Assumption for threads in shear plane of bolted connections for bridges

[Enginerdad]

For bolted steel connections, AASHTO LRFD allows threads to be either included or excluded from the shear plane. Obviously it would be conservative for design to assume that the the threads are included, but is that the standard industry assumption? Or is there some provision somewhere that requires the threads to be excluded, which would allow me to safely assume the less conservative condition?

 

[BridgeSmith]

Whether threads are included in the shear plane for design is typically dependent on the thickness of the plate on the nut side of the assembly. We specify 7/8" HS (Grade A325) bolts for pretty much all our bolted assemblies. For those, if the plate on the nut side is 1/2" thick or more, we typically exclude the threads from the shear plane for design, because if the proper length bolts are used (which our specs dictate), the threaded portion won't extend into the shear plane. For plates less than 1/2" thick, the threads could be in the shear plane, so we design accordingly.

Edit: I just noticed the 1/2" dimension is in the spec. See C6.13.6.1.3b (AASHTO LRFD, 9th Edition)

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

 

[Enginerdad]

BridgeSmith, thank you! Even though that reference is specified for flange splices, I don't see why it wouldn't apply to other connections as well. In my case it's a diaphragm connection that is designed for jacking load. It's not guaranteed which way the bolts will be inserted, but if both the connection plate and channel web are >= 1/2" I'll use threads excluded, and if not I'll use threads included. I really appreciate the information!

 

[dik]

Nearly all the design I've encountered, the threads are included in the joint.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[Tomfh]

I’ve never excluded the threads. There’s enough things that go wrong as it is.

 

[rb1957]

by design, threads shouldn't IMHO be in the shear plane. By analysis they can be, as a conservative measure.

Maybe other industries don't control the plain shank dim'n as well as mine, maybe it's a case of "what if the installer uses a bolt with a narrower plane shank (and no one can check once it's installed".

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[BridgeSmith]

Maybe other industries don't control the plain shank dim'n as well as mine,

Probably not. We just spec ASTM F3125, which are supplied in 1/4" length increments, with a set threaded length, based on the nut and washer thickness. For thin plates on the nut side, the designer can either do the math to determine if the threads will be in the shear plane (and then specify the bolt length), or just design assuming the threads are in the shear plane, and move on.

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

 

[Enginerdad]

I really appreciate the discussion from everybody here. It's always been my practice to conservatively design including the threads in the shear plane, and I think I'm going to continue on with that practice unless it's particularly important to really refine the design. Additionally, since nearly all connections in bridges are slip-critical, it's usually the slip limit that controls the bolt pattern anyway. But if bolt shear ever limits my design I'll be sure to take a closer look at it.

 

[Lomarandil]

Maybe other industries don't control the plain shank dim'n as well as mine

I guarantee my contractors don't control their bolt sourcing and inventory with one quarter of the attention yours do.

We also end up with field tolerance issues (fit up, weld heat distortion, etc) that mean a nominally correct bolt length ends up a thread or two short of full nut engagement. That can often be allowed on the tension capacity, but it also means there is that much more of the threaded shank potentially in the shear plane.

 

[dik]

On any of the shop drawings I've reviewed, 1/4" increment seems to be the norm.

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[BridgeSmith]

There's the thread runout region that needs to be avoided, so the nut doesn't bottom out, and the full shank, I believe, has to extend 1/8" past the shear plane for the threads to be excluded, so that combined with the 1/4" length increment, is why the minimum 1/2" plate thickness is in the AASHTO spec. If you need the extra shear strength for a connection with a thinner plate, you can do the math and see if the standard dimensions work, or specify a specialty bolt with a specific thread length.

For bridge design, I can't see it becoming an issue, anyway, since the capacity of connections with plates thinner than 1/2" is typically controlled by the bearing capacity of the metal around the hole in the plate, and not by the shear capacity of the bolt, with the possible exception of a connection with filler plates of substantial thickness. However, we only use filler plates on girder flange splices, where we almost never use splice plates less than 1/2".

So basically, in steel bridge girder design, we exclude the threads from the flange splice connections, but include them for web splices and most other connections.

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

 

[Enginerdad]

BridgeSmith ASTM F3125 bolts unfortunately don't come in various-length threads. The thread length is fixed based on the bolt diameter.

 

[BridgeSmith]

Correct, enginerdad. That's why I said if the designer needs a very specific thread length to be able to exclude the threads, a specialty bolt would need to be spec'd. The 1/4" increment on the bolt length gets you close enough, if you have a 1/2" between the washer and the shear plane, again, assuming the correct length bolts are used, i.e. ones that have full thread engagement but not more than 1/4" projection, with plates at their design thickness and tight against each other, which is required for slip-critical connections.

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