Tension Rupture Photo

[DerChad]

Does anyone have a photo of tension rupture failure at a bolted connection? Please post the link. Thanks.

 

[rb1957]

like from a pressure vessel exploding ?

 

[DerChad]

I meant tension rupture in mild steel at the bolted end of a tension member (structural steel). You can see the equation for it in Chapter J4.1 (Eqn J4-2) of AISC 13th Edition Steel Construction Manual.

 

[JAE]

Did you google search for one?

 

[DerChad]

I did...and had no luck. I'm beginning to consider the possibility that tension rupture at a bolted connection exists only in the imagination of a certain steel code (which shall remain nameless).

 

[rowingengineer]

Connections Teaching Toolkit By the ASIC, has a few nice photo's in it, don't remember if there is any photo's of Tension Rupture.

Arguing with an engineer is like wrestling with a pig in mud. After a while you realize that they like it

 

[hokie66]

You may have more luck asking an academic who does testing to failure. Most of us here are practicing engineers, and we don't want to see failed connections.

 

[DerChad]

I work for a living too and agree that it's not something I want to see associated with anything I've designed. As an ASD guy, tension rupture was something I was told about. Basically, I was told that there was such a thing but that it would never control a typical connection.

Have a look at the equation for tension rupture. Assume a channel (MC shape, say 6") bolted with (8) 3/4" A325 bolts in one line. Assume a bolted connection and a shear lag factor of 0.9 (code limitation). Now do something silly and calculate the hole area based upon a diameter of zero. Calculate the allowable load and put it aside. Now check the same channel for tension on the gross section. With the bolt hole area at zero inches squared, tension rupture controls.

If we check bolt hole bearing, block shear, and the Whitmore section, I find it hard to get my head around the fact that tension rupture controls, especially when it is by such a wide margin. I was hoping that seeing a photo of a section failed in such a manner would help me to understand.

In ASD, the section is controlled by the compression side, with bolt spacing of 18" in the stitch bolts. The connection was sized appropriately such that it would never control.

Any insight would be appreciated. I'll keep digging on my own but with all of the experience in this website, especially those that have many years of experience with LRFD, I'm hoping that I'll find the magic words here that will set off the light bulb above my head.

 

[connectegr]

Google pictures of the Hyatt bridge failure. The connection deviated from the EOR's design and resulted in a "tearout" of the bolts. I have copies, but not on my cell.

For other examples the best source will be lab result photos.

In many practical cases, yielding will cause re-distribution of forces. Failure may not occur, but a change in structural performance will result.

http://www.FerrellEngineering.com

 

[JAE]

Hyatt wasn't, exactly, a tension rupture failure. It was a bolt pull-through of a seam weld at flange tips.

 

[connectegr]

JAE,
I agree that the failure was through the seam. But, that was the only "real life" example of tearout I could think of with pictures.

DerChad,
Tension rupture is not an LRFD phenomena, it is simply a change in terminology from ASD. Rupture (LRFD) = Fracture (ASD). The only real change is that 13th Edition considers block shear a limit state for welds and bolts. Where tearout and block shear were previously consider in bolted connections only.

http://www.FerrellEngineering.com

 

[nutte]

Have a look at the equation for tension rupture. Assume a channel (MC shape, say 6") bolted with (8) 3/4" A325 bolts in one line. Assume a bolted connection and a shear lag factor of 0.9 (code limitation). Now do something silly and calculate the hole area based upon a diameter of zero. Calculate the allowable load and put it aside. Now check the same channel for tension on the gross section. With the bolt hole area at zero inches squared, tension rupture controls.

I don't follow. For tensile rupture, you get a capacity of 0.5*Fu*U*A. For your shape and A36 material, Fu is 58 ksi, U is 0.9, so you get a maximum stress of 26.1 ksi on the gross area.

For tensile yielding, it is 0.6*Fy*A, resulting in 21.6 ksi over the gross area. Tensile yielding would control.

 

[slickdeals]

You can see a rupture/fracture video here. Not on a bolted connection though.

http://www.youtube.com/watch?v=sEcQimHk03g&feature=PlayList&p=46A77FD0AA59F888&index=13

 

[USCeng09]

Perhaps the 1st photo is something you are looking for:

 

[DerChad]

Nutte:

You're right, tension rupture does not control when using ASD. I worried that someone would misunderstand my post just after I entered it. I should have been more clear in that I am looking at an LRFD problem but lamenting the fact that it's not ASD. In LRFD (13th Edition, AISC) tension rupture controls by a fairly wide margin. I talked about my preferred ASD solution and explained how it made more sense to me but I'm being pushed into an LRFD analysis.

 

[DerChad]

USCeng: You illustrate my point exactly. That's a block shear failure and not a tension rupture example. It's interesting to me that tension rupture controls in LRFD and yet we are not able to find an example in a photo of the issue. You'd think that if it controls all bolted channels in tension in LRFD, there would be countless photos of such failures.

It doesn't typically control in ASD so you can understand the lack of photos.

Perhaps we need someone from AISC to explain this and provide us with a photo.

 

[nutte]

How does it control for LRFD? The ratios are the same for both methods. phi*Fu*U=39.15 ksi, phi*Fy=32.4 ksi, where the phis are 0.75 and 0.9 respectively. Yielding still controls.

(Also, the 13th edition gets rid of the 0.9 upper limit on U.)

 

[DerChad]

nutte:

As I stated earlier, the U = 0.9 is a code mandated maximum value for us (TIA/EIA 222-G). In AISC 13th Edition, based upon our vaious channel sizes, we would have a U value that would between 0.94 and 0.97, if we weren't held to 0.9 by TIA. If we assume that the hole is zero inches squared, An = Ag. The steel we are using is A572 GR 65 (Fu=80ksi). Rupture => 0.75*Fu*0.9 = 54ksi. Yielding => 0.9 Fy = 58.5 ksi. Rupture controls. The U value would have to be something greater than 0.975 for yielding to control. And this is with NO area removed for holes.

Once you start removing area to come up with An, the end result is that all of our channels are controlled by tension rupture by a wide margin. If you were to change the U values to AISC 13th values, tension rupture would still control but not by as wide a margin.

 

[nutte]

It doesn't typically control in ASD so you can understand the lack of photos.

It doesn't typically control for A36 material, maybe, where there is a larger difference between Fy and Fu. It's controlling for your material, whether you use ASD or LRFD, because Fy and Fu are much closer to each other.