girder in place of the wide flange, unless there are issues with availability of the W18x76, I don't see how a plate girder of similar dimensions would be more economical. We would generally only consider a plate girder where a WF would work if the plate girder was at least 30 lb/ft lighter.

Rod Smith, P.E., The artist formerly known as HotRod10

 

[tmgczb]

Sorry,it should not be "composite sections" but "built up sections"

 

[hokie66]

The world we live in today can be a strange one. While I agree with others that welded beams are rarely substituted for rolled sections, industrial conditions are fluid, and some welded beam lines exist in competition with rolling mills. Perhaps that is the situation where member tmgczb is located.

https://www.bhullargroup.com.au/welded-sections.php

 

[JStephen]

If using the rolled section involves shipping a single beam halfway around the world, that built up section may look a whole lot more competitive.

 

[tmgczb]

These beams are going to be used for piperacks in oil field located in Iraq. Are there any basic requirements for built up sections?

 

[tmgczb]

But I consider fillet weld should not be as good as full penetration butt weld

 

[dik]

Does this help?

[URL unfurl="true"]https://res.cloudinary.com/engineering-com/raw/upload/v1641166891/tips/Prop-W_BUu_agztay.sm[/url]

[URL unfurl="true"]https://res.cloudinary.com/engineering-com/image/upload/v1641166891/tips/Prop-W_BUu_qgkvji.pdf[/url]

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[dik]

and for the actual section, rather than the 'test' one:

[URL unfurl="true"]https://res.cloudinary.com/engineering-com/image/upload/v1641167558/tips/Prop-W_BUu_imrp47.pdf[/url]

[URL unfurl="true"]https://res.cloudinary.com/engineering-com/raw/upload/v1641167558/tips/Prop-W_BUu_kphnnc.sm[/url]

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[tmgczb]

This is all about calculation of section properties seems not to mention weld.

 

[desertfox]

Hi tmgczb

This should help I think;-

https://www.newsteelconstruction.com/wp/wp-content/uploads/TechPaper/AD411.pdf

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[dik]

thanks, the SMath is a work in progress... Regarding desertfox's submission... a couple of questions (I'm not familiar with the codes used).

What is the 345? MPa? and from where? and the factor 106/109?

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[tham1]

Dik,

A reference for the steel would be BS EN 10025-2. As the flanges in the example desertfox posted are 30 THK it drops to 345MPa 16mm <= t < 40mm, ratner than t < 16mm.

As for chi, I'm not completely sure. Just had a quick skim through EC3, chi is listed as a buckling reduction factor referencing curves. Nothing with the w subscript though.

 

[TLHS]

Do other country's standards still have typical welded wide flange sections? We still have them in Canada, but they're pretty niche. Basically I'm wondering if the engineering in this case can just be "replace the W18x67 with a WWF18x__"

If that's not an option, and the dimensional and material specs are pretty comparable to the original beam, it's likely that you don't need to go through all the web girder design requirements (confirm all the width to thickness requirements etc to make sure though), but you'd need to design the web weld for shear flow and for transfer of any compressive/tensile loads between components. There may also be qualitative requirements for built up sections in whatever governing code you're looking at (minimum welds, spacings, sizes, etc)

 

[TLHS]

Also, the proposed weld looks like a full strength weld, which kind of bypasses what I said in my last post re: needing to check the capacity vs shear flow and transfer between elements. Haven't checked with the specifics, but the dimensionality looks about right. This is not the most economically efficient method, since you could likely justify a stitch weld or a reduced leg based on actual loads, but it's not crazy. It looks like this is what's done for Canadian WWF members up to a web thickness of 20mm unless you specify something custom.

 

[dik]

Thanks, tham

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[BridgeSmith]

For the design of built-up plate girders, there are a few extra design checks; primarily shear on the flange to web welds, which it should pass easily. There are also a few checks that probably wouldn't be critical for a rolled beam, but may be for a welded plate girder, such as fatigue, if you have substantial cyclic loading.

Minor increases in design complexity aside, material/fabrication cost is most likely to be the deciding factor. I would contact the vendor and ask why they made the suggestion.

Rod Smith, P.E., The artist formerly known as HotRod10

 

[SandwichEngine]

In the US and elsewhere, PEMB companies create built up sections for many buildings. They typically only have the web welded to the flanges on one side and that weld size is minimal. Doing a shear flow calc would usually show that even a minimum sized weld wouldn't need to be full length for strength. Typically they are full length on one side but they definitely don't need to be multi-pass welds on both sides. That's overkill and like someone stated above, I'd be concerned with the welds blowing through the web.

I'd ask your fabricator if they have a pull-through auto welder like this one.

https://youtu.be/XklUKO4vix4

If they do, then they'll be able to create perfect welds and the beam should function as intended as long as the section properties are adequate.

It also wouldn't surprise me if it was less expensive than using a wide flange beam. I've recently seen several jobs switched from open web steel joints to built-up beams due to material shortages.