Doubler Plate design for beam web

[BAGW]

Hi,

I have a situation where I need to splice the existing beam. I cannot add flange plates due to construction restraints and need to transfer the splice forces through the beam web splice plate only. The tension force is 300k and the beam web is 0.5''. Beam web does not work for block shear strength. So I am planning to add doubler plate at the web to account for thicker web. Any inputs? Is there any resources out there which talks about the design?

Thanks

 

[KootK]

So the load transferred is predominately axial then? I might need a sketch in order to get the gist of this but, as I consider it now, I'm having a hard time imagining how doubler plates would help with this. Once you transfer the loads into the doubler plates, would you not be left with the same block shear issue at the web? How much web length are you able to access on either side of your splice? Can you weld stuff to the inside of the existing flanges at all?

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[rb1957]

maybe nest angles on the inside of the flanges and the web ?

another day in paradise, or is paradise one day closer ?

 

[rb1957]

if "tension" is axial load, then any area of the beam can carry it. Traditionally we use the caps, but if the splice is on the web then there'll be some shear lag load transfer from the caps into the web splice.

if you wanted to be "sexy" about the design you could make the splice plate as a "fish mouth", tapering towards the ends, as the load gradually shears from the caps to the splice.

another day in paradise, or is paradise one day closer ?

 

[hokie66]

rb1957, what do you mean by "caps"? Flanges?

 

[BAGW]

@ Ktook

Adding doubler plate make the web thickness larger thus helping the block shear. Am I missing something here?

The force is around 500kips tension. Web alone is sufficient for 350kips. Adding doubler plate on both sides of the web, increases the web thickness thus block shear capacity. The weld around the double plate will be designed to resist all 500kips of force.

Yes, I can weld the doubler plate to the web. The doubler plate will be 20'' long.

 

[wannabeSE]

I’m confused. Is this a beam or is it a hanger with a tension only load?

 

[BAGW]

its beam splice. Collector beam

 

[hokie66]

Sounds like a tie to me. My definition of a beam is a member primarily loaded in bending.

 

[WinelandV]

I am also confused... if it's block shear that's the problem, why not increase the edge distance of the first row of bolts or increase the number of bolts? If it's a field welded splice, why not increase the length of the weld plate?

If I've misunderstood something, perhaps a sketch of the situation is in order?

 

[KootK]

Adding doubler plate make the web thickness larger thus helping the block shear. Am I missing something here?

I suspect so but it's pretty hard to tell without a sketch. As I see it, the splice plate pulls on the doubler plates which, in turn, pull on the original beam web in block shear, as was originally the case. If that's the case, then the only improvement would come about if, somehow, the perimeter of your doubler plates were somehow substantially larger than the welded perimeter of your splice plates were. And if the doubler plates could be made that large, then I don't see why the splice plates themselves simply couldn't be made that large, obviating the need for the doubler plates altogether.

Yes, I can weld the doubler plate to the web. The doubler plate will be 20'' long.

What I wanted to know was how far into the existing beam you could potentially take the splice plate. Like winelandv, I don't understand why you cant extend the splice plate until the welded or bolted perimeter is sufficient for block shear.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[BAGW]

I am attaching the sketch.

Cannot achieve the block shear capacity by extending the splice plate with additional bolts. The plate becomes too large.

I can access upto 24'' inches plus or minus.

 

[KootK]

Thanks for the sketch. Having seen it, I stand by the concern that I voiced in my last post. I think that your block shear perimeter is now just the perimeter of the doubler plate rather than the perimeter of your bolt group. So you've increased capacity a small amount but at a large cost. And your capacity would be no better than it would be if you just made the splice plate welded and the size of the doubler plates, omitting the doubler plates.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[CANPRO]

It looks like you're using the welded double plate to engage more web along the length of the beam...but only by about maybe 1 or 2 bolt spacings. So it looks like its a lot of site welding for little benefit. Why not skip the bolted splice and just make the welded double plates continuous across the splice?

Be careful with how close your double plates get to the beam flanges - you'll restrict access and won't be able to weld all around as you're showing.

Is there any reason you can't add plates to the underside of top flange and to the bottom flange as well?

 

[rb1957]

thx for the sketch. If the load is tension, axial?, then what is "block shear" ?

isn't the problem "merely" making a load transfer into the doubler good for 500,000 lbs ?
and accounting for shear lag (the load from the caps flanges).

however you do this, working within the inside of the beam flanges, and now also with a length limitation, there is a limit on how much load you can ship. If the load required is bigger than the capacity then ... redesign.

another day in paradise, or is paradise one day closer ?

 

[BAGW]

Maybe I can eliminate the doubler plates and use only the splice plate as pointed out. Will also stress on contractor if there is a possibility of adding flange plates.

Going back to what I had been doing, the doubler plate length is selected based on weld requirement. The block shear for the doubler plate itself is not a problem. To just give you an idea, web can handle 70% of the tension force in block shear.

Adding to this, doubler plate thickness was chosen so that the web thickness + doubler plate thickness works for the block shear. The doubler plate alone cannot carry the entire 500kips of the load.

 

[KootK]

I still feel that there's a fundamental misunderstanding of load path here BAGW. The web doubler does nothing to improve the kip/in block shear capacity of the web. It merely increases the block shear perimeter for the kip/in of the original beam web. This is quite different from the shear capacity improvement that doublers do give you in, say, panel zone reinforcement schemes.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[WinelandV]

rb1957,

Shear lag needs to be accounted for, but the block shear failure also needs to be accounted for in tension connections. It's a simultaneous failure of tension rupture at the part perpendicular to the load and shear yielding or rupture at the part parallel to the load. I'm not sure if it's in the AISC LRFD 1st or 2nd editions, but it is in the AISC 13th edition - Section J4.3. I've attached a snip of tension member showing the shear and tension areas.

So in BAGW's situation, it's checking for a 3-sided shape pulling out of the base material.

 

[BAGW]

@Koootk, why wont the doubler increase the kip/in capacity of the web. Now the doublr plate acts like part of the web as it has weld capacity to transfer all the tension force. If the connection has to break out in block shear, it has to pull out the web and the doubler plate as a system.

 

[KootK]

The sandwich that is the web + doublers will all rip out together in block shear. And when they do, that ripping will be around the perimeter of the doubler plates but just outside of the doubler plates. So you'll be ripping through lonely web with no aid from the doubler.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.