Friction Contribution to Lateral Force Resistance

[AThor]

Hello all. I have a somewhat theoretical question relating to calculating the required tie down force for resisting lateral forces, particularly seismic.

I have attached a sketch to illustrate the scenario I'm thinking of. Let's say I have a rigid box type structure, say 10'x10'x10'. It rests on a bearing pad at each corner. It is not really part of any building structure, just sitting on pavement. We don't want it to slide or tip over. Using ASCE equation 13.3-1, I calculate a seismic lateral force of 1.2*Wp.

The crux of my question is, should I factor in the friction between the bearing pads and surface when calculating the force in any anchor tie down brackets for the base? Let's say I estimate a friction coefficient of 0.5. Then, seismic force = 1.2Wp, and friction force = 0.5Wp. Will the anchors take a load of 1.2Wp or 0.7Wp? There will definitely be friction, I'm just wondering if it is something usually calculated, or just ignored, as a sort of additional safety factor.

Let me know if my reasoning makes sense, and if anybody has thoughts on this type of situation. As always, the help here is much appreciated!

 

[DETstru]

I don't include any friction resistance. Unless you're going to specify how to treat the surfaces to achieve your assumed coefficient you don't have much control over what it will be.

The only time I've used friction (outside of elements touching soil) is when I was doing some design for temporary stadium seating for an event and they didn't want to use a ton of anchors in their clean slab. The AHJ was making a fuss over an earthquake happening during a 1-day event so they needed to find a way to prevent lateral slip without turning the slab into swiss cheese.

 

[atrizzy]

Even though at the base you've calculated your shear to be 0.7Wp, this reduction (if deemed applicable) would only apply to the shear component of the anchorage. The friction resistance wouldn't help the overturning. Not sure if you were asking/considering this or not.

 

[AThor]

Thanks @DET, that makes sense.

As a follow up question, when you're calculating the component seismic force, what do you use for z/h when you have an object like this box, or temporary bleachers, or anything sitting on a ground surface not really connected to a building. I had just been using z/h = 1, but thinking about it, z=0 makes more sense I think. It seems like the z/h = 1 is to account for the amplified motion at the top of a building during shaking. If I can use z=0, it would cut my seismic load down to the minimum required instead of 1.2Wp, which would be easier to design for.

Yes, I agree @atrizzy. The overturning is a matter of the height and balancing the moments between the weight and seismic. No moment arm for friction.

 

[haynewp]

Seismic can also act vertically at the same time to reduce the dead load.

 

[AThor]

Yeah, that's true. I'd have to factor in the 0.2Wp vertically, and use the load combinations with earthquake forces that have a 0.9 load factor on dead. So after that, the friction contribution would be fairly small, especially since I'd probably have to use a conservative coefficient of friction.

 

[DETstru]

AThor,
From your description, it seems like you should be using Chapter 15 to design the anchorage, not Chapter 13.
Chapter 13 is for "nonstructural components that are permanently attached to structures".

 

[Deker]

ASCE 7-10 Section 13.4 requires that friction be neglected for anchorage of nonstructural components. Z is taken as the height of the connection with respect to the base. So if your component sits on the ground and is only anchored at the base, Z=0.

 

[AThor]

Yeah, Det, I think you're right, the "permanently attached to a structure" phrase kind of makes Chapter 13 not applicable. I am not as familiar with chapter 15, but I think equation 15.4-5, for rigid nonbuilding structures might be most applicable. Interestingly, that equation is 0.3Sds*Ie*W, which is exactly the same as the minimum Fp in Chapter 13, which I would have used if I used chapter 13 with z=0. So, it seems like that range of a base shear of around 0.45W is what I should be using.

Ah, thanks for pointing that out Deker. That helps clear up the friction question. And it does seem like z=0 is correct also, based on my above paragraph.

 

[briancpotter]

A value of 0.5 for friction is far, far too high. Concrete cast directly against earth, an extremely rough and tightly connected interface, typically only gets a value of 0.3 or so. If this is some sort of pre-fab structure made of steel or a polymer, with a smooth surface (which sounds like its exposed and thus might get wet?), it's going to have a much lower coefficient than that.

Which is generally the rub (heh) with trying to use friction resistance - it's hard to establish a reasonable coefficient, which forces you to be so conservative to the point where it's not worth including them, even if the code doesn't specifically preclude it.

 

[AThor]

Yep, that's the gist I'm getting. By the time you factor in the unknown coefficient of friction, the vertical seismic uplift effects, and dead load factor 0.9, the friction effect is almost negligible, so might as well just treat it as a small extra safety factor.

Thanks for all the quick advice. Just talking through it was very helpful.

 

[P1ENG]

I use friction, but it is usually a conservative value (between 0.15 and 0.3 per materials). Someone said it earlier, but your net lateral after vertical seismic and load combination factors is (ASD): 0.7*Cs*Wp - μ*(0.6Wp-0.7*0.2*SDS*Wp)
Effective friction simplified: μ*(0.6-0.7*0.2*SDS)*Wp

You already realized it, but it does little. Sometimes it can help though, especially if seismic is small and you don't want anchors.

Juston Fluckey, SE, PE, AWS CWI
Engineering Consultant

 

[tony1851]

In mechanical engineering, friction is always there when you don't want it; in structural engineering, it's never there when you DO want it.