Is Roof Live Load considered transitory to allow 1/3 stress increase for bearing? 1

[ZiggyKS]

Working on a 180-ft long span PEMB and getting huge kickout loads. Geotech took "Easy Button" and used presumptive capacities from IBC Table 1806.2 for clay soils resulting in lateral resistance of 100 psf/ft depth passive and 130 psf shear strength of soil. At least they gave us 2,500 psf bearing w/ 1/3 increase. I did preliminary size based on 250 psf/ft passive for Dead + Collateral and 250x4/3 = 333 so I limited myself to 300 psf/ft for D +C +Lr
They didn't check live load reduction so we got 90-kip kickout force for Lr = 20 psf. We had them rerun with reduction and was able to drop that down to 70-kip lateral for Lr = 12 psf. I can deal with the footprint for vertical load and end up with a 12'x10' ftng so no issues there. Frost depth is 18" so the perimeter grade beam is 24" deep to give a little cushion.
CHK 1: Apply Dead load + Collateral load case and resist that with values as specified.
CHK 2: Apply D +C +Lr load case and resist with or without 1/3 increase?
CHK 3+: Use the PEMB load cases for W and E and check as normal w/ 1/3 increase or w/o increase as needed.

I gave constr team three options since the Geotech gave me less than expected lateral resistance.
► OPT 1: Step 1, drop footing from 2'-0" to 4'-0" below grade to gain passive wedge. Step 2, treat grade beams like T-Beam flanges, then drop them to bottom of footing and extend outward 8'-0" each side of the footing to gain passive wedge width. After 8'-0" taper trench back up to 24" depth. 6 x1.50' grade beam width = 9.00' so 8'-0" extension seemed reasonable. This gets me plenty of passive resistance without relying on any friction or tie backs. Original grade beams were designed as 8" wide earth formed trench (12" bucket assumed) to coincide with 8" bypass girts. I cannot get constr. team to use turn-down slab. So, I would also have to transition the grade beam trench width from 18" to 12" which I don't think will cause too much heartache w/ constr. team.
► OPT 2: Same as OPT 1 but instead of add'l excavation for the footing I would leave the footing where it is and have them trench in a key below the footing and extend that out for the grade beam. Same amount of passive wedge plus I get a little benefit from soil shear in front of key under the footing.
► OPT 3: I still want to drop the footing from 2'-0" to 4'-0" but instead of extending the grade beams out I would run a grade beam back 12' to another longitudinal grade beam forming a 'T'. Then I get soil shear for the footprint in front of the Tee plus passive on the footing and grade beam above the footing. Constr. gets to keep their 8" wide grade beam (12" bucket trench) in this case and I get a funky design utilizing the T-grade beam as a deadman. Since this approach is a little outside the norm I thought I would throw this out to you all. Would I also need to check the passive wedge in front of the T-deadman in case the failure plane is the passive wedge, or can I assume the soil mass moves laterally like on a retaining wall.

Q1) I think 12 psf Lr is a temporary load case and we should be able to use 1/3 increase for lateral and vertical bearing pressures. What say you?
Q2) Which Option would you choose and why?

 

[SE2607]

No 1/3 increase for Lr, only reduction for TA.

What is this collateral load? AFAIK, that's not an ASCE-7 load case.

My choice:
OPT 4: Hire another geotech and make him/her earn their money.

 

[ntattose]

I don't think the 1/3 increase is allowed in this case. Also, lowering the footing results in moment on the footing which will increase your bearing stress. In a case like this it's common to restrain the base of the PEMB column against the horiz. load. For lighter loads hairpins are typically used. For loads like this I recommend ties rods that run all the way across the building and tie the base plates together.

 

[ZiggyKS]

ntattose: We considered full tension ties between columns but we've got utility trenches in between supporting equipment so that prevents us from doing that other than by going deep. We also considered the hairpins to engage the slab but the slab is discontinuous because of the trenches and concrete on soil friction factor for subgrade drag doesn't help much. That is one reason why I wanted to tie grade beams back to a longitudinal grade beam. On one side of the building the grade beam is the trench. On the other side we would have to trench in a grade beam. Didn't want to give away all the details ...

SE2607: You've never heard of collateral loads? These are typically uniform phantom dead loads reserved for things like lighting, conduits, HVAC ductwork and piping. MBMA guidelines give suggestion then you work from there based on the type of occupancy. For metal buildings you can easily get to 7.5 psf if not 10 psf as additional dead load.

 

[ntattose]

I have had problems with trench drains and floor pits myself. In this case passive is a good solution. You didn't include friction, but that will help quite a bit, and you should be able to use both passive and friction together. Also, check with the geotech, but usually you can place some crushed rock in the bottom of the excavation and increase your friction coefficient.

 

[WinelandV]

Without digging in to the IBC, I was under the understanding that you couldn't use the 4/3 stress increase unless you also used different load combinations than the typical ASD combos.

... alright, I looked it up. 1605.3.1.1 specifically DISALLOWS the use of the stress increase with the combinations of 1605.3.1. 1605.3.2 in the 2018 IBC has the different load combinations. 1806.1 mentions "The values of vertical foundation pressure and lateral bearing pressure given in Table 1806.2 shall be permitted to be increased by one-third where used with the alternative basic load combinations of section 1605.3.2 that include wind or earthquake loads."

Here's the snip of 1605.3.2. Once you account for the omega factor as 1.3, suddenly there's not much benefit to using the 1/3 stress increase.

IMO, don't use the stress increase for any of the loads as it's more hassle than it's worth.

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

 

[SE2607]

ZiggyKS -
Thanks for clarifying Collateral loads. I don't do PEMB design, but consider such loads as misc. dead loads.

 

[ZiggyKS]

Good call on IBC not allowing 1/3 increase for Lr. Sure seems like that would acceptable since it is such a temporary load and the reality of 12 psf x180' x30' bays = 64.8 kips is outrageous for a warehouse/industrial setting. 12 psf equates to a 300 lb worker on a 5'x5' grid or 180x30/25 = 216 workers per bay ... like that is ever going to happen.

 

[Aesur]

Since you cannot use tie beams, which would be almost always typical for this size of a PEMB, and because hairpins only give around 10 to 15 kips resistance max typically when considering passive turndowns in combination with friction, it is my opinion that the only two options you have are dropping your footings or using deep caissons. Typically our minimum top of footing depth is 3' to utilize soil above for uplift as 99% of PEMB's I have designed had footings controlled by uplift. The few I have done sim to the size you mention utilized caissons to grab skin friction for uplift resistance.

How are you planning on dealing with concrete breakout from that thrust load, that seems like it may be quite a challenge without providing a large pedestal that extends outside the face of the building, normally above grade as the column sits on top of the slab, or without dropping the column to sit below the top of slab so you can extend the footing/pedestal outside the footprint of the building while being below exterior finished grade, which is frowned upon by PEMB guys?

 

[ntattose]

Aesur - The anchor bolt breakout is a good point. My fix would be shear lugs. As for the PEMB column sitting on top of the footing; that's always my preference, but it requires bypass girts which is sometime a problem with architects. I have never had a problem with the PEMB supplier regarding that.

 

[DanKile]

ZiggyKS - While 200+ workers in a bay is unrealistic, some of that live load will also be material that is staged for the work being done. Even had some resent failures from workers putting too much material in one area.

Aesur - Not a problem for PEMBs to recess base plates, you just need to tell them this in the contract. It might be different with a supplier versus a manufacturer.

 

[XR250]

Do you not have snow in your area? I agree that 20 psf live load is kinda dumb for checking a large trib. frame as never in history has that much material been staged on a roof. Would def. use it for checking smaller trib members though.

 

[Aesur]

Not a problem for PEMBs to recess base plates, you just need to tell them this in the contract. It might be different with a supplier versus a manufacturer.

I agree it shouldn't be an issue, the usual process we see (that I don't really like) is that the PEMB guys are brought on board before the EOR/Foundation designer in my local market, this means they get what they want and get crazy change orders if anything changes. The solution is simple, bring the EOR/Foundation designer on board earlier, but hard to convince some architects/owners/contractors that.

My fix would be shear lugs.

Good idea, I will need to dig into that a bit more, I believe this was added to ACI chapter 17 anchoring to concrete in the current version. We don't need these that often, but I can see some scenarios where this may be beneficial to use.