Steel Cranked Beam End Supports 1

AFP0592 · Jun 10, 2025

Hi All, I would appreciate your thoughts and feedback on a cranked beam support over a masonry wall.

I'm designing a cracked beam to support a new suspended floor inside an existing roof. To avoid damaging the ceiling below and to maintain the lateral stability of the roof, we will need to keep the ceiling joists. Due to its height in relation to the existing masonry wall, the cranked beam seems the only solution in this case. See photos with the scheme and the assumed support diagram.

Now, I have a transverse beam that would provide lateral and vertical supports for the new cranked beam. The beam is embedded in the external walls.

By designing the cranked beam as a roller support, I have a small horizontal deflection of more or less 1mm in the cranked beam. I may be overthinking this, but just wondering about the cranked beam triggering friction against the wall and the overall stability and the reality of a roller in that case.

I can not have a pin there due to the lateral stability of the wall and the high thrust/axial load that is going to trigger on the cranked beam.

Thank you for your thoughts and views on this.

GeorgeTheCivilEngineer · Jun 10, 2025

I don't see much of an issue with that.

milkshakelake · Jun 10, 2025

I'm not sure the bottom reaction point would be a roller. Aren't your ceiling joists restraining the thrust? I would personally model it as a pin. The horizontal reaction (i.e. "shear" in the vertical part of the cranked beam) would add to the thrust. Therefore, I would reinforce the connection of the ceiling joist to the CMU wall.

I think the roof diaphragm itself can take some thrust, dumping it into the parallel shear walls, and it gets wishy-washy in wood design fashion. But I'd reinforce those ceiling joist connections just in case.

If it's a pain to reinforce those connections and you need to keep the ceiling in place, the thrust might just wishy-wash away (because you have a steel beam at the end of the cranked beam, and you have the diaphragm as well). So maybe it's not 100% necessary. I'm just giving something to think about.

IceNine · Jun 10, 2025

I've always done what you're proposing. Make the cranked beam stiff enough that the horizontal deflection with a roller support is minimal. Then you don't have to worry about the thrust, let the wall push out 1mm.

XR250 · Jun 10, 2025

I would not worry about any of this.

Eng16080 · Jun 10, 2025

I would also not worry about a 1 mm displacement from your analytical model. With both the ceiling and roof diaphragms, I doubt this movement will occur. If it does, I still wouldn't be too concerned.

Wood construction moves through shrinkage and seasonally by more than the 1 mm you're concerned about.

driftLimiter · Jun 10, 2025

I worry about this until I see the lateral support deflection. Get it small enough and I stop worrying about it.

Greenalleycat · Jun 10, 2025

Can't you just drop the beam down to the plane of the ceiling joists anyway?

HTURKAK · Jun 11, 2025

Have a look to the first picture , the top of 215 mm masonry wall , Is that RC capping beam ? If so , is this beam in a closed form all around and on the top of interior walls ?
If so , IMO this set up is OK. Can you post the plan ?
Check the masonry stresses and ensure that less than allowable. If necessary , you may consider adding pillars ( fin wall set up) under the cranked beam supports.

AFP0592 · Sunday at 5:15 PM

milkshakelake said:
I'm not sure the bottom reaction point would be a roller. Aren't your ceiling joists restraining the thrust? I would personally model it as a pin. The horizontal reaction (i.e. "shear" in the vertical part of the cranked beam) would add to the thrust. Therefore, I would reinforce the connection of the ceiling joist to the CMU wall.

View attachment 10610

I think the roof diaphragm itself can take some thrust, dumping it into the parallel shear walls, and it gets wishy-washy in wood design fashion. But I'd reinforce those ceiling joist connections just in case.

If it's a pain to reinforce those connections and you need to keep the ceiling in place, the thrust might just wishy-wash away (because you have a steel beam at the end of the cranked beam, and you have the diaphragm as well). So maybe it's not 100% necessary. I'm just giving something to think about.

Hi Milkshakelake,

Thank you for your post and for prompting some further reflection.

Having reviewed your message and the other responses, I feel more confident in concluding that the connection is not acting as a fully free roller. While there may be some horizontal thrust present, it is likely to be limited—certainly less than what would be expected if the support were assumed to be a perfect pin, which was previously resulting in a reaction force of around 25–30 kN under a relatively small UDL applied to the top member.

Another point that helped inform this view is the overall system behaviour: the joists would be properly connected, and the steel beam would be embedded in the external and flank walls, contributing additional restraint and stiffness to the structure.

Many thanks again for the useful discussion.

AFP0592 · Sunday at 5:23 PM

Greenalleycat said:
Can't you just drop the beam down to the plane of the ceiling joists anyway?

Unfortunately, I can't. The client wants to avoid damaging their existing ceiling.

HTURKAK said:
Have a look to the first picture , the top of 215 mm masonry wall , Is that RC capping beam ? If so , is this beam in a closed form all around and on the top of interior walls ?
If so , IMO this set up is OK. Can you post the plan ?
Check the masonry stresses and ensure that less than allowable. If necessary , you may consider adding pillars ( fin wall set up) under the cranked beam supports.

It's a loft conversion. So all the masonry around. The B4 and B5 will be a cranked beam.

Thank you for the help!

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

Steel Cranked Beam End Supports 1

AFP0592

Structural

Attachments

Recommended for you

GeorgeTheCivilEngineer

Structural

milkshakelake

Structural

IceNine

Structural

XR250

Structural

Eng16080

Structural

driftLimiter

Structural

Greenalleycat

Structural

HTURKAK

Structural

AFP0592

Structural

AFP0592

Structural

Similar threads

Part and Inventory Search

Sponsor