Structural Lightweight Concrete 7

[pce2000]

I am looking into using lightweight concrete for some structural beams and slabs. I'm trying to find some references that go above and beyond what the ACI code presents. My main source of concern is embed plates and relating the values given by PCI with the lightweight concrete. I am not confident in the PCI method for lightweight concrete in the 90 to 95 lb/ft^3 range.

Any references or comments would be appreciated.

 

[Taro]

PCI and ACI adjustment factors for lightweight concrete are very similar. Can you be more specific about the issues you are not confident in?

 

[pce2000]

Looking over my question, I can see it wasn't clear. My main concern is the factor applied to the root of the compressive strength in both ACI and PCI when you do not know the splitting tensile strength.

According to ACI 11.2, you should multiply the f'c^1/2 by 0.75 for "all-lightweight" concrete and 0.85 for "sand-lightweight" concrete (If you do not know the splitting tensile strength). At the present time, we do not have a splitting tensile strength, so I'm looking at these factors.

The same factors are used for Lamda in PCI. I mention PCI because of their handy embed tables. What are the limitations on the lightweight concrete before these factors are not applicable?

It just seems the lighter the material, the more brittle the concrete will become and shear and pullout will be the biggest problem. I also worry about the bond of the rebar in the concrete. I would like to find any design research that anyone feels beneficial in the design of lightweight components. At present my library is restricted to the ACI code and the PCI code.

Thanks

 

[Taro]

The ACI 318 commentary (R11.2.1.2) states that "the multipliers are based on data from tests of many types of structural lightweight aggregate concrete". I would guess that these factors are derived from a statistical analysis of empirical tests on typical lightweight concrete (85-120 pcf) and result in a comparable factor of safety as normal weight concrete provisions.

As for bond issues, note that ACI 318 equations for reinforcing development length include a multiplier of 1.3 for lightweight concrete. This is equivalent to using a 0.75 factor on the (f'c)^1/2 term. The bond issue would seem to be addressed by the lambda factor.

 

[pce2000]

Thanks for your input. Let me tell you the situation I have encountered and see what you think.

I recently started working with a lightweight concrete building manufacturer who produces in a high volume. As in all designs, they want a cheaper structure which in this business means lighter buildings (for shipping purposes). They hired a concrete technologist to develop a design of the concrete that went from 117lb/ft^3 down to 90 lb/ft^3. They tried this new mix on four buildings and two of them have already failed due to pullout. The first building was clearly overloaded in fabrication.

The second failure was the roof picks. The roof embed plates pulled out in a cone failure during transport at the plant. All four plates pulled out and it was a textbook picture of cone failure under an embed plate.

I looked through the previous engineers calcs and his design seemed to be accurate by ACI and PCI methods. The compression tests of the concrete are also acceptable, but the concrete still failed. This is the source of my doubt of the factors. The concrete that failed is extremely brittle. Too brittle for my normal weight concrete design background to be comfortable with.

I am getting some cores taken from the building and sent to the lab for "splitting tensile strength" tests. I am then going to compare that with the factors presented in ACI 11.2. Do you have any other suggestions besides a bigger safety factor??

I am trying to develop as much knowledge about the design of lightweight concrete structures as possible. Any references that would help out with this or similar lightweight issues would be appreciated.

Thanks and sorry for the long winded reply, but this situation is a cause of great concern. Luckily, the buildings with this mix is not leaving the plant.

 

[Ron]

pce2000....be careful in using only compressive strength as your basis for much of your work. As the unit weight decreases, the relationship between compressive and tensile strength does not hold true to the empirical relationships of normal weight concrete. This is due to the reduced shear strength of lightweight aggregates as compared to normal aggregates. Remember that lightweight aggregates have much higher void volumes than normal weight aggregates and air doesn't produce much shear strength. By the same token, you are probably introducing more air in the concrete mix to achieve such low unit weights. Same problem.

You are on the right track by getting the splitting tensile tests. You might also do some mock-ups and do some direct pullout tests so that you can evaluate the cone-shear tension capacity of the concrete.

 

[JAE]

Another move you could make is in the details of the embed plates. Specifying even small #3 bars running through the anticipated shear cone could make a world of difference in creating a more ductile type failure than the brittle shear cone failure you experienced.

There is a table in the TRW Nelson Stud Catalog that indicates areas of ductile, semi-ductile, and brittle failures. While this doesn't directly address lightweight concrete, your previous pop outs may have been due to a combination of lightweight concrete as well as the location of reinforcing relative to the cones.

Perhaps the embedments could be specified with extra reinforcing placed to assist the shear cone resistance.

 

[kronosconcrete]

It's also easy to have the water cement ratio be wild. If one adds the right amount in the beginning, the aggregate absorbs so much that there isn't enough for the hydration. Also, one can add so much to compensate for the absorption that it's too high. I've had alot of experience mixing 80 pcf concrete at 4000 psi with a six sack mix and a quarter inch minus aggregate with a dry weight of 55 pcf. Getting the water right can make one crazy short of presoaking.