Welding High Tensile Light-Gauge Sections

[ADT]

I frequently see steel-framed buildings using light-gauge (viz. purlin type section) frames, having welded connections at the maximum moment points, ie. knee (column to rafter) and apex of a hipped roof.
It's my understanding that the heat of welding generally speaking destroys the high tensile capacity of steel. So to have a welded connection at maximum load points is a serious contradiction.
Am I missing something? Are there valid weld methods (lower temp; inert gas protection etc etc) which address this? How should one specify such weld procedures, which will frequently be on site?

Please, you mechanical types, help put this ignorant civil & structural engineer straight

Anthony Tugwell
Project Director & Consulting Engineer - now in Australia

 

[GregLocock]

Can you get a spec for the steel?

No, inert gas won't help.

Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[EngineerTex]

I mainly design cranes, derricks and the like, and I spend a good amount of my time modifying existing equipment that's already in the field, consequently, I have an idea or two on the subject, so I'll give it a shot.

1) I don't weld to anything that doesn't have original mill test reports documenting the material. Because of this, it's hard to respond to the term "high tensile" steel. T-1 (ASTM A-514) can be quite successfully welded in the field without much difficulty. It's what bridges are made out of. It's true that the heat affected zone (HAZ) does have different properties than that of the remainder of the metal, but the properties are *different*, not *destroyed*. You'll get a certain amount of hardening and it will be more susceptible to fatigue, but a proper welding procedure executed by a qualified welder on a well-designed joint can result in a connection that comes close to being as strong as the parent material.

2) The points of highest moment aren't necessarily the points of highest stress. A properly-designed joint between, for example, two W-flange beams at 90-degrees to each other can be at the point of highest moment, but if you stiffen your webs at the point where the flanges intersect, you're going to go a long, long way to developing a smooth transition of moment from one beam to the other.

3) If you look at a modular steel buliding, you notice that both the columns and the rafters are tapered. As you noticed correctly, the knees are where the highest moments are and, of course, this is where the beam section is the deepest.

So, with all that in mind, if you're noticing that non-tapered beams are being used in a knee at a point of highest moment, I would assume that the engineer sized the purlin for the highest stress and what you've got is the opposite question for what you asked, which is this: Why do the engineers waste material by oversizing the beams in the areas of the lowest stresses?

And there's my two cents on the subject

-T



Engineering is not the science behind building things. It is the science behind not building things.

 

[EngineerTex]

--Sorry, I didn't address the *main* question you were asking.

I suggest three books:

"The Procedure Handbook of Arc Welding" from the James F Lincoln Arc Welding Foundation

"The Design of Weldments" by Omer W. Blodgett

"Solutions to Design of Weldments" by Omer W. Blodgett

Here's the bookstore:

https://ssl.lincolnelectric.com/foundation/store.asp?PID=16&cat=8

All three of these books are sold EXTREMELY cheaply by Lincoln Electric as a service to education, especially considering the technical nature and the incredible amount of information contained within. Where else are you going to get three books containing more than 600 total pages of highly-technical information for less than $50 all together?

If you're going to design welds or weldments, do not do it without looking at these books.

Then talk to a welding consultant. He/she will give you insight into thing you might not have thought of.

A quick war story: A steel building company bought a bunch of steel plate with a purchase requirement of a minimum yield strength of 36ksi. They welded up the plates, started construction and before the roof was on, every weld cracked and the whole thing came down. Obviously, I'm leaving out a lot of details, but the culprit turned out to be that the steel supplier met the 36ksi minimum yield requirement by supplying steel with a carbon content close to that of tool steel and a yield strength of about 130ksi. This problem may have never arose in bolted or riveted construction, but as you correctly point out, there's a lot to be considered in what happens with a welded connection.

Again, I really suggest that you talk with a welding engineer.

-T

Engineering is not the science behind building things. It is the science behind not building things.

 

[hokie66]

Cold formed steel is commonly welded in Australia for fascia trusses and small portal frames. I don't know the specifics of welding, other than it is arc welding. As you are practicing in Australia, suggest you contact Bluescope Steel technical services for advice.

 

[ADT]

Many thanks all for your comments. I'll make contact with Buildscope in OZ and see what I can get from them, (sometimes an extended process!). I'll contact Lincoln and see if they'll deal with a bloke from down under.

Regarding designing economically, the buildings I'm referring to are medium to large industrial buildings and most commonly use standard galvanised purlin sections (C-profile sections) as main frames, so tapering such light sections is not an option. Sometimes simple knee braces are used instead; occasionally back-to-back section stiffeners.
The grade of the steel is usually referred to here as G500 or G600 (viz. 500MPa or 600MPa yield strength).

Anthony Tugwell
Project Director & Consulting Engineer - now in Australia

 

[BillBirch]

Anthony,
Are they welding the galvanised sections? I should hope not. The problem with that includes welding current tracking across the galvanising (hence lack of penetration), pollution of weld pool, and most importantly, the creation of toxic fumes.
Regards,
Bill

 

[ADT]

Bill, yes the practice is common, even ubiquitous around here. Even if the job has bolted connections you almost always have end plates for the bolting, and these plates, (eg. 8 or 10mm plate) are welded to the galv light gauge section at knee and apex.

I'm trying to get a handle on the best practice issues, in order to address, via approval and other avenues, some potential problems.

A while ago I created & designed a range of such farm and commercial buildings for a fabricator/supplier of the buildings. These were light gauge frame buildings but were different in using very high strength rivets to lapped plates at connection points; a far superior solution. We also did significant actual testing as design backup, which gives significant economies compared to the more usual theoretical design analysis.

Anthony Tugwell
Project Director & Consulting Engineer - now in Australia

 

[hokie66]

Anthony,

It is Bluescope Steel. If you get to the right people in technical services, they are usually quite helpful.

If your base metal thickness is 1.5 mm or more, the steel will be G450, 450MPa yield.

In my experience, portal frames made of purlin sections are small buildings, not "medium to large industrial buildings". I suppose it depends on how you define large.

Welding of these sections is common and well established. I don't know the exact techniques involved, but have confidence that our fabricators do.

 

[ADT]

Yes, Hokie, as you say, under 1.2mm is G500 or G550. 1.6mm and above is G450. My mistake.

To define large, I should have mentioned that these buildings are all "Standard Designs" using the light gauge material. So "large" for these is perhaps up to 20m span and 7.5m frame centres. Obviously not very big by aircraft hangar standards

It's the welding techniques I want to confirm and pin down ... so if anyone knows about this aspect I would be pleased to hear.

Thanks for the input.

Anthony Tugwell
Project Director & Consulting Engineer - now in Australia

 

[hokie66]

20 metre spans are very large for portals built of cold formed purlin sections. Of course, the sections may be deeper than I am thinking about, because the manufacturers can make a lot of different shapes. If you are in Region A, maybe---

 

[hokie66]

By the way, I think you are in the wrong forum. The Welding, Structural Engineering Other Topics, or Structural Steel would be better choices.