Stiffness contribution of concrete to beam below

[canwesteng]

I've got fairly large centrifuges on an elevated slab supported on steel below - the centrifuge will be directly above the structural steel but anchored to concrete. The slab won't be composite with the beams below, but will possibly have some impact on the stiffness of the supporting beams, through either friction between the slab/concrete deck and the beam causing some composite action, or through the slab acting as a continuous beam over supporting steel, spreading the load out to adjacent beams that wouldn't otherwise see the load. Because the floor will be low tuned during centrifuge operation, any extra stiffness is detrimental.

Currently, I'm planning to assume slip occurs between the concrete deck and the steel beams, and to spec isolation joints on the concrete above beam centrelines. I'm not sure about the validity of the first assumption though, and not sure about the neccissity of the first. I'm wondering if anyone has any feedback for either approach.

 

[EireEng]

Hi canwesteng,
Unfortuantly I'm not replying to provide any answers or advice here, however I am facing a similar challenge regarding the support of a centrifuge on an elevated slab on steel within a building structure,
I am wondering did you manage to get a workable solution here and if so, could you provide some info on how best to approach the problem,
I am just starting into researching for this so any information would be great while I try build up the knowledge base,

Your proposal to isolate the concrete deck from the beams below is an interesting one to help reduce vibration transferred to supporting steel below,

Much appreciated,

 

[WinelandV]

I did one of these back in 2019, inside an existing facility. I anchored directly to the steel, but also through the concrete - essentially the concrete was just mass to help with the vibrations. I didn't try to make the concrete composite to the beams, and I didn't try to pick up the extra stiffness due to the concrete. See the pics below for final install design. I did have the chance to be at the site while they were using the centrifuge - you could tell it was running (there was some vibration in the slab), but it didn't feel like anything crazy was happening. Just a "yep, the machine is on".

The centrifuge weighed 5,250#, but I can't seem to find the bowl speed as I look back. In my RISA model, I'm getting a natural frequency of 6.7Hz, so I'm going to guess that the bowl is spinning faster than that.

Please note that is a "v" (as in Violin) not a "y".

 

[canwesteng]

I ended up adding enough studs to get composite action between the slab and the beams. The actual stiffnesses using the code provisions for composite beams ended up working out almost the same as if the slab was modelled as a plate element in the model. Some literature suggests the vibrations won't be sufficient to break friction, but that is only considering small vibrations, and I'm sure during run down my centrifuges would break friction. I recommend some sensitivity testing to vary the stiffness of the concrete deck, say +/-50%. It may seem like a lot, but changes in stiffness don't generally have a huge effect. As wineland did, I bolted through concrete straight to steel - the load the anchors actually see is quite low though. In general between the mounting plate and the centrifuge you have isolators with limited or no uplift capacity anyway.

FWIW I had three centrifuges, approx 20T each, running at 25Hz+

 

[Lomarandil]

As a side note, in addition to friction, during demolition we often (but not always) find a non-negligible amount of adhesion between concrete deck and a steel beam underneath. All the more reason to envelope your analysis